Frameworks are supposed to reduce a hard problem to a set of moves you can make. Most product-market fit frameworks fail that test — they give you a diagram to nod at, not a decision to make. You stare at the pyramid, agree it's true, and go back to work no wiser about what to do next.
So let's be practical. Four frameworks have genuinely earned their place, and the confusion clears the moment you realize they aren't competing — they're answering different questions. One tells you how to build toward fit, one tells you whether you have it, one tells you how to improve it, and one tells you how to keep it as you scale.
The four frameworks, at a glance
| Framework | Best for | Core lens / metric | Where it stops |
|---|---|---|---|
| Olsen's PMF Pyramid | Building toward fit | Target customer → needs → value prop → features → UX | Diagnostic, not a live number |
| The Sean Ellis Test | Measuring fit | % "very disappointed" (40% line) | Tells you if, not how to build |
| Superhuman PMF Engine | Improving fit | Segment by very-disappointed users | Needs a measured baseline first |
| Balfour's Four Fits | Scaling fit | Market · Product · Channel · Model | Assumes you already have PMF |
1. Dan Olsen's Product-Market Fit Pyramid — how to build it
From The Lean Product Playbook, Olsen's pyramid has five layers. The bottom two are your market: your target customer, and their underserved needs. The top three are your product: your value proposition, your feature set, and the UX. Fit is the strength of the relationship between the product you built and the market underneath it.
Why it's useful: it turns "get product-market fit" into a stack of specific, sequenced decisions — pick the customer, find the underserved need, define the value prop that meets it, then the features, then the experience. Most fit failures are really a crack low in the pyramid: a great UX built on a need nobody urgently has.
Where it stops: the pyramid is a build-and-diagnose tool, not a measurement. It helps you construct fit and locate why you're missing it — but it won't hand you a number that says whether you got there. For that you need the next one.
2. The Sean Ellis Test — how to measure it
The most-used PMF framework in the world is a single question: "How would you feel if you could no longer use this product?" The share of engaged users who answer "very disappointed" is your score, and 40% is the threshold that has historically separated products with fit from products without it.
Why it's useful: it's the rare fit framework that produces a real, repeatable, comparable number. You can track it over time, benchmark it, and act on it. It's the closest thing PMF has to a thermometer.
Where it stops: it measures fit, it doesn't manufacture it. A low score tells you that you're not there without telling you which layer of the pyramid to fix. The measurement is the starting gun for the work, not the work itself. (One caution: the score is only as honest as your sample — who you survey matters as much as how many.)
Run the measurement framework in minutes
PMFtracker is the Sean Ellis test as a living system: it surveys your engaged users, calculates your "very disappointed" score on a clean sample, and tracks it over time — so the measurement layer of your framework runs itself.
Measure your PMF score free → 14-day free trial · No credit card3. The Superhuman PMF Engine — how to improve it
Rahul Vohra turned the Sean Ellis question from a one-time reading into an improvement engine. The move: segment users by their answer. Study the people who'd be "very disappointed" to find who your product is truly for and why they love it — then double down on them. Separately, look at the "somewhat disappointed" fence-sitters and win over the ones who cite reasons you can fix, while ignoring the rest.
Why it's useful: it converts a static score into a roadmap. Superhuman used it to climb from 22% to 58%. It's the only framework here that tells you what to build next to raise the number.
Where it stops: the engine needs fuel — a measured score and enough responses to segment. It's a framework for improving fit, which means you have to be measuring first.
4. Brian Balfour's Four Fits — how to scale it
Balfour's insight is that product-market fit alone doesn't make a company grow. Sustainable growth needs four fits aligned: market-product, product-channel, channel-model, and model-market. A great product in the right market can still stall if it doesn't fit the channels you can afford or the business model those channels imply.
Why it's useful: it explains the startups that have obvious fit and still can't grow. Once you've validated PMF, the Four Fits is the framework for making it scale instead of stall.
Where it stops: it assumes you've already got PMF. It's a scaling framework, not a getting-there framework — which is exactly why it comes last.
The synthesis: one framework, four stages
Here's the reframe that makes all four click: they're not rivals, they're a sequence. Each hands off to the next, and together they map cleanly onto the stages of product-market fit:
- Build the right product for the right market — Olsen's Pyramid.
- Measure whether you got there — the Sean Ellis test.
- Improve the score toward strong fit — the Superhuman engine.
- Scale without losing fit — Balfour's Four Fits.
The engine that drives the whole loop is the measurement in step two. Without a live PMF number, "build" is guessing, "improve" has no target, and "scale" is flying blind. That's why the practical version of a PMF framework always centers on a score you keep taking — and why the teams who reach durable fit treat measurement as continuous, not a one-time audit. Before you scale on any of it, it's worth validating the fit is real and not a mirage.
Put the framework on autopilot
Measure your PMF score, watch it trend, and segment your very-disappointed users — the measure-and-improve core of the whole framework, running continuously.
Start free → 5-minute setup · No credit card
