Situation
- • Product: UMI.CMS/UMI.RU, starting from 65k commercial users.
- • Release cycle was once every few months.
- • Many clients avoided updates due to breakages.
- • Issue tracker had several thousand bugs, plus 5-10 new bugs per day.
UMI.CMS/UMI.RU — Transition from Boxed Product to SaaS
Role: CTO | Timeline: 2010-2012 | Team size: 10-15 (Dev, QA, DevOps)
- +50%
- revenue growth
- 1M+
- users
- 65k
- commercial users
I led an end-to-end engineering transformation under high uncertainty: there were no formal KPIs at kickoff, but the business objective was clear - stabilize product delivery, restore customer confidence in updates, and build a scalable SaaS operating model.
Role & authority
- Position
- CTO, UMI.CMS/UMI.RU — a Russian CMS/SaaS platform (65k commercial customers) — 2010–2012
- Reported to
- the CEO
- Scope
- End-to-end product ownership, SaaS transformation of engineering, a 10–15 person team, delivery processes, quality and release practice.
Recommendation
"He showed himself as an excellent technical leader. He put the development process to a new level, and the best practices he taught the team significantly improved the quality of the product."
Task
- • Stabilize product delivery without slowing product evolution.
- • Rebuild the engineering team and operating model.
- • Establish a repeatable quality, release, and deployment system.
- • Create the foundation for scalable SaaS growth.
Action
- • Rebuilt the team, replacing the conservative core and hiring 50%+ new specialists.
- • Introduced quality control processes, structured bug tracking, and QA practices.
- • Implemented DevOps practices and CI/CD pipelines for reliable delivery.
- • Built a distributed test farm across dozens of environments.
- • Migrated product update mechanics to a safer and more predictable model.
- • Drove adoption through hands-on leadership, coaching, and personal example.
Skills applied
- • Team leadership and rebuild: hiring, replacing a conservative core, onboarding, and quality rituals.
- • QA and engineering culture: structured bug tracking, defect control, shared acceptance rules.
- • DevOps and CI/CD: predictable releases, delivery automation, and safer rollbacks.
- • A distributed test farm scaling verification across dozens of environments.
- • Change management: lead-by-example, coaching, and day-to-day methodological support.
- • End-to-end product lifecycle ownership as CTO—from process design to release practice.
Result
Release cadence
from months to ~weekly
New bug inflow
from 5-10/day to <=5/week
Business impact
+50% revenue
and user base: 65k -> 1.8M
User base
grown to 1.8M users
Early stabilization became visible in about 6 months: the product became safer to update and enterprise customers were no longer afraid of new versions. After ~1.5 years, the outcome became publicly visible: the user base grew to 1.8M, demand for new features increased, and scaling became predictable.
Case Materials
- • blog (ru): dobryakov.com/blog
- • blog (en): dobryakov.net/blog
- • distributed asynchronous systems course: YouTube playlist
Grigoriy Dobryakov — AI-Driven Head of Engineering
Revenue +50%
65k → 1M+ users