12 Mental Models Every PM
Should Know (2026)
The thinking tools that compound over a PM career — second-order effects, inversion, Pareto, Goodhart's Law, and more. With product examples for each.
Practice Mental Models Daily — Free →1. Second-order effects
Every decision has downstream consequences the first look misses.
💡 PM example
A free tier that boosts signups (first-order) but trains users never to pay (second-order).
2. Opportunity cost
What are you NOT doing because you're doing this?
💡 PM example
Every feature you build blocks another feature. 'Adding one thing' is never really free.
3. Pareto principle (80/20)
80% of outcomes come from 20% of causes.
💡 PM example
Find the 20% of users whose feedback drives 80% of insight. Spend your time there, not on edge cases.
4. Inversion
Solve problems backwards — what would cause failure?
💡 PM example
Instead of 'how do we launch well?' ask 'what would make this launch fail?' Then prevent that.
5. Expected value
Outcome × probability, summed across possibilities.
💡 PM example
A 20% chance of ₹10Cr impact beats a 70% chance of ₹1Cr impact. Expected value: 2Cr vs 0.7Cr.
6. Regression to the mean
Extreme outcomes tend to be followed by less extreme ones.
💡 PM example
A spike in signups after a viral tweet will revert. Don't confuse variance for trend.
7. Confirmation bias
We seek evidence that confirms what we already believe.
💡 PM example
If you're convinced a feature will win, you'll notice positive feedback and dismiss negative. Deliberately seek disconfirming data.
8. Goodhart's Law
When a measure becomes a target, it ceases to be a good measure.
💡 PM example
Set DAU as the goal, and your team will find ways to inflate DAU without real value creation. Guardrails prevent this.
9. Sunk cost fallacy
Past investment shouldn't influence future decisions.
💡 PM example
If a feature isn't working after 3 months, killing it is a new decision. 'We've spent so much time' is irrelevant to future EV.
10. Chesterton's fence
Don't remove a fence until you understand why it was put up.
💡 PM example
Before deleting 'weird' features or flows, find out why they exist. Often there's a real reason that isn't obvious.
11. Circle of competence
Know what you know, and be honest about what you don't.
💡 PM example
A consumer PM pivoting to B2B should admit the domain gap and partner with experts — not fake expertise.
12. Leverage
Small inputs that produce outsized outputs.
💡 PM example
Fixing onboarding (high leverage) beats shipping 10 new features (low leverage). PMs should hunt leverage continuously.
FAQ
Why do mental models matter for PMs?
Mental models are pre-compiled thinking patterns. Instead of reasoning from scratch every time, you recognise patterns and apply the right model. PMs who internalise a dozen models make faster, higher-quality decisions than PMs who don't — especially under time pressure. Think of them as the mental equivalent of well-named functions: they let you think in bigger chunks.
How do you actually learn to use mental models?
Reading about them isn't enough. Three steps: (1) pick ONE model per week, (2) look for situations where it applies and explicitly name it — even in your own head, (3) after 2 months, you'll use the model automatically. Rotating through 12 models takes ~3 months and changes how you think permanently.
Build PM Thinking in 2 Minutes a Day
Daily scenarios that force you to apply mental models — not just read about them.
Start Free Trial →