Monitoring is one of the key control measures in public procurement, but its effectiveness depends on more than detecting violations. It also depends on how quickly auditors respond. If monitoring starts too late, the contracting authority may already have begun paying, or the contractor may even have performed the contract in full. At that point, influencing the process is almost impossible. Hence the key question of prevention: does the control system act ahead of time, while negative consequences can still be prevented or at least mitigated?
This study proceeds from the hypothesis that, in terms of prevention, monitoring of competitive procurement is most effective if it starts after the winner is determined (the notice of intent to conclude a contract is published) and ends before the procurement contract is published, except for direct contract reporting. In this window, the contracting authority has already chosen the winner, but no contractual obligations have yet arisen between the parties. These are the best conditions for detecting and remedying violations in time. This prevents negative consequences and avoids situations where the problem can only be fixed by terminating the contract. As a result, procurement becomes more predictable, risks for businesses and contracting authorities fall, and trust in the public procurement system grows. Hereinafter, we refer to this window as the “optimal period.”
That said, monitoring launched after the contract is concluded still matters: it makes it possible to check whether contract amendments are lawful and whether the contract complies with the procurement terms. They are necessary too, but should not make up the majority. As for direct procurement (without using the electronic system), monitoring simply cannot begin before the contract is concluded, so starting it as soon as possible after the contract is published becomes particularly important.
Summary
In 2024–2025 and Q1 2026, over three-quarters of monitoring procedures in competitive procurement began after the contract had been published. Yet control gradually shifted to earlier stages. In 2025, the share of monitoring procedures launched between the determination of the winner and the publication of the contract more than doubled compared with 2024, from 10.8% to 23.8%. The time to the start of monitoring after contract publication also shrank: from 35 to 11 days for competitive procurement and from 41 to 20 days for direct procurement.
However, in 2025, only about 5% of monitoring procedures in competitive procurement took place entirely within the optimal period defined by the study, between the determination of the winner and the publication of the contract. Overall, auditors were starting control earlier, but completing it before the contract was concluded proved extremely difficult. The median duration of monitoring was 16–17 calendar days, while the period between the determination of the winner and the conclusion of the contract may be as short as 5 days.
To make control more preventive, conditions must be created for completing monitoring before the contract is concluded. One option is to consider how to streamline the monitoring process.
Another solution could be to suspend procurement for the duration of pre-contract monitoring. This would allow auditors to complete control and the contracting authority to remedy any violations identified before the contract is concluded. Still, it must be clearly defined when the procurement is unblocked, and safeguards against deliberate delays must be put in place.
General statistics
We examined public procurement monitoring activities launched in 2024, 2025, and Q1 2026. The analysis is based on data from the electronic procurement system as of April 2026. During the period under review, the State Audit Service and its interregional territorial bodies (hereinafter, state financial control bodies, SFCBs) launched 24,009 monitoring procedures. We excluded 68 repeat procedures for procurement that had already been monitored. We also excluded 2 monitoring procedures of defense procurement in which much of the information was restricted. The final sample thus comprised 23,939 monitoring procedures.
During the period under review, auditors monitored open tendering, simplified procurement, negotiated procedures, and direct contracts, as well as several requests for suppliers’ proposals. In total, 7.8 million procurements using these methods, excluding requests for proposals, were announced in the system in 2024–2025 and Q1 2026. Monitoring therefore covers approximately 0.31% of them by number.
This ratio should, however, be interpreted with caution. The total number of procurements includes a large share of below-threshold procurement, which is mostly not worth monitoring given its low value. Besides, monitoring in a given year may concern procurement announced earlier. The coverage rate should therefore be treated as an indicative, approximate figure rather than an exact ratio, both overall and in the breakdowns by year and procurement method below. Moreover, the State Audit Service generally does not monitor requests for suppliers’ proposals, as it considers that it lacks the authority to do so. This should change when the new Law on Public Procurement takes effect in spring 2027. Under it, marketplaces will replace the e-catalog, and auditors will be able to monitor such procurement.
A similar caveat applies to value. The total estimated value of procurement under the methods listed above was almost UAH 3.17 trillion. The value of procurement covered by monitoring was almost UAH 420 billion, or 12.85%. These figures exclude direct natural gas procurement by Gas Supply Company Naftogaz Trading LLC within Naftogaz Group.
By number of procurements, monitoring coverage remains low for the vast majority of methods. Open tendering accounts for the lion’s share of all monitoring procedures, yet only 3.21% of announced open tenders are covered. For simplified procurement, the figure is 0.95%. The negotiated procedure stands out at 5.2%, but until recently, wartime procurement rules did not allow contracting authorities to use it. As a result, there were only 173 such procedures during the period under review.
By value, coverage is substantially higher and better reflects the focus of control. Notably, auditors are increasingly targeting procurement with larger budgets year by year. In 2024, monitoring covered 14.88% of open tendering and just 2.69% of simplified procurement by value. By Q1 2026, these figures had risen to 21.87% and 14.93%, respectively. Since coverage of these procedures by number remains roughly unchanged, this growth in value terms points to deliberate prioritization: auditors are focusing more and more on higher-value procurement. Direct above-threshold procurement, meanwhile, remains a category with consistently low coverage, at about 5% by value.
Construction work (CPV division 45) was monitored most often: 8,259 procurements, or 34.5% of all monitoring procedures. This group also had the largest total estimated value: UAH 237.79 billion, or 56.74% of the total estimated value of monitored procurement. For construction, we analyzed data at the CPV division level. For other procurement items, we used the class level (four-digit codes), which allowed a more granular analysis.
Apart from construction work, fuel and timber account for a significant share of monitoring procedures by number. By value, the largest shares go to road maintenance and to individual uniforms, mostly for the military. Among the most common and most expensive types of monitored procurement, individual uniforms have one of the lowest shares of procurements with violations found (33.72%), while fuel has the highest (80.73%).
Analysis of monitoring in competitive procurement
To assess how preventive monitoring is, we need to know at what stage of procurement it begins and when it ends relative to key procurement events. We therefore defined the key events against which we assessed the start and end of monitoring: the determination of the winner (publication of the notice of intent to conclude a contract) and the publication of the contract.
When monitoring begins
In this study, we distinguished three main periods in which monitoring begins:
- before the winner is determined;
- from the day the winner is determined, i.e., the notice of intent to conclude a contract is published (inclusive), until the contract is published. For short, we call this “between the determination of the winner and the publication of the contract” or simply “after the determination of the winner”;
- after the contract is published (including that day). For short, we call this “after the publication of the contract.”
Most often, auditors started monitoring after the contract was published: 15,738 procedures, or 75.15% of the total.
The second most common period was “between the determination of the winner and the publication of the contract,” with 3,871 procedures (18.49%). These began in the optimal period defined in the study, when the procurement already has a result but contractual obligations have not yet arisen. Of these, 1.48% began more than 20 days after the winner was determined, i.e., in cases where the conclusion of the contract dragged on.
The third largest group consists of procedures launched before the winner was determined: 1,332 (6.36%). Among them, we also singled out 327 procedures (1.56%) that began at least 60 days after the procurement was announced. This separates procedures that genuinely started soon after the announcement from those where the procurement sat in the system for a long time without a winner.
Open tendering largely shapes the overall trend, so we looked at simplified procurement separately to check whether monitoring there starts at the same stages. There were 962 monitoring procedures in simplified procurement, or 4% of all monitored procurements. Of these, almost 78% began after the contract was published, 186 (19.3%) after the winner was determined, and only 29 (3%) before the winner was determined. Thus, in simplified procurement too, monitoring mostly starts after the contract is concluded.
We also tried to assess when monitoring began relative to the first payment under the contract. We thank the Fiscal Policy Research Center NGO for the payment data. However, this information is available for only 48.11% of procurements in which monitoring began after the contract was concluded. Based on the available data, we can only say this: of the procedures that started after the contract was published, at least 30.41% began before the first payment and at least 17.7% after it.
A year-by-year analysis shows that monitoring start dates are gradually moving to earlier stages of procurement. The share of procedures launched before the winner was determined grew one and a half times, to over 7%. The share launched after the determination of the winner but before contract publication more than doubled, to over 20%.
The trend toward earlier monitoring is observed in both open tendering and simplified procurement. It is more pronounced for tenders: the share of procedures launched before contract publication rose from 15.4% in 2024 to 31.8% in 2025 and 34.4% in Q1 2026. That is more than a twofold increase in two years. The shift is slightly smaller for simplified procurement: 16.7% in 2024, 26.4% in 2025, and 20.4% in Q1 2026.
Meanwhile, the share of monitoring procedures involving protracted procurements remains fairly stable. These are cases where, when monitoring began, one of two things was true. Either no winner had been determined although 60 or more days had passed since the procurement was announced, or the contract had not been published although 21 or more days had passed since the winner was determined. In open tendering, the combined share of such procedures even rose slightly, from 3.1% in 2024 to 3.9% in 2025 and 4.3% in Q1 2026. An early-stage start should therefore not be equated with a swift response, since the stages themselves sometimes last unusually long. Auditors may be choosing such procurements precisely because of the delays, for example because they expect deadline violations there. It would be worthwhile, though, to assess how effective this monitoring is: whether it detects significant violations and whether those violations are then remedied.
Preventiveness by region
The Western Office Department in Ivano-Frankivsk Region has the highest share of monitoring procedures launched before contract publication: 55.87%. In other words, this body starts most of its monitoring before the contract is published, a result unique among all the bodies studied. The central office of the State Audit Service (36.34%) and the departments in Zakarpattia (34.2%), Zaporizhzhia (30.5%), and Vinnytsia (29.1%) Regions also perform well above average. At the bottom are the departments in Kherson (9.67%), Ternopil (12.04%), Volyn (12.11%), and Luhansk (12.58%) Regions. There, over 87% of procedures start after the contract is published.
Monitoring is initiated fairly evenly across the central and interregional territorial bodies. The State Audit Service of Ukraine accounts for the most procedures, but the figures for regional bodies do not differ much. Regional departments also conduct similar volumes of monitoring.
Preventiveness by monitoring grounds
SFCBs launched most monitoring procedures, 16,441 (78.5%), on their own initiative. Another 3,509 (16.76%) followed triggered risk indicators. All other grounds (requests from public authorities and civic associations, media reports) together account for less than 5%.
Monitoring launched on auditors’ own initiative was more preventive. The share of such procedures that started in the optimal period rose markedly, from 11.49% in 2024 to 31.34% in 2025. In contrast, only 2.5–7% of procedures based on triggered risk indicators began between the determination of the winner and the publication of the contract. This difference may be partly explained by the fact that some risk indicators could only be triggered at later stages of procurement, after the contract was published or amended.
How late monitoring begins after contract publication
Across the sample, the median time from contract publication to the start of monitoring was 18 days, but the figures differ sharply from year to year. In 2024, the median was 35 days. In 2025, it shrank more than threefold, to 11 days, and in Q1 2026, to just 7 days. In most cases, this time ranged from 0 to 240 days, though there were extreme cases: the latest monitoring began 2,141 days (almost 6 years) after the contract was published. Note, however, that some long lead times may be due to the publication of contract amendments. These include supplementary agreements extending the contract term, which could, among other things, trigger automatic risk indicators.
Even where monitoring begins after the contract is published rather than before, the sooner it starts, the more scope remains to make a real difference. Among such procedures, start times shift earlier every year. The share launched within the first week after contract publication rose from 13.81% in 2024 to 41.79% in 2025 and 50.69% in Q1 2026. The share started in the second week also grew slightly, while later starts are now less than half as frequent.
When monitoring ends
Monitoring duration may depend on the procurement item (reviewing documents in works procurement usually takes longer), its complexity, and the number of lots. It may also depend on whether auditors need additional explanations and documents from contracting authorities. Yet our analysis shows that duration does not vary with the stage at which monitoring began: the median is 16–17 days for all categories.
Given that the period between the determination of the winner and the conclusion of the contract may be as short as 5 days, completing monitoring before the contract is concluded is almost impossible, even if it starts fairly early.
Unsurprisingly, 88.49% of completed monitoring procedures ended after the contract was published. At least 22.29% of procedures were completed before the first payment, and at least 20.34% after it. Recall that we have first-payment data for slightly less than half of the procurements with contracts in our sample, so this breakdown cannot be made for the remaining procedures.
Monitoring ends at early stages much less often: 3.55% of procedures end before the winner is determined, and 5.72% after the winner is determined, including on that day. Almost another 2% were completed after the procurement had been canceled, and 0.18% were not completed for various reasons.
The trend is positive, though slow. The share of procedures completed after contract publication is falling, from 91.1% in 2024 to 86.82% in 2025 and 85.2% in Q1 2026. Meanwhile, the share completed after the determination of the winner but before contract publication grew from 4.17% in 2024 to 8.09% in Q1 2026. This stage is among the most important for monitoring. Key decisions on the winner have already been made, but the contract has not yet been concluded, so it is still possible to stop a contract that would be concluded in breach of the rules.
In sum, most monitoring both begins and ends after the contract is published, when violations are rather difficult to remedy. Auditors are trying to increase the share of preventive monitoring, but this control measure takes longer than the period between the determination of the winner and the conclusion of the contract. Completing monitoring in the optimal period is therefore extremely difficult.
Above all, the share of procedures that began after the determination of the winner but ended after the contract was published rose substantially, from 6.35% in 2024 to 17.69% in 2025. The share of procedures conducted entirely within the optimal period also grew: 336 (3.74%) in 2024, 526 (5.37%) in 2025, and 156 (7.17%) in Q1 2026. Still, even with this growth, the share remains too small to speak of a systemic shift to preventive control. This category also includes procurements in which the contract was never signed.
Little monitoring overall is conducted in the optimal period, but practice varies significantly among SFCBs. The undisputed leader by share of such procedures is the Western Office Department in Ivano-Frankivsk Region, at 11.24% (110 procedures). That is more than double the sample average of 4.86%. Other standouts are the departments in Zaporizhzhia (8.95%), Zakarpattia (8.5%), and Vinnytsia (7.23%) Regions and the State Audit Service (6.27%). At the other end are the departments in Volyn (0.78%) and Luhansk (1.32%) Regions, the Eastern Office (1.97%), and the department in Kherson Region (2.12%). For these bodies, a full monitoring procedure in the optimal period is a rarity.
Frequency of violations detected in competitive procurement monitoring
Auditors detect violations in 62.31% of monitoring procedures, and this figure has been largely stable over the years.
Violations were found in 69.3% of procedures launched before the winner was determined, though this category is relatively small. For procedures launched after the determination of the winner and after the publication of the contract, the figure is lower but almost identical: 61.7% and 61.9%, respectively.
The downside of late monitoring is that some violations can only be remedied by terminating the contract. Termination means losses for businesses that have already started performing the contract. It also means delays for contracting authorities, which may now be unable to meet their needs on time. We cannot comprehensively verify how many contracts contracting authorities terminated specifically to comply with obligations imposed by auditors. However, even an aggregate analysis shows that they do face this problem. At the time of the study, 29.15% of the 11,700 contracts in competitive procurements where auditors found violations had been terminated. Where monitoring found no violations, only 4.82% of contracts had this status. By comparison, 3.12% of contracts have been terminated to date in completed competitive procurements that were announced in 2024–2025 or Q1 2026 and were never monitored. This comparison shows the scale of the difference between procurements with and without monitoring. The results should be interpreted with due regard to how the samples were formed, though. For monitoring procedures, we used the publication date of the decision to start monitoring; for procurements, the date the procurement was announced. The two datasets are therefore slightly offset in time. Comparing them gives an approximate but telling picture.
Sample of monitoring procedures conducted in the optimal period
Of the 1,018 monitoring procedures conducted in the optimal period, violations were found in 670 procurements (almost 66%). For most of these lots (almost 70%), the contracting authority canceled the procurement for the lot and no contract was concluded at all. In the group without violations, by contrast, cancellation was the exception (9.48%). Even more tellingly, 405 (87%) of the 465 canceled procurements were canceled before the monitoring conclusion was published. In these cases, the contracting authority canceled the procurement without waiting for the state auditors’ conclusion. This may suggest that the very launch of monitoring influences the contracting authority’s subsequent decisions on the procurement.
To find out which violations auditors identified and which remedies they required, we extracted and grouped this information from the monitoring conclusion files using Claude and ChatGPT. Most often, in about 70% of cases, auditors identified violations in the review of tender proposals and determination of the winner. Contracting authorities failed to reject bidders whose proposals did not meet technical or qualification requirements, did not confirm experience in performing similar contracts, or contained incomplete and contradictory documents. The opposite also occurred. Bids were rejected without sufficient grounds, bidders were not given the chance to correct discrepancies, or a bidder with a weaker or improperly prepared proposal was declared the winner. Such violations significantly affect the outcome of procurement, and it is important to detect them before the contract is concluded.
Another large group of violations concerned the preparation and publication of documents, found in about 40% of procedures. Tender documentation did not comply with the Law and Procurement Specifics No. 1178. It contained excessive or unlawful requirements, improperly defined qualification criteria, and errors in technical requirements and draft contracts. In two cases, contracting authorities set “virtually impossible contract deadlines” in the tender documentation. Auditors also recorded untimely or incomplete publication of information; no justification for technical and quality characteristics or for the estimated value; breaches of the procedure for amending tender documentation; failure to provide complete explanations and documents at the auditors’ request.
Some conclusions contained several obligations at once or alternative remedies. Examples are “publish the documents or, if this is impossible, reject the tender proposal” and “prevent similar violations in the future.” In such cases, each obligation was counted in every relevant category. The total number of entries in the table may therefore differ from the actual number of SFCB conclusions.
The largest group consists of cases where the conclusion set no specific obligation because the contracting authority had already canceled the procurement during monitoring (215 cases, or 32.09%).
Among auditors’ direct requirements, the most common is cancellation of the procurement: 197 cases (29.4%). It was sometimes listed as an alternative to other obligations.
As for actual compliance, information published in the electronic procurement system shows that contracting authorities reported fully remedying the violations in 151 of the 197 cases (76.65%).
However, in at least:
- 2 cases, the conclusion was challenged in court even though the contracting authority had canceled the procurement;
- 25 cases, the contracting authority disputes the conclusion and is challenging it in court;
- 1 case, the contracting authority had already concluded the procurement contract, so the e-procurement platform offered no technical means to cancel the procedure. Since the violation could not be remedied, explanatory work was conducted with the procurement officer;
- 5 cases, contracting authorities took other actions instead of cancellation. In one of them, to avoid derailing the project, the contracting authority concluded a supplementary agreement. Under it, the contractor undertook to cover, at its own expense, the risks of the construction work and of inflation, because the bidder had not set aside funds for them in the contract price;
- 2 cases, proceedings to impose administrative liability were opened;
- 24 cases, the outcome is unknown, as the system contains no information on either remedying the violation or opening proceedings.
The next most common requirements after cancellation are the two mildest forms of response: economic training and explanatory work (13.43%) and prevention of violations in the future (12.54%). Almost half of these were the only obligation in the conclusion, not an add-on to others.
Contracting authorities were required to cancel the decision on the winner in 37 cases (5.52%). Six of these conclusions were challenged; in 2 cases the outcome is unknown, and in the rest the violations were remedied. Auditors required rejection of a bidder’s tender proposal in 25 procedures (3.73%). Here, 2 conclusions were challenged and 1 outcome is unknown; in the rest, the violations were fully remedied. Auditors also required termination of the contract in 21 cases (3.13%). This resulted in 1 court claim, 1 administrative proceeding, and 3 unknown outcomes; in the remaining cases, contracting authorities remedied the violations. In some of these termination cases, monitoring had formally ended before the contract was published, judging by the dates. In fact, the contracting authorities had already concluded the contract and simply failed to publish the information on time, hence the requirement to terminate it. In the other cases, termination appeared as an alternative: auditors stated up front that if the contract had already been concluded, it should be terminated.
Excluding cases where the contracting authority canceled the procurement before monitoring ended, about 57% of all obligations in optimal-period conclusions are specific actions that actually change the course of the procedure. These are cancellation of the tender, annulment of a decision, and rejection of a bid.
Analysis of monitoring of procurement without using the electronic system
Procurement without using the electronic system differs fundamentally from other procurement methods, including in what it allows monitoring to do. Here, monitoring can begin only after the contract has been signed and a report on it published in the electronic system. We therefore analyzed such procurement separately and excluded it from the overall time calculations.
Procurement without using the electronic system accounts for 2,998 monitoring procedures, or 12.52% of the total sample. Of these, at least 1,486 (49.56%) concerned above-threshold procurement. Another 11 procedures were still ongoing at the time of the study (April 2026).
The median time from publication of the procurement contract report to the start of monitoring more than halved. It fell from 41 days in 2024 to 20 days in 2025 and 19 days in Q1 2026. The distribution of start times confirms this trend: auditors are now more than twice as likely to start monitoring within two weeks of the report’s publication.
Yet even with this acceleration, almost half of monitoring procedures for direct contracts still begin three weeks or more after the contract report is published. Such a late start significantly limits their practical value. If the contract has already been performed by the time of the check, remedying the violations found will be virtually impossible. Control then merely records the problem but can no longer solve it.
The duration of the procedures themselves remains steady: 15 days in 2024–2025 and 16 days in Q1 2026. Add the 19–20 days of waiting between contract publication and the start of monitoring, and the entire control process in most cases ends no earlier than 34–36 days after the contract is published.
Of the 2,987 completed procedures, 1,792 (60%) found violations. In 2024, auditors detected violations in almost half of procedures (49.1%). In 2025–2026, the share exceeded 70%: 71.9% and 72.7%, respectively.
As of the data download (April 2026), almost half of the contracts (815, or 45.5%) in procurements where monitoring found violations remained active. Another 600 contracts (33.5%) had been performed, and 377 (21%) terminated.
Conclusions
In 2024–2025 and Q1 2026, the vast majority of monitoring procedures in competitive procurement, over three-quarters, began after the contract had been published. In other words, control mostly records violations when the scope for real intervention is significantly limited or nonexistent. Over this time, though, the State Audit Service began stepping in early much more often. The share of procedures launched after the determination of the winner and before contract publication rose from 10.81% (2024) to 23.8% (2025). Still, even a start at this stage does not always allow monitoring to be completed before the contract is published. By law, monitoring may last up to 15 working days, and the study data broadly confirm a similar duration. The median was 16–17 calendar days, and some procedures took longer. Reasons include a complex procurement item, large volumes of information to process, and the time needed to obtain additional explanations and documents from contracting authorities. The mandatory period between the determination of the winner and the conclusion of the contract, by contrast, has been 5 days in recent years.
Auditors conducted only 1,018 monitoring procedures (4.87%) entirely within the optimal period defined in this study. Their share grew from 3.74% in 2024 to 5.37% the following year. These procedures found violations about as often as those conducted at other stages of procurement (60.08% versus 62% in 2025). In about 70% of cases, auditors identified violations in the review of tender proposals and determination of the winner, which significantly affect the procurement outcome. This shows that an earlier start does not necessarily mean a lower likelihood of detecting violations. And it is early monitoring that potentially offers more opportunities to stop contracts with such violations from being concluded.
Auditors also began launching monitoring sooner after contract conclusion. In 2024, they stepped in on average 35 days after the contract; in 2025, after just 11.
Procurement without using the electronic system accounts for a relatively small share of the sample: 2,998 monitoring procedures (12.52%). For this type of procurement, preventive control is objectively impossible, as monitoring can begin only after the contract is published. Still, the State Audit Service now also starts monitoring direct procurement earlier, though the change is smaller than for competitive procurement. The median time from publication of the procurement contract report to the start of monitoring fell from 41 days in 2024 to 20 days in 2025.
Recommendations
Given the above, it is important to find solutions that make monitoring within the optimal period possible.
The new Law on Public Procurement, which takes effect in spring 2027, already extends the minimum period between the determination of the winner and the conclusion of the contract to 10 days for most types of procurement. This will give auditors more time to complete monitoring before the contract is concluded. However, this alone may not suffice. Pushing the period out further would delay all procurement, although only a few percent of procedures are monitored.
Auditors could also look at ways to streamline the monitoring process so that control takes less time. Such changes are unlikely to fully solve the problem of completing monitoring before contract conclusion, but they will still improve the situation. A shorter monitoring period could also be considered for procurement through marketplaces (the future equivalent of Prozorro Market), as this stage will involve fewer documents to review. Auditors should then be able to extend monitoring up to the general time limit if they need to analyze additional materials during the review.
Finally, procurement could be suspended during pre-contract monitoring, as happens when a complaint is filed against a contracting authority’s decision. Auditors could then complete monitoring and, if they find a violation, issue an obligation to remedy it before the contract is signed and the business begins performing it. Implementing such changes, however, requires clear rules on how the procurement is unblocked. It also requires safeguards against deliberate delays, for example allowing a procurement to be suspended for monitoring only once.
Research team:
Team lead:
Ivan Lakhtionov, Deputy Executive Director of TI Ukraine for Innovative Projects
Authors:
Viktoriia Hermasheva, Projects Assistant, TI Ukraine
Kateryna Rusina, Project Manager at DOZORRO
