Purpose. Existing figures on women founders mostly reflect investor decisions. This paper looks at the stage before that: how women and men applicants differ at the time of application, and which factors explain the gap.
Data and method. We analysed 1,170 complete applications to UzCombinator Batch 2 that answered the gender question. Users, revenue, paying customers and company registration were compared across groups; gaps were tested within the same startup stage and in a statistical model controlling for age, sector, stage, customer type and freemium model.
Results. 19.1% of applicants are women. The share of applicants with paying customers is three times higher among men (24.2% vs 7.9%). Most of the gap in users and revenue is explained by women applying more often at the idea stage (39.0% vs 20.2%). The gap in paying customers persists after all controls (12.4% vs 22.6%). The share of women is lower in Tashkent city (12.4%) than in the regions (23.2%).
Limitations. The sample covers applicants to a single accelerator; metrics were asked as yes/no questions, and the number of women applicants with users is small (46).
Keywords
ayol tadbirkorlar
startaplar
akseleratorlar
gender tafovuti
hududiy tafovut
O'zbekiston
Full text
Similar publications
Research InstituteCase studyFrom $150,000 to $800,000: UzCombinator Demo Day figures and the startup that secured the most investment2026Investment and venture
Case study·Investment and venture·November 5, 2026
The second batch of UzCombinator has ended, and Demo Day has already passed. Now the question is: what changes and growth were observed in the second batch compared to the first?
Ahmadjon Qahramonov, Mashhurbek Mahmudov, Murodillo Ismoilov and 1 more
Only funded startups are announced in Uzbekistan, so the process of building a startup, including its attempts, concentrations, and failures, remains unknown to the public. This article contributes to filling that information gap using the 3,573 applications submitted to UzCombinator’s (UzC) second batch in June–July 2026, of which 1,360 complete applications form the analytical sample. It describes the applicant pool by sector, stage, customer type, and revenue model, and compares it with the accepted cohort (n=31) and with Y Combinator’s Summer 2025 batch.
AI and education are the most common sectors among applicants, and AI applied to education is the largest AI combination. AI appears mainly as a function added to existing sectors to solve domestic problems. Selection differs sharply from the pool: no idea-stage startup was accepted, although idea-stage founders made up 26.2% of applicants, while revenue- and growth-stage startups were roughly three times more common among those accepted. The accepted cohort is more AI-based (51.6%) and mostly B2B (87.1%), moving it closer to the Y Combinator profile. Subscription is the most common revenue model across the pool.
The study is descriptive and is not a census of Uzbek startups. It covers founders who chose to apply to one accelerator in one window, so founders who never applied or never heard of the program are absent by construction, and therefore, the findings do not establish causes.
Keywords: startup ecosystem, Uzbekistan, accelerators, entrepreneurship, artificial intelligence, emerging markets, Central Asia
This report presents an anonymised, aggregate analysis of the 3,532 applications, and the 4,233 founders behind them, submitted to UzCombinator's second season between 8 June and 20 July 2026.
The founders are young: 66% of those who gave their age are 21 or younger; women make up 24% and solo founders 84%. A third of the applications that named a region come from Tashkent city, the rest from 13 regions and from abroad. Artificial intelligence (406) and education technology (402) lead; business (788) and consumer (775) customers are almost even.
Of the 1,635 teams that named their stage, 41% are at MVP; the step from first users to first revenue is the narrowest. 94% of applications report no revenue yet, 15% are raising now and 1% have raised before. The programme's acceptance rate was 0.65% (23 teams).
The report is intended as an open source of data on the size, composition and stages of Uzbekistan's startup ecosystem. Every figure is given with its base (n).