Uzbekistan is experiencing the early stage of venture market formation. The first funds are emerging, accelerators are conducting selection rounds, and startups are presenting at pitch sessions. Yet systemic success remains limited: most projects do not survive their first year, and those that do rarely scale beyond the country’s borders.

Kun.uz spoke with venture investor Azizbek Kurbanov, founder of the insurance marketplace Sug'urta Bozor and former managing director of a venture fund. We discussed the typical mistakes made by Uzbek startups and what the local ecosystem needs for sustainable growth.

Problem One: Launching Without Domain Expertise

One of the most common mistakes is when founders enter an industry in which they have never worked. Someone with no background in logistics decides to build a delivery service. A team with no understanding of manufacturing processes attempts to develop an ERP system for factories.

“People have never done anything in the industry, have never worked in it, yet they decide to launch a product for that very sector. And they don’t even have an advisor with relevant experience on the team. It simply doesn’t work like that,” Kurbanov notes.

Solutions that succeed in more developed markets have been tested repeatedly. Interface design choices, product-launch sequences, monetization models — all of this can be studied by examining foreign competitors. However, without understanding the industry’s specific nuances, mere copying does not lead to meaningful results.

Problem Two: Conflicts Among Co-Founders

Venture market statistics show that disagreements between co-founders are among the leading reasons startups shut down at an early stage. In Uzbekistan, where a business partnership culture is only beginning to take shape, this issue is particularly acute.

Co-founders spend 10–12 hours a day together. They go through periods without funding together, raise investments together, and make difficult decisions together. Misalignment in values or working styles becomes visible quickly and often proves fatal for the project.

A telling example is the insurance marketplace Sug'urta Bozor. The project scaled to tens of thousands of sold policies totaling more than one million dollars largely because the co-founders immersed themselves in operational work: analyzing user behavior through Yandex.Metrica, reviewing session recordings, and iteratively improving conversion rates. The result: conversion grew from 1–2% to 25%.