Knowing When to Stop Before It Wastes Anyone's Time
Stopped before customer PoC.
Problem
- Construction procurement means fragmented supplier communication, manual comparison, unclear availability, repeated follow-ups, and slow decisions.
- The question: were these pains strong and repeatable enough to justify an agentic AI procurement concept?
- And did the problem deserve deeper validation before anything was built?
What I did
- Framed the procurement pain hypothesis before choosing a product direction.
- Ran customer research and validation interviews with industry stakeholders.
- Built an advisory board of senior professionals from Turkey's construction and tech ecosystem.
- Took part in a pre-incubation program during early validation.
The decision to stop
The project stopped before customer PoC testing, when co-founder strategic misalignment became clear.
Stopping there kept the work at discovery, before any build.
What I would do differently today
- Pick one problem, instead of detailed market sizing for the pitch deck.
- Validate that one problem with many potential users.
- Launch fast. Then decide: move to another problem, or is this worth continuing? If it is, start improving the product.
- Take the co-founder choice slowly. a16z's advice is blunt: never rush it. I would run the early stages with that in mind.