How Can Mid-Sized Businesses Adopt AI Without Big-Business Budgets?

Alexander Vorontsov and Alexander Polomodov discuss AI adoption in companies with roughly RUB 2–15 billion in annual revenue and workforces ranging from a few dozen to several hundred people. These firms rarely have a laboratory or a large platform team; demand starts with a concrete problem such as shrinking margins, logistics, hiring, reporting, or regulation.

The successful case does not begin with buying licenses. The owners of a commerce company interviewed employees, ran a paid prototype competition, and involved a systems analyst who understood the actual tables, delays, and corrections in 1C. Ideas met data constraints and daily work quickly, while sensitive personal and margin data received separate access rules.

A contrasting case shows the cost of an initiative without an owner. At a growing staff-augmentation company, ideas for recruitment and internal mobility fragmented because nobody could connect the business process, data, technology, and adoption. A platform can be technically ready and still produce no business result when operating functions continue to work as before.

The practical route is to observe actual work briefly, select one to three use cases, and define a baseline, acceptance criterion, error cost, and stop condition before building. A smaller organization can move quickly, but it still needs employee incentives, support, and human alignment. AI lowers the cost of creating automation, not accountability for its consequences.