Summary

Over 14 months, contracting authorities concluded 257,000 above-threshold direct award contracts totaling UAH 401.7 billion. This represents approximately 6% of the total number of contracts in the system and nearly 28% of their aggregate value. While the number of such procurements is relatively modest, their financial scale is significant.

At the same time, almost half of the total value (43.3%) of above-threshold direct contracts relates to the procurement of natural gas by a contracting authority on which the government has imposed public service obligations. These are reports on the sale of gas within the Naftogaz group, which became mandatory in 2023. Such procurements may distort the statistics and overstate the problem of direct non-competitive awards in Prozorro, including in the eyes of international partners. 

Overall, the situation regarding non-competitive above-threshold procurement would be improved by reinstating the negotiated procedure. Excluding natural gas procurement within the Naftogaz group, approximately 70% of contract value currently consists of procurement conducted on grounds for which the negotiated procedure was previously required. The negotiated procedure provides for consultations with several suppliers where possible, and for a review mechanism, both of which reduce the risk of infringements and inefficient decisions.

Direct awards can be retained as exceptions for circumstances arising specifically from martial law — for example, where a contracting authority is located in an area of active hostilities. At the same time, a number of grounds should be reconsidered altogether — whether they are in fact necessary.

Volumes and dynamics of procurement

From July 30, 2024, to September 30, 2025, contracting authorities concluded 257,000 above-threshold direct award contracts. Their value at the time of the study stood at UAH 401.7 billion — nearly 28% of the aggregate current value of contracts concluded during that period. While their value share is significant, the number of such procurements is moderate, at 5.87% of the total.

Contracting authority activity is largely governed by the budgetary and planning cycle. A gradual increase in the number of lots and total contract value is observed through to January 2025, followed by a steady decline through March. This pattern mirrors the general trend across all procurement methods and may be explained by a combination of seasonal factors, the desire to use up budget appropriations before the year-end, and procurements for the new year. In the seasonality analysis, we excluded reports on natural gas procurement within the Naftogaz group, as they significantly distorted the statistics.

This means that the planning and management of budgetary resources in the direct award segment of procurement require heightened attention during the peak season, in order to prevent inefficient use of funds and potential procurement risks.

Grounds applied by contracting authorities

Over the 14 months under review, 84.4% of the value of above-threshold direct awards fell within the five most common grounds out of nearly three dozen. These were: 

1. Procurement of natural gas by an entity subject to public service obligations — 43.3%. In four such procurements totaling UAH 82.8 billion, the contracting authority indicated an incorrect ground in the electronic field; these were added manually on the basis of the justification document.
2. Absence of competition on technical grounds — 19.1%.
3. Protection of critical infrastructure and energy facilities — 8.1%.
4. Procurement on grounds of urgent need — 7.6%.
5. Open tendering canceled due to absence of tenders — 6.4%.

By number of lots, four grounds account for 78.1% of procurements:

1. Absence of competition on technical grounds — 29.9%.
2. Open tendering canceled due to the absence of tenders — 18.3%.
3. Procurement on grounds of urgent need — 15.7%.
4. Legal services — 14.2%.

Each of these grounds has its own application characteristics and associated risks.

On the ground of absence of competition on technical grounds, contracting authorities may conduct procurement from natural monopolies (where the absence of competition has been formally recognized by the Antimonopoly Committee) or in other circumstances. To roughly estimate the share of procurements conducted specifically from natural monopolies, we used the AMCU’s consolidated register of natural monopolies as of October 31, 2025. They account for 56.1% of procurements on the ground of absence of competition on technical grounds by number of lots, and 51.8% by value.

In other cases, the absence of competition may arise from vendor lock-in, where changing a contractor is disadvantageous or altogether impossible. In IT service procurement, for example, such dependency may arise from the architecture of the IT product, the absence of comprehensive technical documentation, or the use of niche programming languages. The effective monopolization by developers of the right to further develop solutions they have created can lead to rising service costs and reduced motivation for the contractor to innovate or improve quality in the absence of competition. This issue is examined in greater detail in the study on IT procurement in Prozorro. That said, not every restriction of competition is unambiguously negative or the result of bad faith on the part of the contracting authority or supplier. In a number of situations — in particular those relating to national security, defense capability, or the protection of critical infrastructure — the use of proprietary technical solutions, closed architectures, or a limited pool of contractors may be objectively justified and aimed at minimizing risks to the state. The same applies to other procurement subjects.

In direct contracts on grounds of urgent need, there is a risk that the need is not in fact genuinely urgent — or that it became urgent as a result of the contracting authority’s own actions in failing to procure earlier. The number of procurements on this ground increases toward the end of the budgetary year — between September and December, that number rises by approximately 50%. It is therefore most likely that contracting authorities also invoke this ground simply when they need to use up appropriations within the budgetary year. 

When a contracting authority concludes a direct contract because no one submitted a tender, the terms of that direct award must be identical to those of the original tender. However, whether contracting authorities comply with this in practice — for example, whether they require counterparties to provide all the certificates they requested from bidders in the failed competition — is difficult to verify, since only the contract itself is published in the system. There are instances where contracting authorities conclude several contracts instead of one. 

Among direct contracts for legal services, high-volume procurement of small amounts is combined with a number of large strategic contracts. 65% of total procurement value is concentrated among five contracting authorities: Ukrnafta JSC, National Joint Stock Company Naftogaz of Ukraine, Dnipro Metropolitan Municipal Enterprise, and the Eastern and South-Eastern Interregional Centers for the Provision of Free Legal Aid. By number of lots, however, nearly 97% belong to the last two legal aid centers, which systematically procure legal representation services for the provision of secondary free legal aid. We also investigated how other such inter-regional centers procure legal services. It emerged that they too conclude numerous small direct contracts en masse, but do not indicate the ground in the electronic field. In total, the number of such lots from legal aid centers over the period under review was nearly 73,000, totaling UAH 646 million. 

In the case of legal services, the question arises as to whether direct contracts should in fact be permitted at all for such services. Legal advisory and legal representation services, for example, are procured competitively. Similarly, competitive procedures are used to procure a range of other high-expertise services — such as consulting or audit.

What is being procured

Contracting authorities most frequently used above-threshold non-competitive procurement for the purchase of services. Services account for 57.47% of all such procurements — over 147,000 of the 257,000 lots. In monetary terms, however, services represent only 19.7% of the aggregate contract value. This indicates that the majority of service procurements are small in value but large in volume. 

The situation with goods is quite different. Although fewer in number than services — 40.15% of the total, or over 103,000 lots — goods account for nearly two-thirds of the total value of direct contracts, approximately UAH 255 billion.

Works have the smallest share — only 2.38% by number, or approximately 6,000 lots. In value terms, however, they account for 16.88% of all above-threshold non-competitive procurements (nearly UAH 68 billion). 

If one drills down into CPV codes, the value of direct awards is heavily concentrated in three categories, which together account for over 69% of total contract value. Certain categories generate a substantial share of contract value despite a relatively low number of lots.

Gaseous fuels (09120000-6) is the key category by financial volume. Despite its modest share of the total number of lots, gas accounts for the largest share of contract value. One reason is the obligation on Naftogaz Trading LLC, a gas supply company, to report on its procurement of natural gas from its own parent company and other entities within the group. This subsidiary of NAC Naftogaz of Ukraine conducts wholesale sales of natural gas to industrial consumers, budget institutions, and suppliers.

It is also worth noting that nearly 8,700 gas supply contracts were concluded via direct award following unsuccessful competitive tendering. During the period under review, only 8.3% of competitive procurements of this commodity were successfully completed.

Ukraine has committed to ensuring a free market for electricity and gaseous fuels. At the same time, the government has imposed public service obligations on Naftogaz Trading LLC — the company is required to supply natural gas to budget institutions at a preferential tariff. No separate rules have been established for this situation, so contracting authorities are required to launch open tendering, which no one participates in due to the low estimated value, and then conclude a direct contract with the Naftogaz company at a low price. Occasionally — in summer, for example, when the price of the commodity is lower — competitive tendering does succeed. Overall, however, considerable resources are expended on formal procedures.

Construction works (CPV 45000000-7): For the analysis of this sector, we consolidated all procurements by the second digit of the CPV code, since the procurement item here is defined by reference to the object rather than the classifier. Three grounds account for 75.7% of the total value of above-threshold direct awards in this sector, while together representing only 51.3% of the number of lots. These are: protection of critical infrastructure and energy facilities, urgent need, and the procurement of additional works under the main contract (up to 50%). The need to procure additional works indicates potential issues with the quality of project design and the accuracy of cost estimates at the planning stage, including owing to the complexity of such projects. 

Direct contracts on grounds of absence of any tender submission are high in number but account for a relatively small share of value. Insufficient competition in this sector may point to potential barriers to participation — for example, excessive requirements in tender documentation — or to a limited pool of contractors. 

Steam, hot water, and associated products (09320000-8) are characterized by a high share of procedures conducted on the ground of absence of competition on technical grounds. This is consistent with the nature of the market, where supply is often provided by local monopolies and the possibility of choosing an alternative supplier is limited.

Although the categories Electrical energy (09310000-5) and Distribution of electrical energy (65310000-9) belong to the same sector, they exhibit a different structure of grounds for above-threshold direct awards. In electricity procurement as a commodity, one ground dominates by a wide margin (approximately 70%), both by value and by number of procedures — namely the conclusion of a contract with a supplier of last resort or a universal service provider. A significant share (21%) also relates to absence of competition on technical grounds and cancellation of tendering due to absence of tenders. This structure indicates that in this category the use of direct contracts is substantially dependent on the quality of procurement planning and the level of market activity among suppliers.

By contrast, electrical energy distribution services within a given territory are provided by a single distribution system operator, and tariffs are set by the state regulator. In these circumstances, the predominance of the ground “absence of competition on technical grounds” — 95.5% by number and 97.6% by value of contracts — is objective in nature and results from the natural monopoly position of the relevant operators. 

Thus, despite their sectoral proximity, these two categories exhibit different models of non-competitive procurement and require different approaches to risk assessment.

Since some of the grounds for above-threshold non-competitive procurement are tied to the subject matter of the procurement, the presence of certain categories at the top was expected — for example, natural gas or legal services. Overall, the bulk of above-threshold direct contracts relates to the procurement of various energy carriers. 

In each category, a few principal grounds for direct contracts predominate. The choice of which to invoke may, however, be influenced by a variety of circumstances: technical constraints, absence of competitors, or urgent needs.

Contracting authorities do sometimes indicate incorrect grounds. For example, gaseous fuel was purchased under the ground for the procurement of hematopoietic stem cells. At the same time, among contracts for natural gas with suppliers subject to public service obligations, contracts for the supply of gas to schools do appear. However, such instances were not sufficiently widespread to have a material effect on overall statistics by value.

Helicopters, airplanes, spacecraft, and other motor-powered aircraft (34710000-7): Procurements in this category are primarily directed at meeting the urgent needs of the Armed Forces of Ukraine or are accounted for by contracting authorities located in areas of active hostilities.

Since some of the grounds for above-threshold non-competitive procurement are tied to the subject matter of the procurement, the presence of certain categories at the top was expected — for example, natural gas or legal services. Overall, the bulk of above-threshold direct contracts relates to the procurement of various energy carriers. 

In each category, a few principal grounds for direct contracts predominate. The choice of which to invoke may, however, be influenced by a variety of circumstances: technical constraints, absence of competitors, or urgent needs.

Contracting authorities do sometimes indicate incorrect grounds. For example, gaseous fuel was purchased under the ground for the procurement of hematopoietic stem cells. At the same time, among contracts for natural gas with suppliers subject to public service obligations, contracts for the supply of gas to schools do appear. However, such instances were not sufficiently widespread to have a material effect on overall statistics by value.

Monitoring of direct contracts

The State Audit Service covered 1,062 lots — 0.41% of above-threshold direct awards conducted during the period under review — through monitoring. In competitive procurement, monitoring coverage is nearly seven times higher, at 2.73% of the total number of procurements eligible for monitoring. The same trend is observed in value terms: monitoring covers only 2.3% of the aggregate value of above-threshold direct award contracts, compared with 20.9% in the competitive segment. Even excluding the value of natural gas procurement within the Naftogaz group, as well as procurements from natural monopolies, suppliers of last resort, and universal service providers, coverage reaches only 5% of the value of direct contracts. This gap relative to competitive procurement indicates that direct awards remain largely outside the monitoring focus. Yet they too require oversight, since they do not provide for competition or safeguards such as review before the AMCU. 

The largest monitoring volumes relate to the construction sector (CPV 45) and the energy sector (CPV 09), where contract values run into the billions of hryvnias. Overall, monitoring activity is concentrated in sectors with high financial exposure and an elevated risk of inefficient use of funds. Significant attention is also devoted to industrial machinery (CPV 42) and repair and maintenance services (CPV 50).

At the same time, the top three sectors by number of monitoring instances include accommodation, catering, and retail trade services (CPV 55), with a combined value of over UAH 90 million. In 98% of cases, these involve catering services for educational establishments. Such procurements were monitored primarily by the directorates of the Western Office of the State Audit Service, in particular in Khmelnytskyi and Ternopil regions, where they were examined three times more frequently than construction procurements. Violations found under this CPV related predominantly to the failure to publish, or untimely publication of, the justification for the procurement ground. Accordingly, most monitoring instances resulted in auditors requiring explanatory work and training of authorized persons to be carried out.

Overall, violations in above-threshold direct awards were identified by auditors during monitoring in 67.8% of cases. This figure is slightly higher than in competitive procurement, where it stands at 61.3%. While a gap of 6 percentage points is not critically large, it may indicate that direct non-competitive procurement carries elevated risks of non-compliance with legislation. It is important, however, to understand what types of violations auditors are actually identifying. Doing so comprehensively is difficult, since in the electronic field on Prozorro, the State Audit Service almost invariably classifies the type of violation as “Other violations of procurement legislation,” and details must be sought manually in the text of the audit finding. It would therefore be worthwhile to update the violation categories in the system and apply them consistently, so as to enable analysis of the problems arising in direct procurement.

In summary, above-threshold direct awards warrant greater monitoring coverage. The State Audit Service generally selects sectors for monitoring well — concentrating on construction, energy, and related sectors, where procurement volumes are largest and corruption risks are highest. The Western Office’s excessive focus on catering service procurements for educational establishments is questionable. Such procurements are not particularly widespread, are comparatively small in value, and the violations associated with them largely concern failure to publish or untimely publication of information. The Office would therefore benefit from refining its methodology for selecting above-threshold direct awards for monitoring.

Contract modifications

Modifications were made to contracts in 36,700 lots, representing over 14% of the total number of above-threshold direct award procurements. The aggregate value of modified contracts reached UAH 140 billion, or nearly 35% of the total value of all contracts in this segment. By comparison, in competitive procurement, modifications occurred more frequently: they affected 21.8% of lots and 44.7% of the total value of contracts over the corresponding period. The data indicate that while contract modifications in above-threshold direct awards are recorded less frequently than in competitive procurement, their financial scale remains significant. The majority of modified contracts relate to cases where the ground for using direct awards was absence of competition on technical grounds (41.81% of all modified contracts) or absence of tenders in open tendering (32.53% of all modified contracts). Their shares are high because these are overall the most common grounds for above-threshold direct awards. If one looks at what share of contracts were modified within each direct award ground, the leaders are as follows:

• Procurement of natural gas by an entity subject to public service obligations — modifications in 38.34% of contracts (41.2% for Naftogaz group procurements specifically)
• Procurement of additional works or services from the same economic operator, up to 50% of the price of the previous/existing contract — 28.11% of contracts
• No tender submitted — 25.38%.

In order to assess the justification for these modifications and the potential risk of abuse, it is necessary to conduct further in-depth analysis of their substance.

Comparison with the negotiated procedure 

The negotiated procedure was abandoned in the first year of the full-scale invasion in favor of an even more flexible mechanism. Under Resolution No. 1178, the grounds for the negotiated procedure became the grounds for direct awards. In the early period, such changes may have been justified to meet the needs of contracting authorities, in particular for defense, energy, and critical infrastructure. However, this expansion of contracting authority discretion creates risks of manipulation and circumvention of competitive procedures, whereby a portion of procurements are concluded as direct contracts without adequate scrutiny of the grounds invoked.

Over the comparable period from July 30, 2020, to September 30, 2021 — the years immediately preceding the full-scale invasion — contracting authorities conducted 100,600 negotiated procedures, under which contracts totaling UAH 101.7 billion were concluded. This amounts to an average of 7,200 contracts per month valued at UAH 7.3 billion. This is considerably lower than the current figure — during the period under review, an average of 18,400 above-threshold direct award contracts per month were concluded, valued at UAH 28.7 billion. In other words, the number of such lots has more than doubled overall, and the contract volume has increased nearly fourfold. Two factors should of course be taken into account:

• High inflation over recent years, and
• The increase in the number of grounds for above-threshold direct awards.

In both the negotiated procedure and the direct awards conducted under the wartime special rules, absence of competition on technical grounds remains the leading ground by number and by value of expenditure. Under the negotiated procedure, this ground accounts for nearly 52% of lots and over 50% of contract value. The same ground also ranks first in direct awards under the special rules — nearly 30% of contracts and 19% of total contract value. The list of the most common grounds under the negotiated procedure, which together account for 95.7% of contract value, is as follows: 

1. Absence of competition on technical grounds — UAH 51.8 billion (50.9%) by contract value and 52% by number of lots.
2. No tender submitted — UAH 21.4 billion (21%) and 30% of lots.
3. Need to carry out additional construction works — UAH 15.6 billion (15.3%) and 2.5% of lots.
4. Urgent need (across all sub-paragraphs of this ground in the Law) — UAH 6 billion (5.9%) and 3.9% of lots. 
5. Need to protect intellectual property rights — UAH 2.6 billion (2.6%) and 2.2% of lots.

By type of procurement subject, the distribution in negotiated procedures is similar to that in direct contracts. The only material difference was in the value of goods and services. Previously they accounted for 37% and 48% of contract value respectively, whereas now goods take a larger share — 63% versus 20%. 

The largest expenditure under the negotiated procedure was also concentrated in energy resources and construction. At that time, however, natural gas did not dominate contract values to nearly the same extent. Reporting on natural gas procurement within the Naftogaz group did not yet exist, and procurements from the supplier of last resort were also significantly smaller in volume.

Excluding gas procurement within the Naftogaz group, grounds on which contracting authorities previously conducted the negotiated procedure currently account for 70.2% of the value of above-threshold direct awards. This means that a substantial share of the direct procurement market could potentially have been channeled through a more controlled procedure that ensures transparency and competitiveness.

The application of the negotiated procedure offers significant institutional advantages: it allows for consultations with several potential suppliers prior to contract conclusion and provides a mechanism for the review of decisions. Even where review is rare — occurring in fewer than 1% of cases — the very existence of this instrument has a deterrent effect and reduces the risk of non-competitive procurement, including procurement justified by an alleged absence of competition.

Reinstating the negotiated procedure can thus serve as an effective mechanism for enhancing transparency and minimizing risks in the direct award segment, where oversight and competition are currently limited.

Conclusions

From July 30, 2024 to September 30, 2025, contracting authorities concluded 257,000 above-threshold direct award contracts totaling UAH 401.7 billion. This represents approximately 6% of the total number of contracts in the system and nearly 28% of their aggregate value. While the number of such procurements is relatively modest, their financial scale is significant.

Nearly half of the total value (43.3%) of above-threshold direct awards relates to natural gas procurement by a contracting authority on which the government has imposed public service obligations — that is, to reports on the sale of this commodity within the Naftogaz group. This raises the question of whether such reports on Prozorro are necessary at all: prior to 2023, they were not published.

A large volume of gas procurement — 8,700 contracts — is also concluded as direct awards following unsuccessful tendering. Contracting authorities succeeded in completing only 8.3% of such competitive procurements. A solution must be developed with the involvement of both the Ministry of Economy and the Ministry of Energy, since a quasi-free gas market has now taken shape in public procurement: contracting authorities nominally have the option of purchasing from any supplier, but are in practice compelled to conduct fruitless tendering.

Overall, the situation regarding non-competitive above-threshold procurement would be improved by reinstating the negotiated procedure. Excluding natural gas procurement within the Naftogaz group, approximately 70% of contract value currently consists of procurement conducted on grounds for which the negotiated procedure was previously required. The negotiated procedure provides for consultations with several suppliers where possible, and for a review mechanism, both of which reduce the risk of infringements and inefficient decisions.

Particular attention should also be paid to procurement on grounds of urgent need — the use of this ground increases by approximately 50% toward the end of the budgetary year. It is therefore likely that a portion of such procurements are conducted not so much because of genuinely urgent circumstances as because of budgetary management challenges. This, however, requires further investigation.

In addition, it would be advisable to technically link, within Prozorro, unsuccessful open tendering to the subsequent non-competitive procurement conducted as a result of its cancellation. This would enhance the transparency of such procurement and improve oversight of contracting authorities’ compliance with the terms of the original tender in the context of direct awards.

Direct awards can be retained as exceptions for circumstances arising specifically from martial law — for example, where a contracting authority is located in an area of active hostilities. At the same time, a number of grounds should be reconsidered altogether — whether they are in fact necessary. Some grounds have been indicated in the system in only a handful or a few dozen procurements. It would also be appropriate to explore alternative procurement methods for the purchase of free legal aid services — for example, framework agreements. 

For both the negotiated procedure and direct contracts to function effectively, it is important to train contracting authorities — in particular on how best to determine the value of such contracts and select a supplier.

Finally, above-threshold direct awards are currently subject to monitoring at a rate seven times lower than competitive procurement — only 2.26% of value compared with 20.9%. They require greater oversight, with a focus on higher-value procurements and material violations. 

This research was prepared within the framework of the “Digitalization for Growth, Integrity, and Transparency” (UK DIGIT) project, implemented by the Eurasia Foundation and funded by UK Dev.

The research was produced with the financial support of the UK Government’s International Development Assistance Programme. The contents of this material are the sole responsibility of Transparency International Ukraine; the views expressed do not necessarily reflect the official policy of the Government of the United Kingdom.

This research was developed by

Team lead: 

Ivan Lakhtionov, Deputy Executive Director of TI Ukraine for Innovative Projects

Authors of the research:

Pavlo Dehtiariov, researcher of DOZORRO TI Ukraine project

Kateryna Rusina, Project Manager at DOZORRO

SUMMARY

More than half (55%) of the hard coal procurements checked contained unlawful requirements to provide additional documents and set extremely short deadlines for their submission — conditions that are potentially discriminatory. 

Draft contracts frequently also included provisions under which failure to submit documents within the short deadline was automatically treated as a written refusal to conclude a contract, resulting in the rejection of the tender and disqualification of the participant.

In the representative sample, winning tenders in procurements with unlawful requirements and disqualifications were on average nearly 40% more expensive than the lowest-priced tenders that contracting authorities had rejected.

This practice demonstrates that the simplified RFP procedure sometimes fails to function as intended and partially undermines the logic of the electronic catalog as a fast and accessible procurement tool. To enhance transparency in Prozorro Market procurement, the DOZORRO TI Ukraine project recommends the following:

1. Closing the hard coal category in Prozorro Market.
2. Introducing a mechanism for challenging procurement requirements, which would give businesses a means of protecting their interests and could significantly reduce the risk of abuse. 
3. Enabling monitoring of RFPs by the State Audit Service, to bring this procurement under oversight.
4. Granting the Ministry of Economy the authority to approve standard contracts for specific commodity categories, thereby simplifying their preparation and ensuring legislative compliance, and converting them into electronic contracts.
5. As electronic contracting develops, introducing automated pre-screening of draft contracts in the electronic catalog that, upon detecting potentially unlawful language, would generate alerts for supervisory authorities.
6. Developing guidance notes for authorized persons at contracting authorities, with recommendations on preparing and conducting procurement through the e-catalog.

Together, these measures can make Prozorro Market procurement more transparent, predictable, and user-friendly for both contracting authorities and market participants.

What makes Prozorro Market procurement distinctive

The specifics of procurement through the e-catalog (Prozorro Market) are governed by a separate Procedure for the Formation and Use of the Electronic Catalog (the Procedure), approved by Cabinet of Ministers Resolution No. 822 of September 14, 2020. The core idea behind this method is to simplify and expedite procurement. Qualification — that is, the verification of supplier documents — is handled by the catalog administrator. RFPs are not subject to auditor monitoring, and participants have no right to challenge either the terms or the decisions of contracting authorities. 

An RFP is generated by completing electronic forms in which the contracting authority specifies a limited set of details: the name of the procurement item, estimated value, quantity, delivery period and location, payment terms, and the deadline for submitting price proposals. 

When publishing an RFP, a contracting authority may only specify information on the specifications of the goods and their permissible values within the item specification set by the catalog administrator. The government recently amended the Procedure (Cabinet of Ministers Resolution No. 33 of 14 January 2026), further specifying the list of details that may be indicated. It was clarified that, in addition to the information already mentioned, a contracting authority may specify the delivery terms, the amount, type, period, and conditions for providing and returning contract performance security (where it requires such security to be provided), as well as a requirement that the country of origin may not be the Russian Federation, the Republic of Belarus, or the Islamic Republic of Iran.

The draft contract, which forms part of the RFP, must not conflict with the information the contracting authority has entered in the electronic fields, and may contain other conditions that are provided for by law in respect of such a contract and that the contracting authority considers necessary for the performance of the contract during its term. It is prohibited to include in an RFP requirements directed at the procurement participant or the selected winner, or to demand any additional documents not provided for by the Procedure. This is a prerequisite for the transparency of such procurement and allows potential suppliers to properly assess its terms and make an informed decision on whether to participate.

Hard coal procurement through RFPs

This study focuses on CPV class “Solid fuels” (code 09110000-3), with an additional targeted sample specifically of hard coal procurements conducted through RFPs on Prozorro Market. For the analysis, we selected procurements announced between September 1, 2024, and September 30, 2025. 

The total number of lots included in the analysis is 858.

For coal procurement, contracting authorities opt for the RFP method twice as often as for open tendering — Prozorro Market accounts for about 64% of such lots. The estimated value of RFP-based procurements is also twice as high, representing 62% of the value of competitive coal orders. Significantly more participants come forward for Prozorro Market procurements — an average of 6.58, compared with 2.04 in open tendering. However, the number of disqualifications is also much higher. 

Where contracting authorities reject 16.5% of tenders in open tendering, the corresponding figure in RFPs is approximately 55%. It is important to note in this context that in open tendering, approximately 22% of lots are accompanied by complaints, which underscores the significance of the review mechanism as an instrument for overseeing the transparency and openness of procurement.

All profiles in the coal category on Prozorro Market now require goods to comply with the DSTU 7146:2010 standard (the Standard). Under the Standard: 

• When concluding a coal supply contract for household needs, a certificate of genetic, technological, and quality characteristics must be presented (clause 7.10). In practice, however, the application of this provision of the Standard creates a conflict in law enforcement, since the Procedure expressly prohibits the demand for any additional documents under the RFP. 
• A coal consignment is considered accepted for dispatch once a composite commercial sample has been taken from it in accordance with DSTU 4096 and a quality certificate has been issued (clause 7.3). Accordingly, this certificate may be requested at the time of delivery.

The requirement to provide these additional documents occasionally appears in lot notes — in 6% of procurements. These are optional fields in which contracting authorities may include supplementary information about the procurement — for example, drawing potential suppliers’ attention to terms of particular importance to them. However, in addition to established documents, contracting authorities in 1.4% of lots also required additional documents not covered by the Standard. Notes carry no legal force, but this already reveals at this stage that some contracting authorities conducting coal procurement tend to impose excessive requirements on potential bidders that are not provided for by law.

 

Additional requirements in draft contracts 

For the detailed analysis, we randomly selected 270 lots from the full dataset of 858. The sample was calculated using a 95% confidence interval and a 5% margin of error. The sample is therefore representative and reflects the overall picture of coal procurement through Prozorro Market.

Under the legislation in force at the time of the procurements under review, an RFP must not contain requirements directed at the supplier or any documents not provided for by the Procedure. In the draft contract, as elsewhere, a contracting authority may specify only information on the characteristics of the goods and their permissible values within the item specification set by the administrator.

The Standard does not require any documents to be provided in respect of coal other than the certificate of genetic, technological, and quality characteristics at contract signing, and the quality certificate for the coal consignment at delivery. Accordingly, only the certificate of genetic, technological, and quality characteristics and contract performance security in the form of a bank guarantee may lawfully be required at the contract conclusion stage (Article 27(1) of the Law on Public Procurement). The right to require a bank guarantee is now also expressly provided for in the amended Procedure — the contracting authority may specify in the draft contract the amount, type, period, and conditions for providing and returning such security. Any other documents demanded from participants or the selected winner that are not provided for by law are unlawful.

The findings, however, point to systematic non-compliance with these requirements. The documents most frequently demanded by contracting authorities in draft coal supply contracts were as follows.

Overall, 55% of draft contracts (148) in the sample contained requirements to provide additional documents not provided for by either the Procedure or the Standard. In isolated cases, these were manifestly excessive and included documents with no functional connection to the subject matter of the procurement — for example, a certificate of proficiency in the state language, copies of the coal producer’s financial statements for prior years, or an expert market assessment. In the majority of procurements, however, the violations consisted in requirements to provide documents that formally relate to the supplier’s operations but do not affect the physical or quality characteristics of the goods and are not provided for by law, in particular, quality management system certificates, proof of experience in performing similar contracts, or extraction permits. This points to a systemic and uncontrolled expansion of supplier requirements. A full list of documents whose provision was recorded in the study is available via the table at the link.

In practice, contracting authorities are using the draft contract as a tool for introducing additional barriers to participation — contrary to the simplified nature of the RFP procedure and creating disproportionate and discriminatory conditions for suppliers. This eliminates competition, impedes access for bona fide participants, and undermines the logic of the e-catalog as a fast and transparent procurement tool.

 

When additional documents must be submitted

The review of draft contract terms also revealed that in 52% (141) of procurements, contracting authorities set extremely short deadlines — between 1 and 48 hours — for submitting an additional, sometimes excessive, list of documents required for contract conclusion.

Such tight timeframes place suppliers on an unequal footing, functioning as a barrier and a covert mechanism for selecting a predetermined winner while screening out those objectively unable to meet the requirements in time.

In this context, it is important to determine from what moment the clock actually starts running. The analysis shows that contracting authorities use one of two reference points: 

• the opening of proposals, which virtually coincides with the submission deadline;
• the determination of the selected winner (receipt of an automated system notification).

When publishing an RFP, a contracting authority may set the proposal submission deadline at its own discretion, subject to the minimum requirements of clause 57 of the Procedure — the deadline may not be less than two working days from the date of publication. Formally, this requirement may be satisfied even where the deadline falls on late evening, midnight, or a weekend. 

Particular concern is raised by the combination of nighttime proposal openings and extremely short deadlines. In 25% of lots with short deadlines, suppliers are given between 1 and 4 hours to provide additional documents after the opening of proposals or the determination of the winner. In half of those cases, the proposal submission deadline and automatic opening fall between 11:00 p.m. and 01:00 a.m. compelling suppliers to submit documents during nighttime hours when most responsible personnel are unavailable. 

As regards the moment of determining the selected winner, under clauses 61–62 of the Procedure the winner is determined automatically upon expiry of the submission deadline and evaluation by the electronic system. In practice, however, this long remained in a gray area, with two competing interpretations: immediately upon opening of proposals, or upon the contracting authority’s confirmation. This has now been resolved. Under the amended clause 62, the contracting authority determines the winner and records its intention to conclude a contract in a protocol that is automatically generated and published in the system. Notification of intent is sent automatically within one day of publication. The start of the compliance period is therefore now more transparent and clear to all suppliers.

Even so, where contracting authorities indicated that the winner would be determined “upon receipt of a system notification,” the combination of short deadlines and uncertainty about when that notification would arrive created additional pressure. Winners were compelled to await confirmation without being able to predict the precise start of the compliance period, making timely preparation of documents more difficult. In any case, requirements to provide excessive or unlawfully demanded documents remained unlawful regardless of how that moment was defined. 

Imposing additional requirements through the draft contract not only constitutes a procedural violation but also contradicts the fundamental principles of contract law and public procurement. Contract terms apply only to the parties that have actually concluded the contract. 

Equally legally incorrect is the widespread practice of including provisions such as: “In the event that the Contracting Authority is not provided with such documents (as defined in the draft contract), the Contracting Authority shall treat this as a written refusal by the participant to sign the contract, and shall accordingly reject such participant’s proposal pursuant to subclause 2 of clause 64 of Cabinet of Ministers Resolution No. 822 of September 14, 2020.”

Subclause 3 of clause 64 of the Procedure provides as a ground for rejection only the selected winner’s written refusal to sign the contract. Under Article 207 of the Civil Code, a legal transaction is considered to have been effected in writing if its content is recorded in document(s) exchanged by the parties, including through information and communication systems. A refusal to conclude a contract has legal force only if made in writing by a person with authority to sign it — the selected winner, not merely a procurement participant — and communicated to the other party. The failure of a participant or winner to take certain actions or submit documents, even if such actions are provided for in the draft contract, cannot therefore be treated as a written refusal to sign.

Such conditions create fertile ground for abuse. Formal procedural compliance combined with additional barriers at the contract conclusion stage creates the appearance of competitive tendering while in practice restricting participation to a narrow pool of suppliers who, in certain cases, may be connected, resulting in procurement at inflated prices and inefficient use of public funds.

For example, in February 2025, a gymnasium in Kirovohrad Region procured three types of coal. The draft contract required participants to submit over a dozen documents — including an underground mining permit and financial statements — within 24 hours of the opening of proposals. The contracting authority rejected 8 out of 9 proposals and concluded a contract at the highest price of UAH 299,700, while the lowest offer was UAH 182,000. In another case, the Vanchykivtsi Village Council’s coal procurement required a set of documents within 24 hours of the winner being determined. According to suppliers, proposals were rejected even where the documents were submitted. The contracting authority rejected 8 out of 9 proposals and concluded a contract for nearly UAH 2.6 million, against a lowest tender of UAH 1.5 million. 

Overall, in the analyzed sample, the gap between the lowest and winning tenders in procurements with unlawful requirements or short deadlines amounted to UAH 28.5 million — contracts ended up costing nearly 40% more than they could have.

Such violations have already drawn the attention of law enforcement and courts. A number of criminal proceedings have been opened for misappropriation of public funds in RFP procedures (case numbers 42024222140000071, 12025082210000241, 12025160000000217). A court judgment has also been handed down (No. 573/1300/25) finding that the contracting authority’s actions — in particular, setting a 1-hour deadline for an excessive list of documents — resulted in the rejection of the majority of economically advantageous proposals, contract conclusion at an inflated price, and inefficient use of public funds.

As part of the study, we spoke with suppliers about their direct experience practice of applying additional requirements and barriers at different stages of procurement. Some noted that in most procurements with additional requirements, contracting authorities either tailor conditions to a specific supplier or mechanically copy draft contracts from other RFPs. They also noted that contracting authorities sometimes justify excessive requirements by the desire to minimize their own risk in connection with potential post-delivery inspections. One supplier reported instances of pressure from competitors filing complaints with law enforcement, after which contracting authorities are required to justify the price-to-quality ratio of the coal delivered. 

We also approached individual contracting authorities for their perspective. Most said the primary purpose of such measures is to “play it safe” against the risk of receiving low-quality coal mid-heating-season or dealing with unreliable suppliers or fraudsters. They noted that even where contractual penalties exist, they often lack the resources for lengthy court proceedings. Regrettably, some contracting authorities, when their attention was drawn to potential violations, responded dismissively or claimed not to recall the details. It is also worth noting that some contracting authorities interpret certain legislative provisions at their own discretion — for example, taking the view that requiring documents and contract signature within 24 hours does not conflict with the statutory five-calendar-day deadline for concluding a contract.

Conclusions

55% of RFPs for coal procurement contain requirements to provide documents not provided for by law, and 52% set short submission deadlines.In the representative sample, winning tenders in procurements with unlawful requirements and disqualifications were on average nearly 40% more expensive than the rejected lowest-priced tenders.

These statistics relate exclusively to coal category procurements, not to all RFPs on Prozorro Market. The situation may vary by commodity, and there may be categories where this problem is not systemic. The study does, however, identify specific vulnerabilities in the RFP process and illustrates the potential scale of problems in certain categories.

• The most effective remedy would be a review mechanism in Prozorro Market, currently being developed under the draft new Law on Public Procurement. However, that process is lengthy and the new mechanism will not cover all RFP-based procurements. 
• The situation of unlawful rejections on grounds of alleged refusal to sign would improve with e-contracting, which is also yet to come. But these need not be the only solutions. 

It is therefore worth introducing a range of additional instruments to reduce abuse risks, enhance legal certainty, and strengthen contracting authority capacity.

• Enable monitoring of RFPs by the State Audit Service.

The State Audit Service does not currently monitor RFPs. Whether it has the legal authority to do so is disputed due to ambiguous wording in the procurement rules. It would therefore be advisable either to confirm formally that monitoring is permitted or to amend the legislation accordingly. Even selective monitoring and sanctioning of individual violators can deter future violations.

• Develop and introduce standard contracts.

The e-catalog currently has no mandatory contract templates for commodity categories, and the Ministry of Economy lacks the authority to approve them. The Ministry should be granted that authority and standard contracts approved that set out the permissible terms for coal procurement, leaving contracting authorities to fill in only the key parameters — delivery and payment terms and quantity. This would unify procurement terms, ensure legislative compliance, and simplify preparation without restricting contracting authorities from reflecting their specific needs. These should subsequently be introduced as e-contracts.

• Close the hard coal category in Prozorro Market.

Given that over 50% of procurements in this category contained unlawful requirements, coal procurement through RFPs should be restricted. Procurement would then take place through a method under which businesses may challenge discriminatory requirements and auditors may carry out monitoring. 

• Develop and introduce universal guidance notes for procurement through the electronic catalog.

Such notes will help authorized persons conduct RFPs correctly, apply legislative provisions accurately, and avoid excessive or discriminatory conditions — contributing to more transparent, predictable, and high-quality catalog procurement.

• As electronic contracting develops, introduce automated pre-screening of draft contracts in the e-catalog as a risk indicator for supervisory authorities.
  The algorithm could detect keywords and language conflicting with legislation and the Procedure — such as “certificate,” “declaration,” “permit,” “within X hours of the opening of tenders.” Where such indicators are found, the system could generate risk signals or reports for supervisory authorities. This will become technically feasible only once electronic contracting is in place, as the system currently cannot automatically read and analyze file contents.

These recommendations will help eliminate one method of circumventing the rules in coal procurement and more broadly improve the efficiency and transparency of RFP-based procurement. Efforts must also continue to find approaches that allow contracting authorities to minimize the risk of receiving substandard goods, while protecting bona fide participants from artificially created obstacles and bad-faith practices.

This research was made possible with the support of the MATRA program of the Embassy of the Kingdom of the Netherlands in Ukraine. Responsibility for the content lies with the author and does not necessarily reflect the official position of the Embassy.

This research was developed by

Team lead: 

Ivan Lakhtionov, Deputy Executive Director of TI Ukraine for Innovative Projects

Authors of the research:

Viktoriia Hermasheva, Projects Assistant, TI Ukraine

Kateryna Rusina, Project Manager at DOZORRO

Research results

The average level of implementation of 40 indicators in the “Open Data” block is 23.1%. 

Kyiv achieved the highest score, with 44.3 out of 100 possible points. One position below is Lutsk with 40 points, followed by Kropyvnytskyi and Lviv, each with 35.3 points. The lowest scores were recorded by Odesa (0), Poltava (0), and Chernihiv (9.9).

Kyiv owes its leading position to two factors. First, in 2024, the city authorities approved a key regulatory document that incorporated the Ministry of Digital Transformation’s updated recommendations and began systematically populating the Kyiv City State Administration’s account on data.gov.ua in line with those recommendations. Second, unlike Lutsk, Kropyvnytskyi, Lviv, and Dnipro, Kyiv currently publishes its datasets exclusively on the Unified State Open Data Web Portal, so data transfer (harvesting) issues on data.gov.ua did not affect the city’s results. 

As for the cities ranked immediately behind Kyiv, Lutsk, Lviv, and Dnipro outperformed the capital in the block of indicators related to open data policy. The critical issue for them, as well as for Kropyvnytskyi, was that the Ministry of Digital Transformation had not established monthly automated transfer of data from local open data portals to the national portal. 

The analysts extracted data for the analysis from data.gov.ua on December 5, 2025. As later confirmed in the Ministry’s response to an inquiry from the Transparent Cities program, data transfer from the open data portals of Kropyvnytskyi and Lutsk City Councils was carried out on September 10, 2025; from Lviv City Council, on September 15, 2025; and from Dnipro City Council, on October 28, 2025. As a result, in each of these cities, more than 10 “EU integration” datasets with monthly or weekly update frequency had no chance of passing the verification filter for timely updated resources. 

In two regional centers — Odesa and Poltava — which scored zero points in the study, the same key problems were identified: these cities have no dedicated thematic section on their official websites consolidating key open data information, and as of December 5, 2025, the electronic accounts of Odesa City Council and Poltava City Council on data.gov.ua were empty. That said, the operating context of these two cities is very different.

The official website of Odesa City Council is designed so that the number of “Open Data” pages equals the number of local authority structural units. This design approach does not comply with contemporary European approaches to user-centricity and the creation of one-stop shops/single points of contact. Odesa’s results are also driven by the absence of approved key open data policy documents and the lack of centralized oversight over dataset publication.

In Poltava, the context is entirely different. In December 2024, the city council approved the Procedure for Publishing Datasets of Poltava City Council and Its Executive Bodies in Open Data Format. There is also a dedicated website, “Open Data of Poltava City Council,” where up-to-date datasets can be found. However, the municipality did not take the next steps—it did not consolidate all user-relevant information about this area in one place on the official website, and it does not comply with the state requirement to ensure convenient and clear dataset publication on data.gov.ua under the “single window” principle.

Local self-government policy in the open data

The average level of implementation across the 10 indicators related to local open data policies is 48.8%. The highest result was recorded by Lutsk, with 20 out of 22 points.

The study results showed that 8 of 11 cities have a “single entry point” to open data on their official websites. Only the municipalities of Odesa, Poltava, and Kharkiv failed to provide convenient public access to a section with the relevant information.

7 of 11 city councils published, in that section, links to an administrative act that approved, within the last five years, a unified List of Datasets subject to publication in open data format. The most recent was the order of the Lutsk mayor dated February 26, 2025. All seven documents assign responsibility for each dataset to specific information managers (executive bodies, municipal enterprises, and institutions).

9 of 11 cities paid insufficient attention to the purpose for which data are published—namely, services built on that data. Journalists and researchers can, where needed, download files in CSV/JSON/XML formats, analyze them, and draw appropriate conclusions. But the main purpose of publishing local self-government datasets is to ensure the data are continuously used, including for designing convenient tools for residents (maps, chatbots, mobile applications, etc.). Only the municipalities of Lutsk and Lviv provided links, in their respective sections, to e-services built on their datasets.

As of December, 8 city councils had not published links to decisions on accession to the International Open Data Charter. Some had genuinely not joined it, while Zaporizhzhia, Kropyvnytskyi, and Khmelnytskyi did not provide information on their accession.

In addition, 8 cities do not include, in their approved local self-government List of Datasets, information on the format in which each specific dataset is published. Lack of alignment on this issue at the level of a key policy document leads to recurring problems for information managers during dataset moderation on data.gov.ua, with non-existent formats such as excel, API, and others appearing in resource descriptions. It also makes it easier for municipalities to change publication formats from time to time without considering the problems this creates for regular users.

Open data publication

The average level of implementation across the 30 indicators related to publication of “EU integration” datasets in 9 cities (excluding Zaporizhzhia and Kharkiv) is 15.9%. The highest result was recorded by Kyiv, with 27.3 out of 78 points.

First, the analysts checked whether the required datasets were present. The study found that, of the 270 “EU integration” datasets that city councils were expected to publish in their single electronic accounts on data.gov.ua, 148 (55%) had been published as of December 5, 2025. As for the rest, 112 datasets could not be found at all; in 5 cases, there was no single consolidated city-level dataset; and in another 5 cases, datasets had been formally created but the data were inaccessible. 

The largest numbers of “EU integration” datasets were published by Kropyvnytskyi and Lutsk (28 each), followed by Lviv (26). The accounts of Odesa and Poltava City Councils were empty, while the Chernihiv City Council account contained only 2 datasets.

A positive result was that 7 of 9 cities published the dataset “Roll-call results of voting of local council members at plenary sessions of the local self-government body.” The weakest-performing datasets were “Data on the location of electric vehicle charging stations” (published only by Lutsk) and “Data on operational characteristics of buildings of municipal enterprises, institutions, and organizations where energy management systems have been implemented” (available only in Kropyvnytskyi and Lutsk).

At the second step, analysts checked whether the update frequency specified in the files of existing datasets corresponded to the Ministry of Digital Transformation’s recommendations. Matches were recorded in 70 cases, while 97 cases showed mismatches. In one case, analysts had to record partial correspondence, because for the dataset “Lists of legal regulations,” the Kyiv City State Administration specified the recommended update frequency (monthly), whereas Kyiv City Council indicated its own frequency (annually).

The highest number of matches was recorded in Kropyvnytskyi (21 of 28 datasets), while the highest number of mismatches was recorded in Lviv (20 of 26 datasets). In 11 cases, Lviv specified an update frequency not provided for at all in the Ministry’s recommendations—“immediately after changes are made.” In subsequent analysis, timely updates for these datasets were assessed against the Ministry’s recommended frequency. In 4 cases, Khmelnytskyi indicated that the dataset “is no longer updated.”

After that, program experts analyzed whether city councils complied with the timeliness principle. For each of the 144 datasets, they checked whether the datasets contained resources updated on time in accordance with the dataset file. It turned out that only 42 datasets had timely updated resources, while 93 did not. In 9 cases, this check was not performed because the data update frequency was “more than once a day.”

It was at this stage that Kyiv moved into first place. It had 12 timely updated datasets out of 20. Kropyvnytskyi and Lutsk each had 9 out of 28, and Lviv had 7 out of 26. 

A major stumbling block was datasets that were supposed to be updated monthly. Of 68 such datasets, 52 belonged to cities that publish data on their own local open data portals (Dnipro, Kropyvnytskyi, Lutsk, and Lviv), after which harvesting to data.gov.ua should occur. The analysis showed that only 2 of these 52 datasets contained timely updated resources: Kropyvnytskyi’s dataset with API endpoints for obtaining data from the e-Liky system, and Dnipro’s dataset “Lists of legal regulations,” whose resources were likely added manually to data.gov.ua rather than transferred via harvesting. By contrast, another monthly Dnipro dataset—“Action plan for drafting legal regulations”—was updated on the local open data portal on November 18, 2025, but on December 5, 2025 it was not counted among timely updated datasets on data.gov.ua because the Ministry had last harvested data on October 28, 2025.

Study leaders: availability on data.gov.ua of timely updated datasets with monthly update frequency as of December 05, 2025.

Accordingly, there are solid grounds to conclude that Kyiv’s rise to first place was driven by the fact that, unlike its main competitors, the capital does not depend on transferring data from its own open data portal to data.gov.ua.

The next four steps aimed to verify whether city councils comply, when publishing data, with one of the most important principles—the principle of interoperability. This principle requires “ensuring interaction, combinability, and compatibility of public information in open data format, including through the use in each dataset of unified identifiers of objects, other object attributes, a unified dataset structure, and unified data resource structures, which makes it possible to compare information within a dataset or resource, as well as with other datasets of the same information manager or other information managers.” 

At the fourth stage, program experts analyzed whether timely updated resources fully reflect the dataset structure prescribed by the Ministry of Digital Transformation’s recommendations. This check was meaningful only for datasets that should consist of two or more resources. For example, the dataset “Data on e-petitions” should consist of two resources: petitions (petition data) and votings (data on persons who signed the petition). As a result of checking 15 datasets with timely updated resources, it was found that the structure of only 3 fully complied with the Ministry’s recommendations, 9 partially complied, and 3 did not comply at all.

Next, for each of the 42 timely updated datasets and 9 datasets with an update frequency of “more than once a day,” experts checked whether resource names followed the Ministry’s recommendations. In 18 cases, the recommendations were fully followed; in 12, partially followed; and in 21 cases, not followed at all. It was at this stage that Lutsk finally lost its chance to compete with Kyiv for leadership. It turned out that all Lutsk resources were titled in Ukrainian, whereas English-language names are required.

At the sixth stage, for each of the 42 datasets, analysts checked whether file names of timely uploaded resources followed the Ministry’s recommendations. In 17 cases, the recommendations were fully followed; in another 10, partially followed. If the previous step showed that the Kyiv City State Administration and Kyiv City Council used different languages for resource names (English and Ukrainian, respectively), then at this stage the divergence shifted to Latin-script conventions (for example, where KCSA uploads a file named projects_2025-12-02.csv, Kyiv City Council uses plan-diialnosti-z-pidgotovki-proiektiv-reguliatornikh-aktiv_2025-rik.xlsx), which is why the city’s overall result was assessed as partial implementation. 

Finally, at the last step, for each of the 42 timely updated datasets and 9 datasets with an update frequency of “more than once a day,” experts checked whether the structure of timely updated resources included all attributes (fields) specified in the Ministry’s recommendations. It was found that 15 datasets had resources built exactly in line with the recommendations. These included 5 of Kyiv’s 19 datasets and 5 of Lutsk’s 9 datasets. In 8 cases, partial implementation was recorded. The attributes of resources forming 28 datasets differed from those recommended. 

As a result, out of the 270 datasets reviewed by the program’s analysts, only 6 (2.2%) are published on time and in line with all recommendations of the Ministry of Digital Transformation. These are 4 datasets from Kyiv, 1 from Dnipro, and 1 from Lviv.

 

Key findings and recommendations 

The average level of implementation by 11 cities of indicators in the “Open data” block (23.1%) was substantially lower than the average levels in other Euro Index blocks—“Openness and public engagement” (53.5%) and “E-services” (49.8%). At the same time, the component related to local open data policy (48.8%) is implemented three times better than the component related to dataset publication (15.9%). Alongside municipalities, responsibility for this result also lies with the Ministry of Digital Transformation and Diia State Enterprise, which is within the Ministry’s management scope, because development of methodological recommendations, timely dataset moderation, and regular harvesting from local portals fall within their competence.

The “single entry point” principle, which is an element of European governance, is actively applied by regional centers when it comes to open data policy. In a dedicated thematic section on the official website, most municipalities provide links to the Regulation/Procedure governing key aspects of local self-government work with open data, as well as to the List of Datasets subject to publication in open data format. In most cases, this List assigns responsibility for specific datasets to structural units and specifies the update frequency for each dataset. Many city councils also provide a link to their single electronic account on the Unified State Open Data Web Portal.

At the same time, there are problems with defining the formats in which local self-government data should be published. Few municipalities provide access to the document appointing the person(s) responsible for open data publication. But the biggest problem is the near-total absence of evidence that municipalities initiate or at least track the use of their data for building e-services capable of improving citizens’ lives. 

As for open data publication, since 2022 the datasets of Zaporizhzhia and Kharkiv have been removed by the state from public access, and among cities that do publish data there are still regional centers that have not switched to publication under the “single window” principle in the city council’s electronic account on data.gov.ua. 

Of the 270 “EU integration” datasets that analysts searched for in 9 electronic accounts, 148 (55%) had been published as of December 5, 2025. Of these, only 51 datasets had timely updated resources. This result is largely due to the fact that Dnipro, Kropyvnytskyi, Lutsk, and Lviv city councils—leaders by number of published datasets—use their own open data portals, while the Ministry of Digital Transformation is unable to ensure regular monthly transfer of data from these portals to data.gov.ua.

Publication of timely updated datasets very rarely follows the interoperability principle. Analysis of timely updated datasets showed that only 6 of 51 datasets complied with all Ministry recommendations on unified object identifiers, unified dataset structures, and unified data resource structures.

The findings indicate a gap between indicators formed on the basis of local self-government self-assessment or aggregated national indicators, and the actual situation with open data publication and quality at the local level. 

In particular, in the Digital Transformation Index of Ukraine’s territorial communities, Kharkiv reported publication of 100% of open datasets, Dnipro 99%, and Khmelnytskyi 90%; however, the verification conducted within this study did not confirm these figures. Similarly, Ukraine’s high position in the Open Data Maturity 2025 ranking (4th place, open data maturity level 97.1% compared with a European average of 81.1%) more likely reflects the existence of policies and initiatives at the national level than the real implementation of EU open data requirements in municipal practice.

Data from the Digital Transformation Index of Ukraine’s territorial communities on open data, as of January 11, 2026.

The program recommends that all cities (not only those included in the pilot sample) take into account the analytical findings, specifically: 

• In line with the updated Resolution No. 835, approve a new internal administrative document containing the List of Datasets subject to publication in open data format. Taking into account the Ministry of Digital Transformation’s recommendations, specify for each dataset in the List: 

1. responsibility of specific information managers for preparation;
2. responsibility of specific information managers for publication;
3. update frequency (do not use the option “immediately after changes are made”);
4. format (do not use labels that are not formats—excel, API, etc.).

• Create a dedicated open data section on the official city council website. In that section, provide links to: 

1.  the document approving the Regulation/Procedure on open data;
2. the document approving the current List of Datasets subject to publication in open data format; 
3. document(s) designating the person(s) responsible for open data publication; 
4.  the decision on accession to the International Open Data Charter (if applicable);
5. the city council’s single electronic account on the Unified State Open Data Web Portal.

•  Compile information on services built on local self-government open data (maps, dashboards, chatbots, mobile apps) and provide links to these services in the dedicated open data section.

To ensure proper dataset publication on data.gov.ua, convenient access, and the ability to compare information both within a single dataset and across analogous datasets of other information managers, we recommend:

• Publish open datasets exclusively through the city council’s electronic account on data.gov.ua.
• Apply the practice of creating one city council dataset for each dataset established by Cabinet of Ministers Resolution No. 835, in line with European user-centric approaches.
• Check whether dataset titles in the city council account match the titles set out in Cabinet of Ministers Resolution No. 835.
• Conduct a detailed review of the Ministry of Digital Transformation’s dataset publication recommendations.

• Check whether the update frequency for each dataset, as specified in its file on data.gov.ua, matches the frequency in the approved List of Datasets.

• Prepare the number of resources recommended for each dataset. Resource structures (table fields) must fully match the recommended structures (see the table templates on the recommendation pages). Column names must be in English and fully replicate those proposed by the Ministry. There is no need to duplicate column names in Ukrainian. Abbreviating column names is unacceptable. 
• Provide accurate and complete information in resources. If some data needed to fill cells are missing, the cells must be populated with null.
• Name resource files and resources themselves using the names recommended by the Ministry (e.g., regulatoryList_2026-06-01, titleList_2026-09-30).
• Publish resources only in the formats provided for in the approved List of Datasets (CSV/JSON/XML/GeoJSON). If resources are published in CSV format, use UTF-8 encoding with comma delimiters.
• Verify that each dataset page contains the number of resources recommended by the Ministry. 
• Update resources forming datasets by version updates (see Example 1 and Example 2 of correct updates for datasets containing two resources each). 

Since the Ministry of Digital Transformation is the lead body in the system of central executive authorities responsible for shaping and implementing state policy in open data, is the holder of the data.gov.ua portal, and supervises the portal administrator, the program recommends that the Ministry: 

• Restore public access to those datasets of information holders from Donetsk, Zaporizhzhia, Luhansk, Mykolaiv, Kharkiv, and Kherson regions whose publication does not pose a threat to national security, territorial integrity, or public order.
• Align, with the Ministry for Communities and Territories Development and the State Enterprise Administrator of the Urban Cadastre at the National Level, a unified position on access to municipal urban planning documentation during martial law, and publish it.
• Approve a three-year Open Data Development Strategy for Ukraine.
• Enshrine the “single window” principle in new versions of Cabinet of Ministers Resolution No. 835 or Resolution No. 867.
• Disseminate methodological guidance for information managers on transition to dataset publication under this principle. Pay particular attention to cases where, prior to applying the principle, multiple information managers within one council published identically named datasets either by exporting in open, machine-readable formats or by each providing their own API endpoint.
• Disseminate methodological guidance on resource updates via version updates. 
• Disseminate methodological guidance on conducting an information audit. 
• Clarify or remove inconsistencies in existing recommendations for open dataset publication, including: 

1. lists of legal and individual regulations (excluding internal ones), draft decisions subject to discussion, and the document designating the person(s) responsible for publishing open data; 
2. real-time data on the location of urban electric and passenger road transport;
3. register of burial record books. 

• Disseminate recommendations for the open dataset related to the unified public investment project portfolio of a territorial community, publication of which is required by Cabinet of Ministers Resolution No. 527.
• During moderation of municipal datasets, strictly enforce the “single window” principle so that structural units of local self-government bodies have no opportunity to publish data through their own separate accounts.

Establish regular monthly automated transfer of datasets from local open data portals to data.gov.ua. In case this technical issue cannot be resolved in the long term, officially notify the relevant local self-government bodies of the actions required from them under these circumstances.

The program will prepare tailored recommendations for each city council covered by the study, to serve as roadmaps for improving the practical operation of electronic services. The indicator-by-indicator results for each city are available at the link.

For cities not included in the pilot study, the program team has prepared a self-assessment form to evaluate compliance of the sector with European standards.

 

This research is made possible with the support of the MATRA Programme of the Embassy of the Kingdom of the Netherlands in Ukraine, and with the financial support of Sweden within the framework of the program on institutional development of Transparency International Ukraine. Content reflects the views of the author(s) and does not necessarily correspond with the position of the Embassy of the Kingdom of the Netherlands in Ukraine or the Government of Sweden.

Transparency International Ukraine is an accredited representative of Global Transparency International. Since 2012, TI Ukraine has been helping Ukraine grow stronger. The organization takes a comprehensive approach to the development and implementation of changes for reduction of corruption levels in certain areas. 

TI Ukraine launched the Transparent Cities program in 2017. Its goal is to foster constructive and meaningful dialogue between citizens, local authorities, and the government to promote high-quality municipal governance, urban development, and effective reconstruction. In 2017–2022, the program annually compiled the Transparency Ranking of the 100 largest cities in Ukraine. After the full-scale invasion, the program conducted two adapted assessments on the state of municipal transparency during wartime. In 2024, the program compiled the Transparency Ranking of 100 Cities, and in 2025, it launched an updated format for assessing city councils — the European City Index.

Summary

Accurately estimating the economic impact of these changes is virtually impossible at this stage. We are unlikely to see the real picture until at least the end of 2026, once the new rules have been in force for six months.

Presumably, the changes will further increase the share of procurements won by VAT-paying participants, who already account for wins in 68.16% of competitive lots by number and 92.14% by value. Cooperation conditions for small businesses are likely to worsen.

The largest impact will be on competitive procurements where VAT-paying participants compete directly with non-VAT-paying participants, approximately 21% of lots by number and less than 10% by value.

However, it is very difficult to forecast whether contracting authorities will actually pay more. The change in evaluation methodology alone is likely to have a limited economic effect. If VAT-exclusive bid evaluation had already been in place during the analyzed period (one year and five months), and participant strategies had remained unchanged, contracting authorities would have paid an additional UAH 250 million, which is less than 0.1% of total contract value.

By contrast, changes in expenditure planning and in setting estimated procurement value may have a more serious effect. Under a worst-case scenario—if all non-VAT-paying contracting authorities add 20% to estimated value as a safeguard—budget pressure could increase to more than UAH 60 billion. Even in that scenario, however, competition could still push actual spending downward. 

On top of this, we must account for likely changes in bidder behavior, which are even harder to predict.

Evaluating bids without VAT may also increase price transparency.

In practice, the most difficult adjustment, especially at the beginning, will be for contracting authorities, which will need to adapt mid-budget year. We expect the Ministry of Economy to issue clear guidance with practical instructions on how to calculate estimated value, launch procurement procedures, and conclude contracts under the new rules. 

That said, the most significant impact on the sector will likely come not from procurement-rule changes, but from tax reform—specifically, the draft law abolishing the separate upper VAT-liability threshold for individual entrepreneurs under the simplified taxation system. The current threshold is over UAH 9.3 million and may be reduced to UAH 1–4 million, in line with the rules applied to other entrepreneurs. 

How the new approach to setting estimated value will work

For now, the contracting authority determines how to announce a procurement—whether the estimated value is stated inclusive of VAT or exclusive of VAT. Bids are evaluated based on their total final price, which includes 20% VAT (or another applicable rate) if the bidder is a VAT payer.

If the changes take effect, the contracting authority will indicate the value inclusive of VAT in the procurement plan, but exclusive of VAT in the procurement notice itself. Bids will also be reviewed exclusive of VAT. If a VAT payer wins, the VAT amount will be added to the bid price at the contract-signing stage. In this way, the final contract value will increase, but it will not exceed the procurement value stated in the plan. 

Why exclude only VAT, given that it is not the only tax that differs across businesses? Most likely because VAT is an indirect tax—effectively paid (financed) by the buyer, while the business only administers and remits it to the budget. By contrast, most other taxes—including the single tax for individual entrepreneurs and corporate income tax—are direct taxes borne by the business itself.  

Do the proposals align with EU rules?

In the explanatory note to the draft government resolution, the government stated that the changes are consistent with the European approach to estimated-value setting and bid evaluation.

This claim cannot be fully endorsed. Directive 2014/24/EU, which regulates public procurement in the EU, states that contracting authorities must calculate estimated procurement value exclusive of VAT. At the same time, the directive does not regulate VAT treatment at the bid-evaluation stage. 

VAT payers among contracting authorities and participants

According to BI Prozorro and the YouControl analytical system, non-VAT payers predominate among contracting authorities, accounting for 78.47%. This is expected, since most of them are budget-funded institutions. The remaining 21.53% of contracting authorities are VAT payers—for example, state-owned enterprises. 

If we look at all market participants (in both direct and competitive procurement), the overall status split is similar:

• non-VAT payers — 74.76%
• VAT payers — 25.24%

Note: In further calculations, we excluded CPV 3300 Medical equipment, pharmaceuticals, and personal care products, because some medicines and medical devices are taxed at a 7% rate. Procurements in preferential categories were not excluded from calculations, since separating them requires manual processing.

In competitive procurement, 88.03% of procedures were announced inclusive of VAT. These procedures accounted for 79.56% of the total value.

In competitive lots, the share of VAT-paying participants rises to 50.56%. This can be explained by the fact that competitive procurements are higher in value, while companies and individual entrepreneurs become VAT payers once they reach a certain annual income threshold. VAT payers also win more often. Over the nearly one-and-a-half-year period we analyzed, they won 68.16% of competitive lots. By value, they accounted for 92.14% of contracts.

How competition may change

In previous years, VAT payers argued that they were at a disadvantage compared with non-VAT payers because they had to add VAT to their bid price. However, as noted above, in competitive procurement they already receive roughly two-thirds of contracts by number and 92% by value. Once the changes take effect, VAT payers will likely win more often, but a substantial increase in their contract share is unlikely. Much will also depend on their interest in smaller lots. 

The shift to VAT-exclusive bid evaluation will primarily affect tenders where non-VAT participants compete directly against VAT payers. During the study period, the distribution was as follows:

• • only non-VAT participants — 21.66%
  •  only VAT-paying participants — 57.01%
  • both participant types — 21.1%

It is also important to note that, when bids are evaluated exclusive of VAT, the balance may tilt in the opposite direction, with VAT payers gaining the advantage—because they may claim input VAT recovery on procurements they previously made with VAT included. As a result, even where the nominal price of goods is the same, VAT payers’ effective costs may be lower.  Under such conditions, there is a risk of small businesses being crowded out of public procurement.

Will non-VAT contracting authorities have to pay more?

This is a difficult question to answer. On the one hand, it is possible to estimate, at least approximately, the economic effect of harmonizing bid-evaluation methodology itself. On the other hand, potential changes in how contracting authorities plan expenditures and set estimated procurement value also need to be considered. Of course, bidder behavior will matter as well, and that behavior may also change. However, in the case of bidders, probable scenarios are significantly harder to forecast.

1. Contracting authorities evaluate bids exclusive of VAT: economic effect

Perhaps the main concern procurement practitioners raise about evaluating all bids exclusive of VAT is that, in the end, contracting authorities may overpay. 

Example. A contracting authority allocates UAH 120,000 in its annual plan for a procurement and announces the procedure exclusive of VAT—i.e., with an estimated value of UAH 100,000. Two bidders participate: a VAT payer with a bid of UAH 95,000 and a non-VAT payer with a bid of UAH 97,000. Under VAT-exclusive evaluation, the VAT payer wins. At the contract stage, 20% VAT is then added. As a result, the final contract value is UAH 114,000—which is UAH 17,000 more than it would have been if the authority had signed with the non-VAT payer.

For VAT-paying contracting authorities, this rule change is unlikely to have a material effect. It is already more advantageous for them to announce procurements exclusive of VAT, and they can do so under current rules. If a VAT payer wins, the additional VAT amount included in the contract can later be recovered through input tax credit. By contrast, the changes will affect non-VAT-paying contracting authorities (about four-fifths of all authorities): they cannot recover paid VAT, and the bid that appeared cheapest exclusive of VAT may become more expensive than others once 20% VAT is added.

We analyzed whether winner selection in past procurements would have changed if bids had been evaluated exclusive of VAT, and by how much contract totals would have increased. 

For procurements where VAT payers and non-VAT payers competed against each other, we converted bids to net amounts (exclusive of VAT), identified winners based on those net amounts, and added 20% tax if the winner was a VAT payer. We then calculated differences between the final bids of the simulated winners and the actual winners.

If, during the study period, non-VAT-paying contracting authorities had evaluated all bids exclusive of VAT, total contract value would have been UAH 250 million higher than it actually was.

At the national level, over nearly one and a half years, this is a relatively small amount—0.06% of the total value of contracts awarded by non-VAT-paying contracting authorities. This is because these two bidder types meet in tenders infrequently, in only about one-fifth of cases, and differences between their bids were usually minimal, sometimes even less than one hryvnia. Potentially, this gap in final bid value could widen if the new rules are introduced, because non-VAT participants would need to cut prices more aggressively to win tenders. 

2. To hedge risk, contracting authorities may plan estimated value and expenditures 20% higher

Because contracting authorities are concerned that they may have to pay more after the reform, they may plan for higher expected expenditures. Some may simply budget amounts 20% higher than before to remain on the safe side and ensure they can procure the required quantity.

The worst-case scenario is that all non-VAT-paying contracting authorities do this in all cases. In that event, demand for local and state budget expenditures could increase by UAH 61.78 billion per year. In practice, the scale will likely be smaller, and this pressure may be one-off: after the first year of VAT-exclusive bid evaluation, there will be actual data showing what costs what in Prozorro and how the reform has functioned. Even so, this underscores the need to support contracting authorities in setting estimated value—and not only to account for VAT correctly.

We will see higher-quality price statistics on Prozorro

Now, if a contracting authority announces a procurement inclusive of VAT, a non-VAT payer may submit a bid within the full estimated value. As a result, that bidder may receive funds that were originally set aside in case VAT had to be paid.

Example. A contracting authority announces a procurement with an estimated value of UAH 120,000 inclusive of VAT. Accordingly, UAH 20,000 of that estimated value is VAT, which would need to be included in the contract and paid if a VAT payer wins. Two bids are submitted: a VAT payer at UAH 119,000 and a non-VAT payer at UAH 118,000. The authority signs a contract without VAT at UAH 118,000. That is already UAH 18,000 higher than the estimated value of the procurement item exclusive of VAT.

In practice, because non-VAT-paying contracting authorities are generally indifferent as to which status of supplier they contract with, they usually have no reason to analyze item value exclusive of VAT. Their priority is to obtain the most economically advantageous bid. Therefore, they may mark the estimated value as “inclusive of VAT” simply to preserve the option of contracting with a VAT payer later.

We examined competitive VAT-inclusive procurements in which contracts were awarded to non-VAT payers. Specifically, we calculated estimated value exclusive of VAT and the gap between that figure and the signed contract amount.

In 70.75% of cases (93,500 lots out of 132,100), the contract value with a non-VAT payer exceeded estimated value exclusive of VAT. The total excess amounted to UAH 3.98 billion, or 7.95% of the initial total value of those agreements. 

Two interpretations are possible:

• either prices are higher than they could be, because non-VAT payers can submit bids up to the full estimated value in procurements where VAT is included;
• or estimated procurement value is understated, and the additional 20% set aside for VAT effectively brings it closer to market level.

We also identified a number of State Audit Service monitoring findings in which violations were recorded when a winning non-VAT bid exceeded the estimated procurement value net of VAT. For example, this occurred in a overhaul procurement for Kyiv City Clinical Hospital No. 1, where the contract was signed without VAT at UAH 59.1 million against an estimated value of UAH 61.1 million inclusive of VAT, and in a procurement for green-space maintenance in Vynohradiv, where the contract was UAH 1.67 million without VAT against an estimated value of UAH 1.69 million inclusive of VAT.

This practice is not widespread, and there are also monitoring cases in which auditors did not identify violations in similar circumstances.

In this context, VAT-exclusive bid evaluation may be beneficial: at a minimum, it can improve how market value is reflected in procurement pricing. At the same time, there is a possibility that the changes will also incentivize non-VAT participants to reduce prices.

Will these changes increase tax revenues?

A significant fiscal effect from procurement-rule changes alone should not be expected. Public procurement participants account for only 1.6% of all active business entities. Of course, part of the overall business-entity count consists of individual entrepreneurs who in practice operate under employment-like arrangements. But even if we compare procurement participants only with legal entities, their share is still below 5%. In addition, VAT payers already win 68% of contracts in competitive lots, which account for 92% of the total contract value in competitive procurement. 

In some cases, communities will indirectly finance the state budget, except where VAT is recoverable. This refers to situations in which a VAT payer’s final winning bid (after VAT is added) exceeds a non-VAT payer’s bid. As the calculations above show, however, these amounts are small.

What will change for VAT-paying contracting authorities?

For these organizers, it is more advantageous to announce procurement exclusive of VAT, because even if a VAT payer wins and VAT must be included in the contract value, the tax can later be recovered through input tax credit. In general, they can already announce procurement this way—but the data show this is done in only 13.38% of cases.

In 2024 and the first five months of 2025, VAT-paying contracting authorities signed UAH 25.7 billion in agreements with non-VAT participants. If they had evaluated bids exclusive of VAT, non-VAT participants would have won less often, and the total value of winning bids would have been UAH 600 million lower (accounting for input tax credit). Why this occurs requires additional research.

Contracting authorities can already announce procurement exclusive of VAT, but State Audit Service monitoring conclusions sometimes treat it as a violation when the final contract amount inclusive of VAT exceeds the estimated value stated exclusive of VAT.

For example, in January 2023, state enterprise Dobropilliavuhillia-Vydobutok ran a procurement for pipes. In the procurement plan, it stated an estimated value inclusive of VAT at UAH 7.8 million, but announced the tender itself exclusive of VAT at UAH 6.5 million. A VAT payer won with a bid of UAH 6.48 million exclusive of VAT; VAT was then added in the contract, bringing the final amount to UAH 7.78 million. Therefore, the contracting authority followed the steps the government now proposes to make mandatory. Nevertheless, the Chernivtsi Regional Office of the Western Office of the State Audit Service identified this as a violation, stating that the authority “did not reject the bidder’s tender proposal as one whose price exceeds the estimated procurement value determined by the contracting authority in the open-tender notice.” 

However, this is not a consistent agency practice, and opposite decisions also exist. For example, the Odesa branch of Ukrainian Sea Ports Authority, in its annual plan for territory-maintenance procurement, set the estimated value at UAH 7.8 million inclusive of VAT, while announcing the tender exclusive of VAT at UAH 6.5 million. A VAT payer won, and the contract value inclusive of VAT was UAH 6.54 million. During monitoring, the Eastern Office of the State Audit Service asked about the estimated-value overrun, but after the contracting authority’s explanations, found no violation on this point.

Conclusions

In summary, accurately estimating the economic impact of these changes is practically impossible at this point. We are unlikely to see the real picture until at least the end of 2026, once the new rules have been in force for six months.

What can be stated clearly is that the reform is unlikely to produce a major fiscal effect—it affects a very small share of businesses at the national level.

Presumably, the changes will further increase the share of procurements won by VAT-paying participants, who already account for wins in 68.16% of competitive lots by number and 92.14% by value. Cooperation conditions for small businesses are likely to worsen.

The largest impact will be on competitive procurements where VAT-paying participants compete directly with non-VAT-paying participants, approximately 21% of lots by number and less than 10% by value.

However, it is very difficult to forecast whether contracting authorities will actually pay more. The change in evaluation methodology alone is likely to have a limited economic effect. If VAT-exclusive bid evaluation had already been in place during the analyzed period (one year and five months), and participant strategies had remained unchanged, contracting authorities would have paid an additional UAH 250 million, which is less than 0.1% of total contract value.

By contrast, changes in expenditure planning and in setting estimated procurement value may have a more serious effect. Under a worst-case scenario—if all non-VAT-paying contracting authorities add 20% to estimated value as a safeguard—budget pressure could increase to more than UAH 60 billion. Even in that scenario, however, competition could still push actual spending downward. 

On top of this, we must account for likely changes in bidder behavior, which are even harder to predict.

Evaluating bids exclusive of VAT may also improve price transparency, because it allows comparison of the net cost of goods and services without the indirect-tax component. It may also help reduce prices offered by non-VAT participants.

In practice, the most difficult adjustment, especially at the beginning, will be for contracting authorities, which will need to adapt mid-budget year. We expect the Ministry of Economy to issue clear guidance with practical instructions on how to calculate estimated value, launch procurement procedures, and conclude contracts under the new rules. 

Ultimately, however, there is another major variable: the draft law abolishing the separate upper VAT-liability threshold for individual entrepreneurs under the simplified taxation system. The current threshold is over UAH 9.3 million and may be reduced to UAH 1–4 million, in line with the rules applied to other entrepreneurs. Tax changes may affect the sector more strongly than procurement-rule changes themselves. The number of non-VAT participants would naturally decline, as they would have to register as VAT payers. And VAT administration is fairly burdensome for entrepreneurs; unless it is simplified, this may affect both the number of businesses and the prices of their goods and services.

This material is funded by the European Union. Its content is the sole responsibility of Transparency International Ukraine and does not necessarily reflect the views of the European Union. 

Research team

Project Lead:
Ivan Lakhtionov, Deputy Executive Director of Transparency International Ukraine for Innovative Projects

Author:
Kateryna Rusina, Senior Project Manager at Transparency International Ukraine

Co-authors:
Ivan Lakhtionov, Olha Nos, Liliia Lakhtionova, Anna Yuldashova

2. Full verification as the NACP’s core procedure is inefficient and disproportionate to corruption risks

Disproportionate focus under limited resources. The NACP has limited capacity and can conduct just over 1,000 full verifications annually. Resources are used inefficiently, with most checks targeting lower-level officials presenting minimal risks. Even successful verifications have a negligible deterrent effect. According to Opendatabot, in 2024, there were 358 criminal cases opened for false declarations and 258 in 2025. Only 20 (6%) reached court in 2024 and 17 (7%) in 2025. For illicit enrichment, 102 proceedings were opened last year, but only seven suspicion notices were served. This year, 52 proceedings resulted in just two suspicion notices and one indictment. False declarations also led to dozens of minor administrative fines (17,000 UAH ≈ 350 EUR).

Focus on minor discrepancies. Verifications largely concentrate on small inconsistencies with registers or documents, which have limited impact on corruption levels and weak prospects in court. Only 1 of 200 full verifications in the first half of 2024 avoided TI Ukraine’s criticism. 

Variable quality of checks. The NACP has, in various cases: incorrectly valued assets; failed to use third parties, expert bodies, or national or foreign institutions; ignored tax evasion; poorly assessed cash and cryptocurrency holdings; and failed to disclose explanations from declarants or whistleblower reports for independent assessment. Even in the case of Tetiana Krupa, head of the Khmelnytskyi MSEC, the NACP had to repeat verification after a scandal, as the first one revealed no criminal wrongdoing. 

Lack of proper oversight of NACP leadership declarations. No safeguards exist during verification of declarations filed by the Agency’s own leadership, creating additional risks.

 

Recommendations:

To Parliament: 

• Set out in the Law on Corruption Prevention the content of full verification of declarations to avoid checking minor discrepancies.
• Extend the list of declarants to include leadership of separate units and branches of legal entities, and patronage staff.

To the NACP:

• Standardize and improve full verification practices to ensure effectiveness and impartiality.
• Focus on identifying unjustified assets and signs of illicit enrichment among high-level officials and in high-risk sectors, to achieve the best outcome given the resources spent.
• Introduce additional checks and balances for verifying declarations of NACP employees.

3. Lifestyle monitoring fails to meet European standards

Overlap with full verification and lack of risk-based approach. Lifestyle monitoring overlaps with full verification and mainly detects hidden assets without lawful income, mostly among lower-level officials. This reflects the absence of a risk-based approach. 

No time limits. Current rules allow lifestyle monitoring to last indefinitely, potentially leading to excessive and disproportionate interference in private and family life, contrary to the law and Article 8 of the European Convention on Human Rights. Ukraine risks losing cases before the ECtHR. Some monitoring procedures already exceed one year. The absence of time limits allows the NACP to manipulate the four-year limitation period for civil forfeiture. The Agency has not changed its substantive approaches to lifestyle monitoring, despite criticism from international auditors and the public. 

Lack of transparency in the distribution of lifestyle monitoring checks among NACP officials and the restricted nature of the results. Lifestyle monitoring cases are not auto-assigned, and the broad powers of authorized NACP staff (including issuing binding orders) raise risks of abuse. Recently, the NACP refused to disclose expert conclusions in the case of former SSU cybersecurity chief Illia Vitiuk, where no violations were found — prompting doubts about the Agency’s independence. Unlike full verification results, lifestyle monitoring outcomes are not made public.

 

Recommendations:

To Parliament: 

• Differentiate in the Law on Corruption Prevention the procedures of full verification and lifestyle monitoring; establish appropriate parameters for lifestyle monitoring as a preliminary analytical activity of the NACP concerning a specific possible fact of an offense.

To the NACP:

• Implement the recommendations of TI Ukraine’s 2021 study on lifestyle monitoring regarding narrowing the mandate of NACP authorized officials, limiting lifestyle monitoring to clear timeframes of 4 months, introducing automatic distribution of lifestyle monitoring cases, etc.
• Publish the results of all lifestyle monitoring cases.

4. NACP illegally conceals regulations, content, and results of certain procedures, undermining accountability

The LAC is a type of financial control that identifies inconsistencies between declared information and registers and evaluates them according to certain rules and coefficients. The outcome of the LAC is a calculated risk rating indicator of the declaration, which affects the order of its selection for verification. 

Non-transparent LAC rules. The last publicly available rules (2020) were heavily criticized by TI Ukraine and the Anti-Corruption Action Center. In 2024, the NACP further concealed the rules, which have not been public since 2021, not even to the Public Council at the NACP. International auditors found no legal grounds for secrecy. Current LAC rules, central to the NACP’s risk-based approach, lead to failure to detect signs of administrative or criminal violations in half of all declarations selected for full verification. 

Secret procedures for special bodies. Verification rules for declarations of SSU staff, NABU employees, intelligence officers, and others remain classified, even withheld from international auditors, raising risks of conflicts of interest for the NACP Head.

 

Recommendations:

 

To Parliament: 

• Require publication of LAC rules in the Law on Corruption Prevention.

To the NACP:

• Publish LAC rules and subject them to broad consultation.
• Provide the Public Council at the NACP with access to financial control procedures for special declarants.

 

5. Second external audit of the NACP is blocked

Flawed methodology and criteria. Given the NACP’s opacity and weak accountability, an external audit by international experts in 2024–2025 (the first two years of the current Head’s mandate) is a necessary step for strengthening corruption prevention and verification processes. 

But before that, the Government must update the methodology and criteria for the audit, adopted in 2020, which proved overloaded and inadequate. Three Ministers of Justice failed to amend them. Instead, the Ministry unlawfully allowed the NACP, conflicted as the entity subject to the audit, to heavily influence their content. Current criteria exclude evaluation of the NACP’s lobbying mandate, while the inefficiency threshold for dismissing the Head is unrealistically high. The methodology and criteria also ignore the individual inefficiency and lack of integrity of the Head of the NACP. 

Moreover, the NACP has not published the Implementation Plan for recommendations of the first audit, announced over two years ago.

 

Recommendations:

To Parliament: 

• Establish in the Law on Corruption Prevention that the Audit Commission shall be the body adopting the methodology and criteria for the external audit of the NACP, as is provided for in the audit of the NABU–SAPO.

To the Government: 

• Broadly consult on draft changes, simplifying audit methodology and criteria while minimizing the NACP’s role as the evaluated body.
• Form the Audit Commission based on proposals from international partners.

To the NACP:

• Disclose the status of implementation of recommendations from the first external audit and ensure their enforcement.

However, none of the city councils published, on their official websites or on the websites of structural units responsible for social protection, a separate list of social services available for online ordering by citizens. While digital transformation has already taken place in the field of administrative services (lists of administrative online services can be found on the websites of Administrative Service Centers in 8 out of 11 cities), the sphere of social services has so far been largely bypassed. Notably, at the national level, specialized electronic services have been developed — the Social Portal of the Ministry of Social Policy and the Electronic Account of a Person with a Disability, which allow users to submit applications for services online. However, links to these services are published only on the websites of Kropyvnytskyi, Kharkiv, and Khmelnytskyi, primarily within news items.

Another unexpected finding of the study was the absence, on the homepages of official local self-government body websites, of links to pages dedicated to comprehensive city mobile applications with lists of available services. In total, four such applications are currently in operation: e-Dnipro (Dnipro), Kyiv Digital (Kyiv), Misto (Odesa), and Open Kharkiv (Kharkiv). However, only the homepage of the official website of the Odesa City Council contains a link to the Misto application. Even in this case, the link leads to a website where it is impossible to find clear information on which services are already available to residents and visitors of Odesa. As for the homepages of the Dnipro City Council, Kyiv City Council, Kyiv City State Administration, and Kharkiv City Council, as of October 2025 there were no references to city mobile applications at all.

Ten regional capitals do not provide access to an e-service for searching for medical equipment in municipal healthcare institutions (the only exception is Kropyvnytskyi), nor do they offer any electronic service with up-to-date data on energy management or energy monitoring (the only exception is Kyiv). Nine out of 11 cities do not provide access to an electronic service for searching for medicines purchased with public funds in municipal healthcare institutions (exceptions are Kyiv and Odesa). It should be noted that city councils are not required to collect data on medical equipment, medicine stocks, or consumption of utility services specifically for the purpose of creating electronic services based on them. Local authorities are required to disclose and in most cases, regularly do disclose such data on the Unified State Web Portal of Open Data, in accordance with Resolution of the Cabinet of Ministers of Ukraine No. 835.

In addition to issues related to the availability of services, analysts also recorded difficulties in determining the timeliness of the data provided by certain services. For example, according to the study, 7 out of 11 city councils do not provide access to up-to-date interactive maps of civil protection shelters, which is critically important in the fourth year of the war. In Kropyvnytskyi and Poltava, it was not possible to find any links to such maps on official websites at all. In five other city councils, access to the maps was provided; however, it was impossible to verify whether the data had been updated in 2025.

As regards the application of the single point of entry principle, compliance with this indicator was recorded in only 5 out of 11 cities — Dnipro, Zaporizhzhia, Kropyvnytskyi, Lutsk, and Khmelnytskyi. None of the leaders — neither Kyiv, nor Lviv, nor Kharkiv — have created dedicated sections on their official websites where users could easily find up-to-date links to dozens of electronic interactive services developed by these cities on their own initiative over recent years.

Key findings and recommendations 

A clear correlation can be observed between the absence of an approved informatization program and low results in the comprehensive assessment of electronic services.

The 11 analyzed cities demonstrate very different levels of development of electronic services ecosystems. While the range of openness assessment results for this sample was 28 points, the gap between the highest and lowest scores in the second block amounted to 43 points. This disparity can be explained by the fact that, unlike indicators related to electronic services, many openness indicators are direct requirements of Ukrainian legislation, which city councils cannot ignore.

The single point of entry principle, which is an element of European governance, is applied selectively. Confusion is sometimes observed in the use of the terms services and public services. 

At the same time, clear progress can be seen in the development of digital services supporting the implementation of the Ukraine Facility Plan in the context of reforms such as Improving Preschool Education, Comprehensive Planning of Transport Sector Development, Strengthening Citizen Engagement Tools in Local Decision-Making, and Digitalisation of Public Services. Thanks to these services, residents of regional capitals have access to simple and user-friendly solutions in the areas of education, transport, administrative services, and urban amenities, as well as opportunities to participate in local e-governance.

At the same time, there are significant challenges related to digital services in the areas of local statistics, energy efficiency, healthcare, and social services. This does not contribute to the implementation of the Ukraine Facility Plan with regard to reforms outlined in the Decentralization and Regional Policy, Energy Sector, and Human Capital sections. Most importantly, local self-government bodies are not paying sufficient attention to the development of electronic services in areas whose importance continues to grow each year during wartime.

The program recommends that all cities (not only those included in the pilot sample) take into account the analytical findings, in particular by: 

• Systematizing strategic digital planning and developing and approving informatization programs for the next 3–5 years. These programs should define the purpose, priority areas, and specific objectives of digital development of the territorial community, as well as expected results. For communities, informatization programs should become not a formal document but a roadmap to a digital future. 

• Conducting an inventory of all existing digital services (information and analytical systems, mobile applications, chatbots, interactive maps, interactive dashboards, etc.), including relevant services offered by the state (e.g. Open Budget, the Social Portal of the Ministry of Social Policy), businesses (e.g. DozoR City, EasyWay), and civil society or charitable organizations (e.g. EDEM: Local Petitions, e-Liky). Links to these services should be consolidated in a single section of the official city council website — a single point of entry for electronic services. Duplication should not be avoided: for example, if municipal hospitals and clinics update data in e-Likyhttps://eliky.in.ua/, links to this service should be available both in the Electronic Services section and on the website or page of the healthcare department.

• Developing basic electronic interactive local statistics services, or, where such services already exist, regularly collecting, processing, and visualizing statistical data. Strategic planning, managerial decision-making, investment attraction, and assessment of local policy effectiveness must be data-driven. Providing access to such data does not necessarily require the immediate creation of a full-fledged local statistics portal, as done by Lviv; it may be sufficient to create a dedicated section with public thematic dashboards on the official city council website and gradually populate it, as Kyiv does. 

• Paying due attention to digital transformation in the field of social services. Over 3.5 years of the full-scale invasion, the number of recipients of social services — individuals and families belonging to vulnerable groups and/or facing difficult life circumstances — has increased significantly. Guided by European principles of human-centered and inclusive design, local authorities should extend the existing experience of digital administrative services to the social services sector as soon as possible by: 

1. ensuring access on the website of the social protection unit to online pre-booking for appointments with its departments (example: Lviv); 
2. publishing on the unit’s website a table based on the Classifier of Social Services, listing available services, service providers, and indications of whether services can be ordered online or offline;
3. creating a dedicated page explaining how to order social services online, with links to relevant national digital services (the Social Portal of the Ministry of Social Policy, the Electronic Account of a Person with a Disability, etc.); 
4. ensuring access to a web form for assessing the quality of social services provided (examples: Kyiv, Chernihiv).

• Clearly indicating the reference date of data used or displayed in electronic services (day/month/quarter/year). Data currency may be specified either through an explanatory note (e.g. “Dashboard update frequency: daily” for the Petitions dashboard on the Kyiv City State Administration website) or through a copyright notice (e.g. “© 2025 Municipal Enterprise Center for Information Technology Management” on the Zaporizhzhia shelter map). 

The program will prepare tailored recommendations for each city council covered by the study, to serve as roadmaps for improving the practical operation of electronic services. The indicator-by-indicator results for each city are available at the link.

Summary

Excessively long payment terms are generally not a very widespread practice on Prozorro. Payment terms longer than one month are used in about 15% of goods and services procurement, and only in works procurement does their share reach 31%.

Quick post-payment reduces financial risks for suppliers and makes participation in procurement more attractive. Conversely, excessively long post-payment periods may put off part of the business community. The data confirmed this: as payment terms increase, competition decreases. At the same time, abnormally long terms of more than one year that we identified are often linked to technical errors in filling in electronic fields on Prozorro, whereas the actual terms in the contracts themselves are significantly shorter.

As for the actual speed of payment in procurements with contractual payment terms of over three months, in those cases where it can be checked, approximately 91% of lots are paid earlier than the agreed deadline — on average 57% faster. However, these data cannot be treated as a representative sample for the system as a whole, since within the scope of this study we managed to obtain payment information for only 27.5% of all such transactions. The reason is that more than two third of such procurements fall on those financed from the income from commercial activities of municipal and state-owned enterprises. Data on these payments is not public.

Complaints about payment terms remain isolated, and the AMCU has so far not upheld them. 

What can be done to reduce the share of procurement transactions with long payment terms

• First of all, find out why the AMCU does not uphold complaints about long payment terms.
• Introducing e-contracting with automatic verification of payment terms in contracts would help prevent situations where long payment periods arise due to technical errors rather than reflecting the actual conditions.
• It is worth considering mandatory reporting on contract performance events within defined time limits — this would make both payment monitoring and contract implementation more effective and transparent.
• Consider making information on payments by municipal and state-owned enterprises public. At present, there is no information at all on how they make payments, even though these contracting authorities also operate public funds. 
• A more radical option could be considered — prohibiting excessively long payment terms in public procurement at the legislative level. However, this is a rather risky step that requires much deeper prior analysis and consultations. In some cases, longer terms may have objective reasons.

How contracting authorities specify payment terms when announcing a procurement

Payment terms under a contract (the procedure for settlements) are a key factor of a procurement’s attractiveness for businesses. 

On authorized Prozorro platforms, when creating a procurement, a contracting authority must select the following in the electronic fields:

• the event upon which payment under the contract will be made;
• the type of payment: advance or post-payment;
• the period (in days) within which payment will be made;
• the type of days: working, calendar, or banking;
• the payment amount — in such a way that the total amount across different types of payment equals 100%, which makes it possible to provide for combined payment formats.

Before the electronic fields on payment terms on Prozorro became mandatory, contracting authorities did not always enter the relevant information, so such tenders are marked as “Not specified.” At the same time, payment terms in any case had to be indicated in the draft contract uploaded with the notice and had to correspond to the parameters selected in the electronic fields, where applicable.

How often contracting authorities set unrealistic payment terms

Overall, from January 2023 through October 2025, contracting authorities successfully conducted 975,000 competitive procurement transactions on Prozorro with an estimated value of UAH 2 trillion. 

Of these, post-payment accounted for 82.9% of lots by number. At the same time, when looking at the distribution by contract value, post-payment accounted for 78.9% of the total amount, almost 3.3% for advance payments, and 15.6% for a combination of advance and post-payment.

Another 14.3% by number and 2.2% by value are lots where the form of settlement is not specified. These are predominantly requests for quotations on Prozorro Market.

Post-payment is the most common form of settlement regardless of the type of procurement item. However, while in goods and services advances and their combination with post-payment account for only a few percent of lots, in works their share is higher — 8.6% for the combined type of payment and another 0.2% for pure advance payment.

Thus, across all breakdowns, post-payment is the most widespread settlement form in contracts concluded as a result of procurements on Prozorro. Going forward, we focus specifically on this form. It should be recalled that, in the context of these data, we are talking about the settlement terms that contracting authorities manually entered into the system when announcing procurements.

Which type of days for payment terms is most often used by contracting authorities 

Contracting authorities use different approaches, but most often they choose calendar days — this option is specified in 45.4% of lots. Other types are used in roughly equal proportions: 28.7% banking days and 25.9% working days. At the same time, the difference between these types can be substantial, since the same number will correspond to different actual periods. For example, 10 working days do not equal 10 calendar days, but 14. That is 40% more. Therefore, to ensure data accuracy in this study, we recalculated payment terms expressed in banking and working days into calendar days. We also took into account the fact that, at the legislative level, from April 1, 2023, a banking day was effectively equated to a calendar day.

Which payment terms are most often set in contracts

After downloading the data from BI Prozorro, we grouped payment terms into the following ranges: up to 1 month, 1–2 months, 2–3 months, 3–6 months, and over 6 months.

The analysis showed that the vast majority of procurements — almost 85% by number and 78% by contract value — have payment terms of up to 1 month.

We also examined whether there are differences between the procurement of goods, services, and works. In the first two categories, the main share falls on procurement transactions with post-payment of up to 2 months — these periods cover more than 92% of all contracts.

In the category of works, a different trend is observed: the second most common option is post-payment in the range from 3 to 6 months. It accounts for almost 12% of the total number.

It is also worth noting that in the procurement of goods and services, payment terms increase in parallel with the growth of the average contract value, which may be a logical consequence of larger supply volumes or more complex performance conditions.

By contrast, no such correlation is observed for works. 

Thus, in general, the vast majority of procurement of goods provides for swift payment, which should promote competition — suppliers face lower financial risks. Longer payment terms (over two months) are not typical for such transactions. The same trend applies to service procurement: contracting authorities are generally ready for prompt post-payment, and longer payment periods are used only in isolated cases.

For works, longer payment terms are more typical, which may be linked to technological construction cycles and staged financing. At the same time, this creates risks for contractors, especially small businesses, due to the need for working capital.

Therefore, it can be concluded that excessively long payment terms are generally not a widespread practice on Prozorro. Payment terms longer than one month are used in about 15% of goods and services procurement, and only in works procurement does their share reach 31%.

Correlation between payment terms and competition 

The highest average level of competition is observed in the procurement of goods — 2.31 bidders per lot. For services, this figure stands at 1.52, and for works — at 1.35. 

At the same time, data on the average number of bidders in procurement correlate with payment terms. Although we cannot assess the strength of this factor’s impact as the level of competition depends on more than just payment conditions, faster post-payment is more attractive for businesses. In particular, quicker payment reduces financial risks for suppliers, whereas long post-payment periods may limit access for some of them due to the need for working capital.

In goods procurement, competition decreases steadily: from 2.41 bidders per lots with payment within one month, to 2.1 for a two-month term, and 1.92 for three months. Beyond that, this indicator remains at approximately 1.71. 

For services, the dynamics are similar — competition also declines gradually, though less sharply, as payment terms increase. Overall, it falls from 1.54 bidders per lot to 1.35. Only in transactions with payment terms exceeding six months does it suddenly rise to 1.49. Given the small number of such procurements, this average value is more likely driven by anomalous outliers rather than reflecting a general trend. A similar pattern is observed in works procurement: a gradual decrease from 1.42 to 1.13, followed by an increase to 1.22 in procurements with payment terms of over six months.

Thus, it can be stated that as payment terms exceed one month, competition gradually declines. At the same time, for post-payment exceeding three months, this decline slows down significantly. With this in mind, for the purposes of further analysis, we treat procurement transactions with post-payment of three months or more as having abnormally long post-payment periods and examine them in greater detail.

Analysis of procurement transactions with post-payment and payment terms exceeding 90 days

Distribution of payment terms

Most contracts fall within the ranges of 180–365 days (17.4 thousand lots) and 90–180 days (11.6 thousand lots). Together, these account for more than 96% of all cases, with a total contract value of UAH 2.1 billion. Lots with payment terms exceeding 365 days are relatively rare — only 1,137 lots (3.8%) — but they account for a rather large total contract value of UAH 1.5 billion.

Sectoral distribution

Long payment terms are most often applied in the construction sector. The largest shares of such contracts fall under the following CPV codes:

• 45450000-6 Final construction works — more than 10% of all lots;
• 45260000-7 Roofing and specialized works — 3.7%;
• 45230000-8 Construction of roads and utilities — 3.2%;
• 45310000-3 Electrical installation works — 2.3%;
• 44110000-4 Construction materials — 2.1%;
• 44160000-9 Mains, pipelines, pipes, casing pipes, tubing, and related products — 2.36%.

Beyond construction, the segment of petroleum product supplies (CPV 09130000-9) is also notable, accounting for more than 1.8% of procurements with long payment terms.

Procurement methods

The vast majority of procedures are special open bidding procedures (90%). Other methods are used far less frequently:

• Request for price quotations — 6.4%;
• Simplified procurement procedure — 3.7%;
• Standard open bidding — only 2 cases, which are likely due to errors in filling in the electronic fields.

Financing and actual payment terms

We also sought to analyze when contracting authorities actually make payments — that is, whether it indeed takes as long as originally provided for. However, it proved impossible to conduct a comprehensive analysis of such data.

The reason is that the overwhelming majority of lots with post-payment terms exceeding three months — namely, 57.2% of such procurements — are financed from the proceeds of commercial activities of municipal and state-owned enterprises. In value terms, this share is even higher — 68.6% of all funds fall within this category. These enterprises pay under contracts from their own bank accounts rather than via the State Treasury Service, and therefore information on the actual payment dates is not publicly available.

By contrast, procurement transactions financed exclusively from the state or local budgets account for approximately one third (33.9%) by number of lots and only one quarter (25.1%) by value. It is precisely for these procurements that real payments can be traced through the Treasury system, but they do not provide a full picture of the entire market.

In addition, payment reports on the Spending portal do not always indicate the identifier of the procurement to which a particular payment relates. The Fiscal Policy Research Center supported our analysis by extracting data on the date of the first payment for each procurement wherever possible. Ultimately, such data were obtained for only 27.5% of all transactions with long payment terms.

It should also be taken into account that, under Article 253(1) of the Civil Code of Ukraine, the calculation of a term begins on the day following the relevant calendar date or event specified by the contracting authority as the trigger for the start of the term. However, since information on such trigger dates is not publicly available, as contracting authorities are not required to report them, for the purposes of analyzing procurements with long post-payment terms we treat the contract publication date as the starting point.

The largest number of procurement transactions with payment terms exceeding 90 days were announced by Energoatom JSC and its 11 separate divisions. Together, they account for more than 20% of the total number of such procurements, with a total contract value of nearly UAH 20 billion. 

At the same time, a total of 18 regional electricity distribution companies of Ukraine also constitute a significant share of procurements — both by number of lots (8.58%) and by total value (UAH 36 billion). However, their presence in the market structure is fundamentally different from that of Energoatom. Some of these companies have common or cross-shareholding beneficiaries, their procurements are not concentrated in a single entity, and are instead distributed across various legal entities.

There are also contracting authorities from other sectors that frequently resort to setting long payment terms exceeding 90 days:

• Department of Humanitarian Policy of the Dnipro City Council — 4.4% of all lots, worth UAH 1.66 billion;
• Education Department of the Darnytskyi District State Administration of Kyiv — 2.2% of all lots, worth UAH 3.12 billion;
• Kharkiv Heating Networks Municipal Enterprise — 2.2% of all lots, worth UAH 3.1 billion.

The largest number of procurement transactions with payment terms exceeding 365 days, specifically, 62% — were announced by four contracting authorities, with a total value exceeding UAH 2 billion: 

• Vinnytsia City Council Municipal Enterprise “Vinnytsiamiskteploenerho” — 18.3% of lots;
• Municipal Enterprise “Management Company for Housing Maintenance of the Sviatoshynskyi District of Kyiv” — 17.6% of lots;
• Department of Infrastructure and Landscaping of the Chernivtsi City Council — 14.9% of lots;
• Chernivtsiteplokomunenerho Municipal Enterprise — 11.2% of lots.

The analysis shows that more than 91% of lots are paid faster than provided by the contract, even if the payment period is calculated from the date of contract publication — on average, by 57%. If, instead of the contract publication date, the countdown starts from the performance event, the actual payment period is even shorter. This points to a clear pattern: the payment terms published on Prozorro do not always reflect real financial practices. 

We interviewed several contracting authorities that set long payment terms to understand why they do so. Some explained that the reason lies in the funding cycle. For example, in district heating enterprises, expenses and revenues depend on seasonality — higher revenues in winter and increased maintenance costs in summer. As one such contracting authority told us, they fear penalties for late payment and therefore secure themselves by setting longer payment terms. In some cases, the decision to establish long terms is made not by the authorized procurement officer, but by the internal customer responsible for preparing the technical specifications. And sometimes long payment terms have simply become a routine practice that no one questions anymore.

Appeals to the AMCU

Since long payment terms may deter bidders from participating in procurement, we also analyzed AMCU practice to assess whether such terms are appealed — and if so, how successful such appeals are.

Procurements with long payment terms were challenged by bidders in only 2.7% of cases, which is the system-wide average among competitive tenders. However, not all such complaints concerned payment terms specifically, some related, for example, to lack of transparency or bias in determining the winner. At the same time, tenders conducted through the request for price quotations procedure are currently not subject to appeal at all. 

Procurements with complaints specifically about payment terms are difficult to identify, but of those we were able to find, the complaints were dismissed. For example, in two ongoing road repair service tenders in Mukachevo (UA-2025-03-10-008291-a and UA-2025-03-10-008888-a), the AMCU Commission rejected the complaints (decisions No. 6419 and No. 6430 of April 22, 2025, respectively), stating that the complainant failed to provide sufficient documentary evidence substantiating the need to amend the documentation, and that there was no proof that the established payment terms violated the bidder’s rights or legitimate interests. Similar refusals were found in at least 97 other AMCU decisions during 2024–2025. 

The very fact that such a large number of complaints are dismissed may discourage bidders from filing appeals, particularly when they encounter excessively long payment terms. They may decide that challenging such conditions is not worth the time, even when the contract terms appear unfair.

Anomalous cases and incorrect data

Some procurements feature extremely long payment terms — more than one year, sometimes even up to 1,000 days. The total contract value of such procurements during 2023–2025 reaches UAH 5 billion, accounting for 3.8% of all transactions. From the procurement announcement alone, it appears that suppliers would have to wait almost indefinitely for payment. Seeing such conditions, suppliers think twice before deciding whether to participate. However, once it comes to contract signing, the situation often changes: the actual payment terms specified in the contract itself frequently turn out to be much shorter.

In some cases, the reason lies not in an intention to delay payments, but in errors made when filling in electronic fields in the Prozorro system. 

For example, the Kyiv City Clinical Endocrinology Center (a municipal non-profit enterprise) procured 1,500 packs of paper towels for UAH 74,700. The electronic form specified post-payment within 1,000 days, although both the draft contract and the signed contract with the winner clearly stipulated 30 banking days from the date of delivery.

Another example concerns the procurement of lighting fixtures for Volynoblenergo PJSC worth UAH 717,000. The contracting authority entered a post-payment term of 888 days in the electronic field but did not specify any payment terms in the draft contract. After the contract was signed, it became clear that the actual term was 120 days, and the payment deferral had been included as a non-price evaluation criterion with options of 60, 90, and 120 days, meaning the exact term depended on the selected bid. In any case, entering 888 days in the electronic field was unjustified, even though formally the requirement of the Law (Article 37(1)(37)) to specify “payment terms” was met — payment was to be made after delivery of the goods. 

These examples show that extremely long payment terms sometimes do not reflect real practice and are merely the result of errors made by contracting authorities in Prozorro.

Although average competition in procurements with long payment terms is not always high, there are certain segments that remain fairly competitive, including cases where the number of bidders reaches 10–16 per lot. For example, in the procurement of medical supplies under CPV 33140000-3, some tenders achieved competition of more than 10 bidders with an average payment term of four months. Likewise, in electricity procurement under CPV 09310000-5, the average competition for the contracting authority Ukrprodcontract reaches 10 bidders with similar contract payment terms. Meanwhile, the state enterprise “Creative, Production and Research Center for the Development of Contemporary Art,” under the same CPV code, achieved competition among 16 bidders with a payment term of just over three months. 

Anomalous cases and incorrect data

Some procurements feature extremely long payment terms — more than one year, sometimes even up to 1,000 days. The total contract value of such procurements during 2023–2025 reaches UAH 5 billion, accounting for 3.8% of all transactions. From the procurement announcement alone, it appears that suppliers would have to wait almost indefinitely for payment. Seeing such conditions, suppliers think twice before deciding whether to participate. However, once it comes to contract signing, the situation often changes: the actual payment terms specified in the contract itself frequently turn out to be much shorter.

In some cases, the reason lies not in an intention to delay payments, but in errors made when filling in electronic fields in the Prozorro system. 

For example, the Kyiv City Clinical Endocrinology Center (a municipal non-profit enterprise) procured 1,500 packs of paper towels for UAH 74,700. The electronic form specified post-payment within 1,000 days, although both the draft contract and the signed contract with the winner clearly stipulated 30 banking days from the date of delivery.

Another example concerns the procurement of lighting fixtures for Volynoblenergo PJSC worth UAH 717,000. The contracting authority entered a post-payment term of 888 days in the electronic field but did not specify any payment terms in the draft contract. After the contract was signed, it became clear that the actual term was 120 days, and the payment deferral had been included as a non-price evaluation criterion with options of 60, 90, and 120 days, meaning the exact term depended on the selected bid. In any case, entering 888 days in the electronic field was unjustified, even though formally the requirement of the Law (Article 37(1)(37)) to specify “payment terms” was met — payment was to be made after delivery of the goods. 

These examples show that extremely long payment terms sometimes do not reflect real practice and are merely the result of errors made by contracting authorities in Prozorro.

Although average competition in procurements with long payment terms is not always high, there are certain segments that remain fairly competitive, including cases where the number of bidders reaches 10–16 per lot. For example, in the procurement of medical supplies under CPV 33140000-3, some tenders achieved competition of more than 10 bidders with an average payment term of four months. Likewise, in electricity procurement under CPV 09310000-5, the average competition for the contracting authority Ukrprodcontract reaches 10 bidders with similar contract payment terms. Meanwhile, the state enterprise “Creative, Production and Research Center for the Development of Contemporary Art,” under the same CPV code, achieved competition among 16 bidders with a payment term of just over three months. 

Summary

Over the two years of implementation, the experimental comprehensive recovery project has failed to reach the expected scale: only about 8% of the planned facilities have been restored. For the majority (64%), work either began or gained momentum only at the end of last year.

The state allocated ₴10.82 billion for the recovery of the selected settlements — 93% of the project’s expected cost. However, due to significant delays in disbursing the funds, the project is unlikely to be completed on time. As of May 2025, only ₴2.23 billion — roughly 20% — had been spent.

Due to gaps in the experiment’s conditions, the reconstruction of the village of Moschun was essentially excluded, and later, the rebuilding of Tsyrkuny was put on hold.

According to the conditions, 2025 was supposed to be the final year of the experiment’s implementation. However, according to the Recovery Services, in at least two settlements — Trostianets and Borodianka — work will not be completed before 2026. Will the government give the Recovery Agency and its regional services the opportunity to see the experiment through?

The more than twofold increase in funding and the list of facilities planned for restoration over the past year suggests such a scenario is possible. The final fate of the experiment, however, may be determined by the project review initiated by the government in April of this year. According to the head of the Recovery Agency, international experts will also be involved, and their findings are expected to be made public in the fall.

It is now crucial to take into account the mistakes made in order to avoid repeating them in the future — both during comprehensive recovery, which will be essential for large-scale postwar reconstruction, and during the implementation of other experimental projects designed to test new approaches to rebuilding.

New projects, old problems

At the start of the project, it included 295 restoration sites, distributed as follows:

• 268 – housing
• 10 – social infrastructure
• 8 – administrative buildings
• 7 – road infrastructure
• 2 – utilities.

Over the past year, the government revised this list twice. As of March 2025, the number of sites had grown to 739 — more than double the original number. This substantial increase was driven by the addition of housing infrastructure to be restored in two localities — Posad-Pokrovske and Borodianka.

The expansion of the list of restoration sites created the need for additional funding. As of 2025, the total expected cost of restoration under the pilot project had risen to UAH 11.6 billion — UAH 1.63 billion more than in 2023.

In the final year of the project’s implementation, the funding amount increased significantly. In 2025, the state budget allocated UAH 5.2 billion — more than double the amount allocated the previous year, and nearly as much as in the two previous years combined. On paper, the state has financed the comprehensive restoration of the selected localities in the amount of UAH 10.82 billion — 93% of the estimated value. However, this figure differs drastically from the actual expenditures on the implementation of the experiment.

Over the course of the project’s implementation, 83% of the funds used (UAH 1.85 billion) were directed toward the restoration of Borodianka and Posad-Pokrovske. Yahidne in the Chernihiv region fared the best — over a third (37%) of its restoration funding needs were met. The fewest funds were spent on restoring Tsyrkuny in the Kharkiv region — just over 1% of the expected need.

Overall, according to the contracting authorities (the Kyiv Regional Military Administration and regional Restoration and Infrastructure Development Services), as of May 2025, only UAH 2.23 billion had actually been spent on implementing the experiment. This represents approximately 20% of the total expected cost of the comprehensive restoration project. 

In 2023, of the more than UAH 3.35 billion allocated for the comprehensive restoration of selected localities, only UAH 559 million was used. The effective use of funds was hampered by a lengthy project selection process and delays in financing. As a result, regional restoration services were only able to start announcing tenders five months after the official launch of the pilot project — and just three months before the end of the fiscal year. Later, the lack of funding led contractors to suspend their work.

In 2024, the situation with comprehensive restoration deteriorated further. Due to political instability and continued dismissals of leadership at the Ministry of Infrastructure and the Recovery Agency, project funding was halted, and most of the work came to a standstill. 

The project resumed only in November 2024, when restoration services regained the ability to make payments for completed work. However, funding ceased again at the start of 2025 due to the end of the fiscal year. As a result, restoration services were able to use only UAH 503 million out of the allocated UAH 2.2 billion. Thus, in less than two years of the project’s implementation, the procuring entities were able to fund restoration work and services for just six months in total.

Only in 2025 were funds for the implementation of the pilot project allocated in a timely manner — budget appropriations were opened at the end of March. According to the Recovery Agency, as of May, procuring entities had managed to use UAH 1.19 billion out of nearly UAH 3.5 billion.

No money — no results

Delays in funding during the first two years of the project’s implementation directly impacted the pace and overall condition of the comprehensive restoration of the selected localities.

According to the Restoration Services, as of mid-June 2024, project documentation had been developed for 260 restoration sites, representing 88% of their total number (prior to its subsequent revision). Construction and repair work was ongoing at 211 sites (72%), and only one site was fully restored in the first year of the pilot project — a section of roadway passing through the village of Posad-Pokrovske.

However, during the second year of implementation, the project still did not reach the expected level of progress. In addition to the previously mentioned funding issues, another obstacle was the inclusion of nearly 450 new restoration sites. While the number of sites increased, the project implementation timeline was not extended. 

As of early June 2025, according to the Restoration Services, project documentation had been developed for 492 sites (67% of the total), and construction or repair work was underway at 472 sites (64%). Over the past year, works were completed at 57 sites in the village of Yahidne in the Chernihiv region, bringing the total number of fully restored sites to 58 across two years of the pilot project in five localities.

Procurement transactions

According to data from the BI Prozorro public analytics module, from June 2024 to June 2025, contracting authorities announced 373 procurement transactions related to the pilot restoration project. For comparison, 237 tenders were announced during the first year of the project.

During the analyzed period, the vast majority of the estimated value — 97% — came from procurement transactions conducted through special open bidding, amounting to nearly UAH 8.11 billion. These procedures primarily resulted in contracts for the construction of new sites or major repairs of existing infrastructure.

Meanwhile, 3% of the estimated value came from direct contracts, which mostly involved the procurement of project documentation development, as well as technical or author supervision. Although there were more of these transactions in number — 254 compared to 119 — they involved relatively small amounts. 

In the first year of the pilot project, the distribution by procurement method was approximately the same: UAH 3.1 billion (over 97.5%) went to special open bidding, and the remaining 2.5% to direct contracts. 

Among the leaders in terms of the number of procurement transactions is the Kyiv Regional Military Administration, which announced 154 tenders and awarded contracts worth over UAH 2.85 billion. In second place is the Restoration Service in the Kherson region — 93 procurement transactions totaling nearly UAH 2 billion. Rounding out the top three is the Restoration Service in the Kyiv region, which carried out 77 procurement transactions with a total contract value of UAH 1.1 billion.

The lion’s share of all awarded contracts — 98% of the total value — related to new construction and major repairs of damaged infrastructure. These contracts amounted to approximately UAH 5.9 billion. Another 2% (over UAH 86 million) were for the development of design documentation and related services. Contracts for technical and author supervision services totaled UAH 33 million, accounting for 1% of all contracts by value.

The largest share of all contracted funds — UAH 2.83 billion, or 47% of the total amount — was allocated to the restoration of apartment buildings. Nearly all of this sum was spent in a single locality — Borodianka in the Kyiv region. Another UAH 1.3 billion (21.5%) went toward the repair and reconstruction of private housing, primarily in Posad-Pokrovske.

Almost the same amount — UAH 1.2 billion — was spent on restoring utilities and road infrastructure, including national highways. The smallest portion — UAH 605 million, or just 10% of the total — was allocated to the restoration of social infrastructure facilities.

On average, 2.25 bidders participated in open tenders — nearly twice the average level of competition in construction tenders on Prozorro. However, this is lower than during the same period in the first year of the pilot project (3.6 bidders). The highest competition was recorded in procurements run by the Kyiv regional service — over 3 bidders per lot on average. 

Just like the level of competition, the “savings” rate — how much lower the awarded contract price was compared to the estimated value — also dropped significantly. On average, procuring entities saved 7.5% on competitive lots, compared to 21% in the first year. The Restoration Service in the Sumy region reported the highest “savings” rate, at 22.6% — three times higher than the project average. The Kyiv regional service also demonstrated strong results with over 17% in estimated “savings.” For other authorities, the savings rate remained within the 3–5% range.

An analysis of material prices in cost estimates showed that, overall, contractors used market-level pricing. However, some deviations were observed, mostly incidental rather than systematic. 

For example, in the construction of a multi-apartment residential building at 353 Tsentralna Street in Borodianka, an overpayment of more than UAH 11 million could result from inflated reinforcement steel prices. The cost estimate includes a price of UAH 35,000 per tonne, while the market price ranges from UAH 26,000 to 28,000 per tonne. It’s worth noting that this price is dynamic, meaning the contract value may change during project implementation. 

Another case involves the construction of private houses on Pryozerna and Sadova streets in Posad-Pokrovske, where an overpayment of over UAH 16.3 million may occur due to the inclusion of aerated concrete blocks in the estimate at a price 20% higher than that of market suppliers. Furthermore, the contract uses a fixed-price model, meaning the contracting authority cannot simply adjust the price of materials in the work acceptance certificates.

It is worth noting that during the analyzed period, the State Audit Service monitored 21 procurement procedures, representing 9% of the total number of announced lots. Violations were identified in 19 of them, including issues in tender documentation, disclosure of procurement information, the procedure for reviewing bids, and the terms of procurement contracts. 

Borodianka

In August 2024, the list of restoration sites in Borodianka, Kyiv region, was more than doubled — reaching 77. In particular, 32 apartment buildings, 2 protective structures, a new children’s art school, a dormitory, and the reconstruction of utility networks were added to the project. As a result, the estimated cost of restoring the locality increased by UAH 674 million.

In October, the Kyiv Regional State Administration became the procuring entity for the restoration of 58 sites in Borodianka. Together with the regional Restoration Service, they spent UAH 1.06 billion on implementing the pilot project in Borodianka — only 20% of the locality’s total comprehensive restoration needs. Most of these funds (nearly UAH 800 million) were used in 2025 alone. 

As a result, construction and repair works began over the past year at 66 restoration sites in Borodianka (85% of the total), including major repairs of damaged multi-story buildings and the construction of six new buildings to replace those that were destroyed. 

However, none of the sites within the pilot project have been fully completed so far. According to the Restoration Agency, major housing repairs are expected to be completed by the end of 2025, while the construction of new buildings is scheduled for 2026.

Trostianets

Since 2024, the list of comprehensive restoration sites in Trostianets, Sumy region, has remained unchanged. As before, it includes several major infrastructure projects, such as the reconstruction of the railway station, the station square, the 40th Army Square (which includes over 20 smaller facilities), and the repair of three sections of the Sumy–Poltava road, two of which are located between Trostianets and the neighboring village of Klymentove.

Back in 2023, UAH 663 million was allocated for the city’s comprehensive restoration. This amount was expected to cover the costs of all the listed projects. However, due to the end of the fiscal year, only UAH 80.6 million — or 12% of the allocated budget — was actually spent. 

In 2024, budget complications repeated themselves: out of the allocated UAH 278 million, the regional Restoration Service managed to utilize just under half (UAH 133 million) within two months. By 2025, the estimated value of Trostianets’s comprehensive restoration had increased to over UAH 1 billion, but the state allocated significantly less — only UAH 151 million. In total, from 2023 to 2025, the government allocated UAH 1.09 billion for the pilot restoration of the city, of which only UAH 224.6 million was actually spent.

As of early June 2025, repair work in Trostianets was ongoing at all sites. The Restoration Service in Sumy region expected the major repairs on the Sumy–Poltava road to be completed by the end of June, while reconstruction of the railway station and other projects would likely extend into the following year.

Yahidne

Yahidne in the Chernihiv region is arguably the only locality to show tangible progress within the pilot project.

The village’s comprehensive restoration plan includes nearly 120 sites, yet its estimated value by 2025 was among the lowest (UAH 402 million). This is due to the fact that 96% of the restoration sites are residential properties, and the cost of major repairs for housing is lower compared to other types of infrastructure. Also, the project envisioned the repair of street infrastructure, the village club-library, the construction of a new artesian well, and the creation of a memorial complex.

Over three years, the state allocated a total of UAH 528 million for all these efforts, but due to the same persistent issues with funding delays, only UAH 147 million (28%) had been spent as of May 2025.

According to the Restoration Service in the Chernihiv region, as of the end of April 2025, 56 houses and the village’s street-road network had been restored. In addition, repair works were ongoing at 23 other sites. Overall, major housing repairs in Yahidne are expected to be completed by the end of summer, with new construction projects scheduled for completion by the end of 2025.

However, the implementation of the project in Yahidne was hindered by the specific ownership status of the targeted properties. Under the terms of the pilot, owners of private properties were required to transfer the role of contracting authority to the Restoration Services. In practice, however, private owners ignored this obligation for an extended period or openly opposed it. As a result, the restoration of 21 residential buildings in Yahidne remains blocked due to untransferred procuring entity.

Tsyrkuny

Tsyrkuny is one of the most heavily affected villages in Kharkiv region and, at the same time, the most problematic locality within the pilot restoration project. The village is located 25 kilometers from the Russian border and remains under constant enemy shelling. Under such circumstances, concerns have been raised not only about safety but also about the economic feasibility of restoring infrastructure there.

At the start of the pilot project, only four sites were planned for restoration in Tsyrkuny:

• the village council building (intended to house an Administrative Services Center)
• the local clinic
• a kindergarten
• a lyceum.

The Restoration Service announced tenders for emergency recovery works on the lyceum and clinic. However, only a contract for the reconstruction of the lyceum was signed — and it was later terminated due to lack of funding. In August 2024, both of these sites were removed from the restoration list.

As for the remaining two sites — the village council building and the preschool — project documentation had been developed as of May 2025. Also, Tsyrkuny became the only locality for which a full restoration concept was developed at the request of the Restoration Service. A total of UAH 5.97 million was spent on this work. 

Whether these projects will be implemented remains uncertain. Back in 2023, discussions began about replacing Tsyrkuny with another locality in the region under the government project. However, the idea met resistance from local residents, for whom the pilot was the only hope for restoring the village. As a result, the proposal was soon dropped. Ultimately, in January 2025, the head of the Kharkiv Region Restoration Service announced that due to the security situation, the pilot project in Tsyrkuny had been put on hold.

Posad-Pokrovske

In 2025, Posad-Pokrovske in Kherson region retained its status as the locality with the highest number of approved restoration projects. Since 2024, their number has increased more than fourfold — primarily due to the addition of 388 housing infrastructure sites. As a result, the total cost of restoring the village rose by more than 40%.

Between 2023 and 2025, the government allocated over UAH 3.7 billion for the comprehensive restoration of Posad-Pokrovske — the highest amount among all pilot localities. At the same time, the Restoration Service in Kherson region managed to use UAH 789 million — just 21% of the allocated funding. 

The first 120 sites in Posad-Pokrovske were scheduled for restoration by April 2024. However, due to problems with transferring procuring entity from private owners and delays in funding disbursement, progress was slow and work on several sites was halted altogether. Nevertheless, the village became the first locality where restoration works were completed under the pilot — though not on a residential building, but on a road that runs through the village and connects Mykolaiv and Kherson. 

As of May 2025, the issues with private owners had been resolved, and a substantial budget — over UAH 2.2 billion — was allocated for the village’s restoration (only Borodianka received more funding that year). As a result, construction and repair works resumed and were underway at 364 sites, with one (the road) already fully restored. However, given the scale of comprehensive restoration in Posad-Pokrovske, completing all works by the end of the pilot project does not appear realistic.

Висновки

Over the two years of implementing the pilot project on comprehensive restoration, construction and repair works were fully completed at only 58 sites, approximately 8% of the total. For the vast majority of other sites (64%), work began or resumed only at the end of the previous year.

Between 2023 and 2025, the state financed the comprehensive restoration of selected localities in the amount of UAH 10.82 billion — 93% of the estimated value. However, problems with the actual disbursement of these funds to the procuring entities over two consecutive years effectively nullified the possibility of timely and efficient implementation. As of May 2025, only UAH 2.23 billion had been disbursed — approximately 20% of the total estimated value of the entire comprehensive restoration project.

According to the terms of the pilot, 2025 is supposed to be the final year of implementation. However, according to the estimates of the Restoration Services, in at least two localities — Trostianets and Borodianka — the completion of construction and repair works is not expected before 2026. An analysis of the progress in other localities suggests that this timeline may be realistic for them as well, given the slow pace of work and the fact that many sites have not even entered the active construction phase. 

Will the government allow the Restoration Agency and its regional services to complete the pilot? The doubling of funding and the number of restoration sites planned for the previous year suggests that this scenario remains possible. Still, the final decision may depend on the outcome of the project review launched by the Cabinet of Ministers in April this year. According to the head of the Restoration Agency, international experts will be involved in the review process and are expected to deliver their conclusions by autumn.

Research team

Project Lead:
Ivan Lakhtionov, Deputy Executive Director of Transparency International Ukraine for Innovative Projects

Author:
Andrii Shvadchak, Legal Counsel at Transparency International Ukraine