Context
ERC (erc.ua) is a major Ukrainian technology distributor and wholesale retailer, working across retail, wholesale, e-commerce, and systems-integration channels. Ideas for internal innovation weren’t in short supply.
The organisation had the scale and commercial complexity to generate good ideas from every direction. What it lacked was a credible route for an idea to move from discussion into a contained business test without becoming another large, slow project.
What didn’t exist was any way to decide which of them deserved time, an owner, or a test.
Diagnosis
Big companies don’t usually fail because they’re out of ideas. They fail because nobody owns the stretch between having an idea and proving it.
Every stalled initiative had the same three things wrong with it: unclear owner, fuzzy KPI, slow decision. None of that gets solved by more resources. It gets solved by a process light enough that using it beats going around it.
Without an owner, an idea turns into a recurring meeting. Without a success metric, a test never ends. Without a scope boundary, a small experiment starts looking like a threat to core operations and gets blocked, quite reasonably.
The other problem was bet size. The attempts were too big. Anything large enough to disturb the core attracts the scrutiny of a major project, which is exactly the friction that kills it.
Decisions
- Made the flow lighter, not stricter, so the official path became the fast path
- Required three things before any pilot: defined scope, a named owner, one success metric
- Kept experiments small enough that nobody had a reason to block them
- Got IT, marketing, and business onto the same flow so ideas stopped dying in the handoff
- Treated "close this one" as a legitimate outcome, not a failure
Results
Project lead time dropped 40% on tighter scoping and cleaner handoffs. One pilot that went through the new structure, a loyalty upsell widget, lifted average cart size 12% in split testing.
The playbook stuck, which matters more than the widget. The next twenty ideas now have a route to a test.
The thing I had to resist was the instinct to answer a process failure with more process. That would have rebuilt exactly the friction we were removing.