An administration can report that thousands of forms have been digitised while citizens still lose half a day on a routine procedure. For users, success is not about how many portals have been launched, but how quickly they can solve their problem, how much information they have to repeat and how often they are required to return. The cost of bureaucracy does not always appear in an institution’s budget.
In June 2026, the OECD published a country profile on Romania’s digital government. The 2025 Digital Government Index gave Romania a score of 0.24, compared with an OECD average of 0.70. By contrast, the indicator for open public data placed Romania above the OECD average. The gap illustrates that publishing data and delivering people-centred digital services are two distinct challenges.
For further context, see 99 kg does not mean 99 kg of food: how to read Romania’s food-waste figures, an article that explores an issue directly related to this subject.
The cost of bureaucracy: what makes a public interaction successful
Consider a citizen applying for a document. If they can find the application, authenticate themselves once, use information the administration already legally holds, receive confirmation and track the status without additional trips, digitalisation saves them time. If they download a PDF, print it, sign it, scan it and take it to a service counter, only part of the bureaucracy has moved onto a screen.
Institutions should therefore be assessed on whether the process is actually completed, rather than simply on the presence of an “online service” button.
A simple exercise: what does 15 minutes a week add up to?
Assume, strictly as an arithmetic example, that a person loses an additional 15 minutes a week to repeated form-filling, checks or re-entering the same information. Over 52 weeks, that amounts to 780 minutes, or 13 hours a year. If one were to assign an entirely hypothetical value of 30 lei per hour, the equivalent would be 390 lei a year.
| Illustrative scenario | Time saved or lost | Hypothetical value at 30 lei/hour |
|---|---|---|
| One person, 15 min/week | 13 hours/year | 390 lei/year |
| 100 people, at the same rate | 1,300 hours/year | 39,000 lei/year |
| 1,000 people, at the same rate | 13,000 hours/year | 390,000 lei/year |
These figures are not measured costs of Romanian bureaucracy and do not represent direct budget expenditure. They are a way of thinking about the issue: people’s time has value even when it does not appear as a payment made by the state.
Why simple multiplication can overstate the problem
Not everyone uses the same services, not every interaction happens weekly, and not every minute can be converted into income. A pensioner, an employee and an unemployed person face different circumstances, while time cannot be valued solely in wage terms. A serious study would therefore measure the real frequency of procedures, the time actually spent, and transport or documentation costs; it would not arbitrarily apply figures from an example to the entire population.
The OECD shows where the largest gaps lie
In its 2026 profile, Romania received very low scores on the “Proactiveness” dimension — 0.06, against an OECD average of 0.67 — and “User-Driven” — 0.10, against an average of 0.71. This is a comparative assessment of digital-government characteristics, not a survey of how many minutes each citizen loses.
The report also notes progress on open data: Romania’s OURdata score was 0.59, above the average of 0.53. A country can therefore make progress in publishing datasets while still having considerable ground to make up in proactive services and user experience.
Three indicators more useful than the number of platforms launched
| Indicator to track | Question it answers |
|---|---|
| Completion rate without an in-person intervention | How many users complete everything remotely? |
| Median time from application to outcome | How long does the typical experience take? |
| Number of data points repeatedly requested | How much information already held is requested again? |
| Abandonment and restart rate | How many people give up before completion? |
| Need for support contact | How clearly does the procedure work without help? |
A quality outcome must be verifiable. If an institution announces that cases are “resolved within 24 hours”, it should clarify when the clock starts, whether weekends are included, whether attending a service counter is required, and what percentage of applications meet that deadline.
Why the “once-only” principle is sound but difficult
In a citizen-centred model, the administration avoids repeatedly requesting information that it can legally and securely reuse. Implementation requires technical interoperability, legal clarity and strict access rules. A system that copies any information without controls is not progress; it is a data-protection risk.
The answer is not one huge database containing every detail about citizens. Necessary exchanges should instead be transparent, justified and audited. Effective digitalisation cannot be separated from security and trust.
A civic experiment for a local authority
- Choose three frequently needed services, such as paying a tax, obtaining a document and submitting a request.
- Test the procedures with the consent of those involved and without entering fictitious personal data into live systems.
- Measure the total time, the number of steps and requests for information that has already been provided.
- Record whether a printer, travel or telephone contact was needed.
- Repeat the test with people who have different levels of digital familiarity.
- Publish the methodology and its limitations before comparing institutions.
A well-documented test can produce useful information for the public, but a small sample is not a representative ranking of all local authorities. The results should be read as observations of a specific user journey.
Who is left behind when services become online-only?
An interface that is intuitive for an office worker may be difficult for a person with disabilities, someone without a capable smartphone, or someone living in an area with poor digital access. Reform must also be measured through accessibility, available assistance and reasonable alternatives. The aim is to reduce the burden, not shift it onto vulnerable people.
The best portal is one that means you no longer need to call
Time saved for citizens should become a public indicator, not an administrative marketing promise. Romania has tools and data, but the OECD shows that progress is uneven. A successful reform does not end when an institution announces a new website; it ends when the user’s problem is resolved simply, securely and verifiably.
Read more on InfoBuz
For baseline data already analysed by InfoBuz, see the catalogue of public services and how many can be completed entirely online.
Frequently asked questions
Is the 390 lei in the example money actually lost by every Romanian?
No. The calculation assigns a hypothetical hourly value to estimated time. It is not official statistics, does not describe every citizen and should not be turned into a budget cost without empirical research.
What does the OECD digital index measure?
The index assesses aspects of digital government, policies and institutional capacity. It does not directly measure the minutes each user spends at a service counter, so its score cannot replace monitoring of real-world experience.
Why does an online PDF form not amount to a fully digital service?
If users must print it, sign it on paper or take it to a service counter, the process has not been completed entirely remotely. The degree of digitalisation should be assessed through to the outcome, not only at the download stage.
What would be the first easily verifiable reform?
Publishing actual processing times and the rate of applications completed without travel would provide a starting point. The data should be broken down by procedure and published with a stable methodology, so institutions can be compared fairly.






Discuția cititorilor
Ai o opinie sau o întrebare despre acest articol? Spune-ne ce crezi.
Fii primul care deschide discuția.
Te rugăm să păstrezi un limbaj civilizat și să discuți la subiect. Comentariile spam sunt filtrate automat.