The Product/Market Fit Matrix [Data-Driven Iteration]
โThe classic advice for reaching product/market fit (โbuild something people wantโ) is true but useless when youโre in the gap and need to know what to do this week. The product market fit matrix I use, which I call the PMF Iteration Matrix, is a 2x2 that gives you a concrete focus: two things to work on (Value and Pains) seen from two perspectives (Customer and Product), which makes four quadrants.โ
For each quadrant you write a set of statements and ask real users to stack rank them, so instead of guessing you get a ranked, data-backed list of what to work on next. Value ร Customer sharpens your value proposition. Value ร Product shows which features to double down on and which to monetise. Pains ร Customer reveals your customers' highest-priority problems. Pains ร Product ranks the frustrations to fix on your roadmap. It's a more continuous, data-driven complement to the one-off Sean Ellis 40% test.
"Do whatever is required to get to product/market fitโ โ Marc Andreessen, 2007
In the gap between finding traction and PMF, this quote is about the most descriptive advice you're going to get. Just go build a product that people really want.
That's frustratingly vague for a founder like me living in this gap right now, so I set out to find something more useful. Instead of guessing at random, hoping something I did would eventually trigger this elusive product/market fit, I wanted a product market fit matrix to focus my effort on what would actually take us one step closer to PMF each week.
Here's what I've been using for the past few months:
How the product market fit matrix works
The PMF Iteration Matrix gives me two things to focus on (Value and Pains) and two perspectives to view them from (Customer and Product). I use customer stack ranking surveys to turn the value and pain statements in each quadrant into a ranked list, so I can prioritise what the team works on. It blends rich qualitative insight, grouped by area of effort, with real ranked data. Here's the breakdown:
1. Value x Customer
The right place to start is the "big-picture value" your product delivers for customers. This is basically your overall value proposition: how is their life better after using your product. Most pre-PMF startups are juggling a few value propositions at once to work out what lands best with target customers. Here are the 5 value prop statements I recently asked a sample of our active users to stack rank:
As you can see in the "Score" column, we have a clear winner (thankfully, our current messaging). I've always tested a handful of value propositions in customer conversations and content marketing, and this result is solid validation that I've been pushing the right way, and that we should double down on "better data for decision-making", our best candidate for value/customer fit.
But this matters beyond the value prop itself. Knowing how we improve customers' lives sharpens our sales messaging: more targeted case studies, before-and-after examples, use-case suggestions, impact-led pitches, and our whole monetisation strategy.
2. Value ร Product
Next, I ask users to stack rank statements that tie value to different product features. That gives me data on two questions: (i) which features let our power users thrive, and (ii) what are we under-monetising?
We recently found that our power users often relied on spreadsheet exports from OpinionX to get detailed data for screening follow-up interview participants or comparing segmented results. But exporting also pulls value out of our product, so we decided it should only be available to premium-tier customers. We're now feature-gating exports and instead nudging freemium users to invite co-workers as in-product collaborators (which also spreads awareness of the product; win-win!).
We also found that users who interacted with our sample survey had a much higher activation rate. So we expanded from one sample survey into a full gallery of examples, and changed onboarding to point new users to that gallery instead of dropping them straight into a blank survey.
3. Pains ร Customer
We do periodic โCustomer Problem Stack Rankingโ research projects (a research method coined by Shreyas Doshi) to understand the priority of our customerโs overall problems stack. We recognize that weโre laser-focused on the problem weโre trying to solve โ to help product teams get better data for big decisions like roadmap prioritization โ but in reality, product managers are juggling a whole range of different problems.
For example, when we asked 152 PMs to stack rank their biggest problems in 2021, the top-ranked problem was "Finding PMF for a new product/feature is an ongoing challenge." That result put this very blog post on my content backlog. And we're not the only ones doing this: Fred used Customer Problem Stack Ranking to create viral problem-focused social media content for his startup Bliinx.
Like Fred, we pull the highest-priority problems from these stack ranked lists and use them to fuel our Pain Point SEO strategy. As a result, almost all our blog posts convert to product sign-up at 2% to 25% (0.2% to 0.4% is usually considered good for most software or service businesses). Our guide to Needs-Based Segmentation is a great example.
4. Pains ร Product
So far we've used stack ranked data to (1) tailor our sales, (2) improve our monetisation, and (3) sharpen our marketing. The last quadrant (4) is about improving the product itself.
At OpinionX, we share a quarterly stack ranking survey with our users to help prioritise the roadmap. But instead of asking them to rank the features they want, we ask them to rank their biggest barriers, frustrations and challenges using our product. That decides which product sprints we take on over the next three months. The screenshot below shows our current roadmap prioritisation stack rank. Each statement starts at a base score of 1500, so it's clear which three problem categories we need to prioritise right now:
Who you collect data from matters as much as the data itself
None of this data is useful if we collect it from the wrong people. We collect demographic, firmographic and pricing-tier data on every stack rank participant, so we can later segment the results to show the priorities of our exact ideal customer profile. Below is an example filtered to only my "Pro" customers:
A better way for pre-PMF teams to iterate toward product/market fit
The only widely-agreed method for measuring product/market fit is the Sean Ellis PMF Survey. You ask customers "How would you feel if you could no longer use our product?" and if over 40% say "Very disappointed", congrats, you've reached PMF. If you're under 40%, good luck and try harder next time. Not so useful, really.
Before the PMF Iteration Matrix, I felt stuck in an exhausting scattergun approach with no end. Every user conversation, feature request, or new idea could drag me off-course. Unfortunately, that's everyday life for most pre-PMF teams.
Today, the product market fit matrix reassures me that I'm making real progress each week, using our existing traction to get one step closer to product/market fit. I hope it proves as useful for your team as it has for ours.
Frequently asked questions
What is the product market fit matrix? The product market fit matrix is a 2x2 framework for iterating toward product/market fit. The two axes are what you focus on (Value and Pains) and whose perspective you take (Customer and Product), giving four quadrants. You rank statements in each quadrant with real users to decide what to work on next.
How do you measure product/market fit? The best-known method is the Sean Ellis survey: ask users how they'd feel if they could no longer use your product, and if over 40% say "very disappointed", you've likely reached PMF. It's a one-off snapshot, so the PMF Matrix works better for continuous, week-to-week iteration.
What is the Sean Ellis PMF survey? A single-question survey that asks "How would you feel if you could no longer use our product?" with the answer "very disappointed" as the key signal. The widely-cited benchmark is that 40% or more choosing "very disappointed" indicates product/market fit.
How do you use stack ranking to reach product/market fit? Write a set of value or pain statements for each quadrant of the matrix, then ask real users to rank them. The ranked results tell you which value proposition, feature, customer problem or product frustration to prioritise next, backed by data, not guesswork.
If you enjoyed this essay, the best way you can show your appreciation is by subscribing to our newsletter. We share actionable advice on user research for customer-led startups, and every new subscriber is an extra dose of motivation to make the next essay even better than the last.
When I'm not writing essays like these, I'm busy building OpinionX, the platform for advanced market research surveys. We help thousands of product teams inform their big decisions with real data on what matters most to their users. Create your own stack ranking survey for free in minutes at app.opinionx.co.
Daniel โ๏ธ