Against Product Intuition: Why I Stopped Trusting Mine

The short version

After twenty years in product, I stopped trusting my product intuition, and it made me better. Product intuition is pattern-matching on past experience, and it only works while the world stays similar to the one that trained it. The AI era rotates the patterns faster than any human gut can recalibrate, which means the more experienced you are, the more confidently you apply rules from a world that already disappeared. The fix is not to replace intuition with dashboards. It is to demote intuition from a decision to a hypothesis, then test the hypothesis against a real customer outcome, because AI made testing cheap enough that shipping on gut is no longer defensible. Experience still matters, but its job changed. It is for asking the right question and designing the fast test, not for making the call. Here is why a senior leader argues against his own intuition.

For most of my career, my intuition was my edge. Twenty years across Microsoft, Adobe, Salesforce, and four startups gave me a gut that was right more often than the room. People paid me for that gut. Then, sometime in the last two years, I noticed my intuition was confidently steering me wrong, and the only way to stay good at my job was to stop trusting the thing that had made me good at it.

Intuition is pattern-matching on a dead world

Strip the mystique off product intuition and here is what it is. A large library of patterns, compressed from experience, that fires fast when a new situation rhymes with an old one. That is it. It feels like wisdom. It is statistics, run on the dataset of your past.

Which means intuition is only as good as the match between the world that trained it and the world you are in now. When the world changes slowly, your patterns stay valid, and your gut is a genuine asset. The patterns I built at Adobe about how enterprises buy, the patterns I built at Salesforce about how platforms compound, those held up for years because the underlying world moved slowly enough.

The problem is when the world stops moving slowly. Then your pattern library quietly goes stale, and the dangerous part is that it does not feel stale. Confidence does not decay just because the world it was built on did. You keep getting the same gut signal, fired with the same certainty, trained on a world that no longer exists.

“

Product intuition is pattern-matching on a world that may no longer exist. The patterns do not announce when they expire. Your confidence in them does not change when they go wrong.

”

AI rotates the patterns faster than I can recalibrate

Here is what tipped me from trusting my gut to distrusting it. The rate of change.

Intuition can survive a changing world if it recalibrates at least as fast as the world moves. Mine could keep up for most of my career, because the fundamentals of how software got built and bought shifted over years, not months. AI broke that. What is buildable changed. What customers expect changed. What a product even is changed. And it changed faster than I could update my gut, because updating a gut requires lived reps, and the world now moves faster than I can accumulate reps.

I caught myself doing it. A founder showed me a product idea and my intuition fired a confident no, this does not work, I have seen this fail. I was pattern-matching to failures from a world where that thing was expensive to build and hard to distribute. In the current world it was cheap to build and trivial to test, and my expensive-world intuition was about to talk me out of something the cheap-world reality made obviously worth trying. My experience was the liability, not the asset.

That is the trap of seniority right now. The more patterns you have, the more confidently you apply them, and the more of them were trained on a world that is already gone. Junior PMs have less intuition to mislead them. This is the survivorship problem turned inward, and it is the same mechanism I unpack in why survivorship bias quietly wrecks PM judgment. You remember the patterns that worked in the old world. You do not feel them expiring.

The fix is not data, it is cheap tests

The obvious correction is replace intuition with data. That is not quite right, and the lazy version of this take stops there.

Dashboards describe the past too. They are intuition with more decimal places, and they can be just as stale, measuring a world that is changing under the measurement. Replacing a stale gut with a stale dashboard is not progress.

The real upgrade is replacing trusted intuition with cheap experiments. Use the gut for what it is good at, generating a sharp hypothesis fast, then refuse to act on it until you have tested it against a real customer outcome. The reason this is possible now, and was not for most of my career, is that AI made testing absurdly cheap. When a prototype took six months, you had to ship on judgment because you could not afford to test everything. Now you can. When testing costs an afternoon, shipping on gut alone is no longer prudence. It is laziness with a senior title.

“

The fix for stale intuition is not a dashboard, it is a cheap test. Use the gut to generate the hypothesis. Refuse to act on it until a real customer outcome settles the bet.

”

This is what I mean when I argue for building strategy from signals rather than conviction. The signal is current. The intuition is historical. Weight the current thing.

What experience is actually for now

I am not arguing that twenty years is worthless. I am arguing its job changed, and the people who do not notice the change are the ones it endangers.

Experience used to be for making the call. The senior person in the room had the best gut, so the senior person decided. That was rational when testing was expensive, because someone had to bet and the best bettor should bet. Now testing is cheap, so the value of experience moved upstream. Experience is for asking the right question and designing the fast test that answers it, not for skipping the test.

That is a real edge and a durable one. A junior PM can run experiments too, but a senior one with good pattern recognition designs better experiments, asks sharper questions, and reaches the decisive test faster. The twenty years are not a more trustworthy gut. They are a shorter path to the experiment that settles the argument. That is the version of experience that survives the patterns rotating.

The senior people in trouble are the ones who treat their experience as a license to skip the test. Trust me, I have seen this before. In a slow world that sentence was wisdom. In a fast world it is the most expensive sentence in the building, because the thing you have seen before happened in a world that no longer runs.

The discipline I actually use

When I feel certain now, I treat the certainty as information about my training data, not about reality. I ask, what world was this confidence built in, and is that world still here. Then I build the cheapest test that could prove me wrong, and I genuinely try to prove me wrong, because my gut is the incumbent and incumbents need challenging.

It is humbling to run your own intuition as a defendant rather than a judge. It is also the only way I have found to stay good at product while the ground keeps moving. The leaders who will struggle are the ones whose entire authority was built on a gut they will not put on trial.

Pick one thing to try this week

Find one product decision you are about to make on instinct. Write down what your gut says and how certain you are. Then ask one question, what world trained this certainty, and is that world still here. Build the cheapest test that could prove your gut wrong, and run it before you act. Do this with three decisions and you will find at least one where your intuition was confidently defending a world that no longer exists. That is the gap between feeling certain and being right, and AI just made it cheap enough to close.

Sources: Silicon Valley Product Group, on product judgment · Harvard Business Review, on the limits of expert intuition · Lenny Rachitsky, on experimentation and product decisions