AVODA Group

Ecosystem KPIs a Ministry Could Actually Collect

How many Ugandan businesses started last year, and how many of them are still trading? The state answers the first half within weeks: 28,408 new companies and 21,695 business names in the year to June 2025, by URSB’s count. The second half waits years for an answer. A ministry can close that gap without a survey by building its ecosystem KPIs from records businesses already create in dealing with the state: a registration at URSB, a return or an e-invoice at URA, a monthly remittance to NSSF. Five indicators built that way can be published every quarter, each with its denominator stated.

Key Takeaways

  • Uganda counts business births quickly and business deaths late. Uganda registered 28,408 new companies in 2024/25, while one August 2026 notice removed 90,000 companies that had gone five years without filing an annual return.
  • Survival rates assume a register that records closures promptly, and patent counts assume inventors who file. WIPO’s 2024 profile lists 23 resident patent applications for Uganda.
  • Activity is the better signal. The Eurostat-OECD standard treats a business as active when it shows turnover or paid employment, and URA and NSSF records hold both.
  • Since 2021, URA’s registration system has pulled business details from URSB using the Business Registration Number, which gives a ministry the matching key UBOS said it lacked in 2019.
  • The Five Byproducts are registration births, a twelve-month activity rate, employer births, invoiced sales growth and secured lending to young firms, each drawn from a record the state already keeps.
  • Morten Jerven warns that data demands stretch African statistical offices, so each indicator should cost matching and analysis only, with zero new field collection.

Twenty-eight thousand companies, and deaths recorded years later

URSB’s annual report for 2024/25 counted 28,408 new companies, against 28,414 the year before, and 21,695 business names, down 20.5 percent from 27,297. The bureau’s review of 30 July 2025, a month after the year closed, printed the company count as 8,408, a dropped digit the annual report’s table corrects; the Ministry of Finance’s figure, as The Independent reported it, matches the table. Those numbers arrive soon after the year closes because the act of registering produces them.

Closures arrive by a slower road. On 3 August 2026 the Registrar of Companies announced the deregistration of 90,000 companies. They had been struck off in 2023 after five consecutive years without an annual return, and then let a twelve-month restoration window pass. URSB’s spokesperson, Dennis Nabende, explained what the returns are for: “They tell us and the public whether a company is still active and whether its details are current.” The bureau added that many of the firms were struck off through neglected filing rather than formal closure, including people who registered a company and then left it dormant.

The register’s death count is therefore a batch of companies that went quiet five or more years earlier. The 90,000 equals more than three years of new incorporations at the 2024/25 pace, yet it says almost nothing about which year’s founders stopped trading, or when.

URA’s register looks similar. It had more than 5.2 million registered taxpayers at 30 June 2025; the Auditor General’s latest report, as cited by The Independent, put active taxpayers at about 2.7 million and fully compliant ones at about 1 million. Most are individuals: at 31 December 2023, 204,573 of URA’s 4,064,432 taxpayers were companies and other non-individuals.

URSB’s Strategic Plan IV aims to register 2 million businesses by 2029/30, and the KiriEasy mass registration drive plans more than 235,000 in its third phase, after 3,195 and 6,856 in its first two. Formalization is good policy: a founder with a registered company can sign contracts and pledge assets in her firm’s name. A KPI table that reports only registrations, though, will rise with each drive, dormant entries included.

Survival rates and patent filings in a 2025 Springer volume

Entrepreneurial Ecosystems: Resource Acquisition, Flow and Management (Springer, 2025), edited by Parisa Baig and four co-editors, is about how resources reach founders. Its ninth chapter, by Roberta Dutra de Andrade and three colleagues, sets out the usual scorecard: “Indicators like startup survival rates, employment growth, generated revenue, and patent filings are used to evaluate the impact of entrepreneurial efforts.” Ecosystem vitality, they add, is read from startup density, access to funding and the supply of skilled people. The editors’ conclusion names new firm entries, innovation outputs and technological capability as the indicators that let policy adapt.

Dax Jacobson’s chapter, set in Salt Lake City, adds a warning a ministry should keep beside that list: “Magical creatures have become indicators of success,” he writes of unicorns and gazelles, while the everyday entrepreneurs who make up most of an economy draw little attention.

The scorecard was written for economies where a statistical business register already records births, deaths and employment for nearly every firm. Employment growth, revenue and access to funding travel to Kampala intact, provided each is measured through a record the firm already files. Patent counts and survival rates travel badly, and coverage needs saying out loud.

Twenty-three patents and an eleven percent formal sector

Start with patents. WIPO’s 2024 statistical profile for Uganda lists 23 resident patent applications, about 0.5 per million inhabitants. A count that small moves with a single university lab, and an ecosystem KPI built on it would mostly report noise.

Survival rates break on the denominator. The Eurostat-OECD Manual on Business Demography Statistics treats an enterprise as active in a year when it shows turnover, paid employment or investment, and it records a death only after a unit has been inactive for at least two years. That definition depends on tax and social security records flowing into a statistical register. In 2019 UBOS told a United Nations expert committee that Uganda’s business registers were fragmented, that linking them awaited a common identifier, that quarterly updates were beyond its budget, and that the death rate among informal businesses was high.

Coverage is harder to fix. Rwanda runs an establishment census every three years, most recently in 2023, when it counted 269,326 establishments. Only 11.9 percent of enterprises met the formal definition of keeping accounts and being registered with the Rwanda Revenue Authority, yet those formal firms employed 60.3 percent of the workers counted. Uganda’s proportions will differ. The pattern still applies: administrative records see a minority of firms, and those firms hold a majority of the jobs an establishment count can find. A ministry’s KPIs built on them describe the formal edge of the ecosystem with precision, and every release should say so.

The Five Byproducts

The Five Byproducts are indicators a ministry can publish because businesses generate them in the ordinary course of dealing with the state. Each has a holder, a matching key and a release rhythm, and the key is already in place: since June 2021, URA’s TIN application has filled in business details from URSB once an applicant enters a Business Registration Number, and a 2024 ICTD paper describes URA’s registration system as integrated with both URSB and the National Identification and Registration Authority.

IndicatorHolderDefinitionRelease
1. Registration birthsURSBNew companies and business names by quarter, district and sectorQuarterly
2. Twelve-month activity rateURSB matched to URA and NSSFShare of a quarter’s registrations showing, within 12 and 24 months, a tax return with turnover, an e-invoice or an NSSF remittanceQuarterly, one year behind
3. Employer births and jobsNSSFFirms making a first remittance, and median contributors per young employer at 12 and 36 monthsQuarterly
4. Invoiced sales growthURA (EFRIS)Median change in monthly invoiced sales for firms under three years oldAnnual
5. Secured lending to young firmsURSB security interest registrySecurity interests registered against companies under three years oldQuarterly

Registration births are already counted; their value comes from always appearing beside the activity rate, so a registration drive shows up first as births and a year later as a test of how many began trading.

The activity rate is the survival rate rebuilt for Uganda. Instead of waiting for URSB to strike a company off, it asks whether the company left a trace of trading anywhere the state can see, which is the Eurostat-OECD activity test run on local records.

Employer births come from NSSF, which already computes a compliance rate for its employer register. That rate fell from 57 percent in 2023/24 to 52 percent in 2024/25, which the Fund attributes to a 2022 change in the law bringing every employer into scope regardless of size. A first remittance marks a new employer more reliably than a registration does, and the count of contributors a year later is the employment growth the Springer chapter asks for.

Invoiced sales growth uses EFRIS, URA’s electronic invoicing system, which had 64,428 taxpayers on its register at December 2023. That base tilts toward larger, VAT-registered firms, so each median should carry its firm count.

Secured lending counts collateral, a step short of loans. URSB reported a 66 percent rise in security interest notices in 2024/25. Splitting those notices by the age of the borrowing company would show a ministry whether lenders are reaching young firms.

Every release publishes cohort totals only, with small cells suppressed, and carries one coverage line: the number of firms in the matched population and a plain statement that informal firms sit outside it.

Morten Jerven on the cost of every new indicator

Jerven, an economic historian whose book Poor Numbers examined how African GDP figures are made, would read a list like this with suspicion. In a 2013 essay for African Arguments he wrote that “The much noted demands for evidence-based policy has put the cart in front of the horse,” and that requests tied to donor projects stretch statistical offices: “donors distort data production instead of building up statistical capacity.” A ministry adding KPIs is one more demand on the same offices, and UBOS has said in writing that funds limited its own register updates.

Jerven’s objection sets the terms for the Five Byproducts. Each indicator has to be computable from records already held, with matching and analysis as the only new work. Jerven’s own remedy, in the same essay, is that “The statistical office should work as a data broker.” The Five Byproducts ask UBOS to play exactly that role and ask the ministry to fund the brokering as a line in its own budget. First-year matches will be imperfect, and the published rate should report its failed matches.

A large city with few people in it

Administrative data sees the formal edge of an economy sharply and its informal majority faintly. Nehemiah met the same problem in a city. He rebuilt Jerusalem’s wall and then found that “the city was large and spacious, but there were few people in it.” His first act was a register. He gathered the people “for registration by families,” and before counting anyone he went to the record that already existed: “I found the genealogical record of those who had been the first to return.” Families from Tel Melah and Tel Harsha searched for their records and came up empty. Nehemiah counted in order to fill a city, and he treated a missing record as a question still open before God. The governor’s ruling was provisional, holding them back from the most sacred food “until there should be a priest ministering with the Urim and Thummim.” It settled what the record could settle and handed the rest to a later day and a higher court. The register still carries their names.

The first quarterly release

Registrations from the first quarter of 2024/25 have already had their twelve months, so a ministry can compute that cohort’s activity rate now, with UBOS matching URSB, URA and NSSF records under a data-sharing agreement among the four bodies. Each quarter’s cohort needs four numbers: births, the matched population, the share active at twelve months, and the share of matches that failed. Employer births can follow in the second release; invoiced sales and secured lending can wait until the age-of-firm field is reliable.

Frequently Asked Questions

What ecosystem KPIs can a Ugandan ministry realistically collect?

Five indicators drawn from existing records: registration births from URSB, a twelve-month activity rate matched across URSB, URA and NSSF, employer births and job counts from NSSF, invoiced sales growth from URA’s EFRIS system, and security interests registered against young companies. Each needs matching and analysis, with no new survey.

Why is the business survival rate hard to measure in Uganda?

The company register records closures late and in batches. In August 2026 URSB deregistered 90,000 companies that had gone five years without filing annual returns, which says little about when each one stopped trading. An activity test using tax returns, e-invoices and NSSF remittances gives a timelier signal.

Should patent filings be an ecosystem KPI in East Africa?

For Uganda the count is too small to carry meaning. WIPO’s 2024 country profile lists 23 resident patent applications, about 0.5 per million people, so one laboratory can swing the figure. Employment, invoiced sales and secured lending describe the ecosystem far better, and the state already holds the records for all three.

Do administrative records miss informal businesses?

Yes, and the release should say so. Rwanda’s 2023 establishment census found 11.9 percent of enterprises were formal, yet they employed 60.3 percent of counted workers. Administrative KPIs describe the formal edge of an ecosystem precisely. A periodic establishment census, like Rwanda’s three-year cycle, is the instrument for the rest.

Who should compute the indicators?

The national statistics office, acting as a broker between registers. UBOS already receives URA’s tax register, and URA’s system has matched records to URSB’s Business Registration Number since 2021. The ministry’s job is to fund the matching, set the definitions and publish cohort totals with small cells suppressed.

Next Step

If your ministry or agency is drafting the indicator table for an entrepreneurship or innovation program, set each proposed KPI beside the records URSB, URA and NSSF already hold on a strategy call with Joshua while the table is still in draft. What survives asks only for numbers that arrive on their own, and your statistics colleagues keep their field budget for the establishment census the informal majority needs.

Related Reading

Sources and Evidence

  1. Baig, Montoya-Martinez, Belitski, Theodoraki and Godley (eds.), Entrepreneurial Ecosystems: Resource Acquisition, Flow and Management, Springer, 2025: the book tested here; the editors’ conclusion on entries, innovation outputs and technological capability.
  2. Dutra de Andrade, Rocha, Carvalho and Pinheiro, “From Resource Flow to Impact in Entrepreneurial Ecosystems: Paths to Sustainability, Circularity and Digitalization”, chapter 9: the standard indicator list, including survival rates and patent filings.
  3. Jacobson, “Transitional Everyday Entrepreneurs and their Enabling and Constraining Entrepreneurial Ecosystems”, chapter 12: “magical creatures” as indicators of success; everyday entrepreneurs.
  4. The Independent (Kampala) via AllAfrica, “Uganda: Tackling Tax Burden to Match Expanding Taxpayer Register”, 5 April 2026: 2024/25 company and business name registrations; URA registered, active and compliant taxpayers; KiriEasy figures.
  5. Daily Monitor, “URSB deregisters 90,000 inactive companies in massive crackdown”, August 2026: the 3 August 2026 notice, the 2023 strike-offs, the restoration window and Dennis Nabende’s statement.
  6. Daily Monitor, “URSB collects Sh312b from business registration”, 2025: Strategic Plan IV target of 2 million registered businesses.
  7. URSB, “URSB sets the pace for the new financial year with bold targets and reforms”, 30 July 2025: 66 percent rise in security interest notices; URSB’s role in security interests over movable property.
  8. Uganda Revenue Authority, Ministerial Policy Statement FY2024/25: taxpayer register at 31 December 2023, non-individual share, on-time filing ratios, EFRIS register.
  9. Uganda Revenue Authority, “URA further simplifies TIN registration”, press release, 11 June 2021: URA and URSB systems synchronized on the Business Registration Number.
  10. Scarpini, Santoro, Waiswa, Arewa and Nabuyondo, “Enhancing Taxpayer Registration with Inter-Institutional Data Sharing: Evidence from Uganda”, ICTD African Tax Administration Paper 35, 2024: URA registration integrated with NIRA and URSB.
  11. NSSF Uganda, “NSSF earnings grow by 11% to a record UGX 3.5 trillion”, press release, 10 September 2025: compliance rate of 57 and 52 percent; the 2022 change bringing all employers into scope.
  12. Uganda Bureau of Statistics (Peter Opio), “Country Progress Report on the Business Registers: The Experience of Uganda”, UN Committee of Experts on Business Registers, June 2019: fragmented registers, the missing common identifier, funding limits on updates, high informal death rates.
  13. National Institute of Statistics of Rwanda, Establishment Census 2023 report: 269,326 establishments; formal share of enterprises and of employment; census cycle.
  14. WIPO, Intellectual property statistical country profile 2024: Uganda: resident patent applications.
  15. Eurostat and OECD, Eurostat-OECD Manual on Business Demography Statistics, 2007: activity definition, two-year rule for deaths, administrative sources of business registers.
  16. Morten Jerven, “Why we need to invest in African development statistics”, African Arguments, 26 September 2013: the hostile witness; data demands, per diems and the statistical office as data broker.
  17. Nehemiah 7:4-5, 61-65, NIV, Bible Gateway: the register of returning families and the provisional ruling.
  18. Uganda Registration Services Bureau, Annual Report Financial Year 2024/25: Table 7, new companies registered (28,414 in 2023/24, 28,408 in 2024/25) and business names (27,297 and 21,695).

Leave a Comment

Your email address will not be published. Required fields are marked *