What a stopped project taught us
Stopping early isn't failure. It's the hardest and most profitable decision in a build cycle.
In the Lab, we document the explorations we stop as thoroughly as the ones we continue. Same placement, same level of detail.
This isn't transparency for its own sake. Stopped projects produce a kind of knowledge that successful ones never do.
The bias that prevents stopping
An exploration in motion generates its own momentum. Every week invested raises the perceived cost of abandoning it, and the team starts looking for reasons to continue rather than reasons to stop.
The mechanism is well known and entirely useless to know: understanding it doesn't protect you from it. What protects you is a decision made before you start.
So when we state a hypothesis, we fix two things at the same time: the date we'll decide, and the signal that would make us stop. Written, dated, shared. When the day comes, the question isn't "do we want to continue?" but "is the signal there?".
What a stopped exploration leaves behind
We stopped an exploration after six weeks. The initial hypothesis was plausible, several interviewees had confirmed it, and the prototype worked.
It was stopped for one reason: we couldn't articulate what would have disproved it. Every attempt produced a broader restatement, never a more testable one. A hypothesis you can't break isn't a hypothesis — it's an opinion you're attached to.
Three things remained.
A map of the ground. The interviews still hold. They describe how a profession actually works, independently of what we wanted to do with it. That material was used elsewhere, some months later.
A reliable negative signal. We now know a direction leads nowhere, and we know why. That's a durable saving: the same idea will come back in different packaging, and we'll have the answer.
A correction to method. We added a step: explicitly state the disproof criterion before building anything. If we can't, the exploration doesn't start.
The real cost
The cost of a project stopped at six weeks is six weeks.
The cost of the same project stopped after launch is several months, a budget, a team's credibility, and often the conviction that it shouldn't have been explored at all — the most expensive of the three.
What it takes
A culture where stopping isn't an admission. That can't be declared: it gets built by making stops visible, documenting them seriously, and never treating them as awkward subjects.
Which is why our stopped explorations stay published on this site, with their status and what we learned. A lab with only successes isn't a lab. It's a showroom.
Category EXPLORE · Reading time 5 minutes