Product Judgment vs Frameworks: What No Framework Teaches

The short version

Twenty years across Microsoft, Adobe, and Salesforce taught me the one thing no PM framework can teach: judgment under ambiguity, which only comes from being wrong at scale, repeatedly. Frameworks are training wheels. They teach you the moves, RICE and JTBD and opportunity solution trees and the rest, but they cannot teach you which move to ignore, what the data is not saying, or when to override the method entirely. That part is earned, not read, and it forms only by making consequential decisions and being wrong enough times that the pattern finally clicks. The AI era over-rewards framework fluency, because frameworks are exactly the explicit, testable knowledge machines are good at and interviews screen for, while earned judgment is tacit and slow and hard to demonstrate in a loop. That is exactly backwards for where things are going, because judgment is becoming the scarce asset. Here is the difference between product judgment and frameworks, and why the thing you cannot teach is the thing that matters most.

This is a capstone post, the one I have been circling for a while. I have read the books and taught the methods and run the rituals. And the thing I actually trade on, the thing that makes me worth listening to, is not any of that. It is two decades of being wrong in consequential ways and surviving long enough to learn from it. No framework gave me that. No framework could have.

Frameworks are training wheels, and that is a compliment

I want to be fair to frameworks before I come for them, because the dismissive version of this argument is lazy.

When you cannot yet balance, training wheels are the difference between riding and falling on your face. A framework gives a junior PM a safe default, a way to not make the most obvious mistake, and a shared language so their team knows what they are doing. RICE stops you from prioritizing by whoever shouted loudest. An opportunity solution tree stops you from jumping to the first solution. Jobs to be done stops you from confusing the feature with the need. These are real services, and a PM who skips them to "use their judgment" before they have any is just falling on their face with extra confidence.

So I am not saying frameworks are useless. I am saying they are training wheels, and the entire point of training wheels is that you take them off. They get you riding. They do not teach you to balance, because balance is not a thing you can be told. It is a thing your body learns by almost falling a hundred times. Judgment is the balance. The framework is the wheels. Confusing the two is the central mistake of a lot of product careers.

“

Frameworks teach you the moves. Judgment is knowing which move to ignore. The first you can read in an afternoon. The second takes a decade of being wrong.

”

What judgment actually is, and where it comes from

Judgment under ambiguity is pattern recognition built from consequence. It is the sense, often hard to articulate, that this customer is telling you something different from what they are saying, that this metric is going to mislead you, that this is the moment to depart from the playbook and this other moment is not. It is knowing what the data does not contain.

And it comes from exactly one place: making real decisions under real stakes, watching them play out, and being wrong enough times that the pattern finally forms. There is no shortcut. You cannot read your way to it, because the lesson is not propositional. It lives in the gap between what you predicted and what happened, and you only get that gap by predicting and being wrong, over and over, with something actually on the line.

I learned the most from the calls I got wrong. The feature I was sure customers wanted and they did not. The market timing I misjudged. The reorg I drew around a people problem that shipped a fault line into the product. Each of those was a wrong prediction at real scale, and each one installed a piece of judgment that no book contained, because the book did not know my specific situation and judgment is mostly about the specifics. This is the same reason I keep warning about survivorship bias in product advice: the lessons that transfer cleanly into a framework are usually the ones that mattered least, and the ones that mattered most are too situational to package.

The AI era is rewarding exactly the wrong thing

Here is where this stops being a reflective career post and becomes an argument about right now.

Frameworks are explicit, teachable, repeatable knowledge. They are exactly the kind of thing AI is excellent at, exactly the kind of thing that is easy to test in an interview, and exactly the kind of thing that is easy to mistake for competence. So the AI era is quietly cranking up the reward on framework fluency. The candidate who can fluently articulate every method, apply RICE on command, sketch a clean opportunity solution tree on the whiteboard, looks impressive and screens well in a 45-minute loop. Meanwhile the thing that actually matters, earned judgment, is tacit, slow to build, and almost impossible to demonstrate in a structured interview. So it gets under-rewarded precisely as it becomes the scarce asset.

This is dangerous, and it is the through-line of everything I write. When build capacity was the constraint, framework fluency was a reasonable proxy for value, because the bottleneck was execution and frameworks help you execute. But the constraint has moved to judgment and taste, and frameworks are the one thing that does not help you there. An AI can already apply your framework better than you can. What it cannot yet do, and what it makes scarcer by comparison, is the earned judgment to know when the framework is wrong for this specific situation. We are building a hiring and promotion machine that selects for the commoditized skill and screens out the scarce one.

“

AI can apply your framework better than you can. What it cannot do is know when the framework is wrong. We are rewarding the part machines do best and starving the part only humans earn.

”

How to build the thing that cannot be taught

If judgment only comes from being wrong at scale, then the move is to maximize the number of consequential decisions you make and to treat the wrong ones as the actual curriculum.

Make as many real calls as you can get your hands on. Not opinions in a doc, decisions with consequences, where you predicted something and reality graded you. Then review the misses honestly, because the gap between your prediction and the outcome is the only place judgment is manufactured. A PM who makes ten consequential decisions a year and reviews their misses will out-judge a PM who read forty books, every time.

And here is the gift of the AI-native present, the one bright spot in an otherwise backwards incentive. Cheap experiments let you be wrong faster. When building collapses to nearly free, you can run more real tests, make more real calls, and get more reality-graded predictions per year than any previous generation of product people could. The same shift that is commoditizing framework execution is also handing you the fastest judgment-building machine ever built, if you use it to make and review a higher volume of real decisions instead of just generating more demos. Be wrong faster, on purpose, and review every miss. That is the whole curriculum, and it is the one no framework contains.

What to do this week

Stop trying to learn another framework. You have enough. Instead, pick one real decision you are facing this week, write down your prediction before you decide, the specific thing you believe will happen and why, and then actually check it against what happens. That written, checkable prediction is worth more to your judgment than the next ten frameworks you could read, because it creates the one thing judgment is made of: a gap between what you thought and what was true.

Do that every week for a year. Keep the predictions, review the misses, and watch the pattern recognition form. The frameworks will still be there when you need a default. But the thing that makes you worth listening to, the judgment under ambiguity that no method can teach, only comes from being wrong at scale and paying attention. Start being wrong on purpose, faster, this week.

Sources: Silicon Valley Product Group on product judgment and senior product leadership; Lenny's Newsletter on PM frameworks and decision making; Harvard Business Review on tacit knowledge and judgment under uncertainty.