The dead horse rule: decide on closing before you start

I have had 5 failed startups, and in almost every one the most expensive mistake was not in the product. It was that I could not stop in time. The decision to close cannot be made at the moment things have gone badly: by then you are no longer defending the project, you are defending the fact that you were right. So the deadline and the numbers get written down in advance, while you are not in love yet.

Why the decision cannot wait

The sequence is always the same. You put in money, time and belief. The more you have put in, the stronger the feeling that closing now means writing all of it off. So you add another month, a little more money, one more attempt. With every step the decision to close gets more expensive and harder.

From there comes a thought that took me far too long to reach. At the moment things have gone badly, you are no longer the person who can judge soberly. You are inside it. You are invested. Which means the decision has to be made in advance, while you are still impartial.

The tool is called the dead horse rule. I heard it stated plainly from my aunt and from Pavel Osokin on a podcast, and before that I learned it the hard way.

How it works in practice

Before you start you write down two things. A deadline, usually three months or six. And a list of metrics you will use to decide when that date arrives. Not “let us see how it goes”, but concrete numbers: this many paying customers, this conversion, this response rate on first contact. You write it while the enthusiasm is at its peak and there is still no reason to lie to yourself.

On the appointed date you simply check against the list. You do not discuss it, you do not look for excuses, you do not say “but we have one client almost closing”. You check.

The metric has to be about the market, not about you

This is the subtlety that decides everything. Your own activity is under your control, which is exactly why it is easy to fool yourself with it. The market is not.

The left column always looks convincing in a report and says nothing about whether anyone except you needs the product.

The uncomfortable part

A list like this is needed not only by a new project. It is needed by the one you are probably dragging into its third year right now. The difference is that writing it there hurts: you have to state honestly at which numbers you admit you are not going further.

What cheap development changed

Over the last year this story has gained a new side. When testing a hypothesis cost a year and a half and hundreds of thousands, letting go of it was nearly impossible. Now a prototype comes together in a week and for small money, and the attachment does not have time to grow.

Cheap development does not cure timelines. It cures the inability to quit.

What a defeat actually is

A project closed on time is not a defeat. The defeat is the project you keep feeding with money and attention for another two years after it is already clear the horse is dead.