Product-Market Fit (PMF) is a term that has been used for so long in the startup ecosystem that has lost its meaning. The concept of PMF can radically help startups focus, prioritize and build the right things at the right time. Unfortunately, founders are not usually able to explain what PMF means in a simple and actionable sentence.
With this in mind, we want to share an easy-to-implement approach to “measuring” and “monitoring“ PMF that founders can use to make sure their startup is always working toward it. Spoiler alert: PMF is not one single discrete event.
What exactly is Product-Market Fit?
The evolution of the concept
Paul Graham, Co-Founder of Y Combinator, made the concept part of YC’s mantra: “make things people want.”
All these statements are great, right? But, as entrepreneurs, it is not so clear how to turn these concepts into actions, measurements, processes etc.
Why PMF is important
It is counter-intuitive to start with the why before we understand its definition, but ultimately PMF is essential because, without it, scaling a startup is too risky! Startups that have weak signals of PMF should allocate scarce resources towards optimizing for learning what customers find truly valuable about their product or service before it is time to focus on and invest in growth.
We mentioned before that PMF is not a discrete event, so let’s elaborate on what we mean by “without it.”
A never-ending approach
Let’s start by questioning a few preconceptions about Product-Market Fit:
- First, is the idea that detecting PMF is straightforward and obvious. This is by no means a simple thing. All startups are focused on showing the evolution of a set of Key Performance Indicators (KPIs) like Monthly Recurring Revenue, Monthly Active Users, etc, but these are outcomes from having strong PMF; the framework we propose first to startups is aimed at optimizing first the inputs to these outcomes.
- Second, there’s the belief that, once you achieve product market fit, you can’t lose it. This is simply not true. Founders fall into the trap of envisioning their business following an exponential growth curve. The reality is that the market is so dynamic that your users’ and customers’ behavior change over time, as does the competition. So in reality startups follow something closer to an S-Curve. Hence, you will need to keep active surveillance to make sure that you continue to work on the right things that will deliver the value at every moment in time.
- And third, the anticipation that the discovery of PMF is a “big-bang” event or discrete milestone. To counter this view, we like to share with our startups the concept of a PMF Continuum. Early in the program, we emphasize helping founders understand that PMF is a process, a never-ending approach that startups need to follow to keep PMF under surveillance.
The best thing we can do given an environment of extreme uncertainty is to gather 📡 qualitative and quantitative signals that provide evidence that we are in a good market with a product that can satisfy that market, as Marc Andreessen stated back in 2007.
Assessing where a startup lies on the PMF continuum helps answer the following questions, key in early-stage startups:
- Does the problem we solve exist?
- Is the problem important enough?
- Is the market for our product a “good” market?
The main objective of the startup team is to strengthen the PMF signals in order to confirm—to investors and other stakeholders—that working on initiatives to scale the business will pay off with growth.
Looking for Signals
What are these qualitative and quantitative signals? And how do we read them?
1. Qualitative Signal: Listening to Customers
These signals relate to the relationship your users and customers have with your product or service. You need to get this information somehow. There’s one question that we keep bringing up during the first week of the program to all team members of the startup:
❓No judgement! When was the last time you had a “conversation” with a user?
It is true that there’s saturation of the use of the term customer centric. We have all heard that startups need to be customer focused, customer obsessed, customer centric, etc. No matter what you call it, the important thing here is to understand the importance of it and encourage all team members to actually do the hard work. Yes, you really need to talk to customers. Apart from the benefits that this will have on your product strategy and development, it is a great way to acquire and validate qualitative PMF signals.
Some examples? They might seem placed here for comedic purposes, but they are not. It Is important to work toward the following signals:
Passion and Love:
Pace and Speed:
Ultimately, as Lenny Rachitsky mentions in his article: there’s three things that you should aim for:
- Passion – people get excited when talking about your idea.
- Skin in the Game – your users willingness to pay (from users to customers)
- A Clear Why – users’ pain points well known
1/ What is “product/market fit”? I’m not sure I can give you a definition. But maybe I can share what the subjective difference is in how it feels when you have it and when you don’t. Founding a startup is deciding to take on the burden of Sisyphus: pushing a boulder up a hill.— Emmett Shear (@eshear) July 27, 2019
2. Quantitative Signal: Historical Usage Data
The “many people use our product” statement. The problem is that it’s not accompanied by context. It sounds cool, but what are they really saying?
Our suggested approach is very pragmatic. Go as far back in time as possible and gather usage data in a User-Level Event Log. Like this:
This very basic data set can unlock the door towards calculating two of our favorite measurements: user retention and engagement.
The main goal of retention is to have your users consistently returning after their first use of your product or service.
Retention is important to growth and showcases signals of good PMF:
- If you are not keeping your existing users, acquiring new users fills a leaky bucket
- You might grow based on new users for a while, but eventually you will use up all possible acquisitions
- User numbers will plateau, then decline
But what are you measuring as the event, just a logged-in user? This is key. First, the team needs to identify the metric that best represents the value of the product (what some call the “North Star metric”) and determine what is the ideal frequency with which that metric should happen. Some examples:
With the value metric identified, the idea here is to plot the evolution of users triggering that metric over time. If the curve flattens off at some point, it means some users are sticking with your product over the long term. This means you have probably found product market fit for some market or audience.
Plotting a retention curve on users acquired at a specific date only shows part of the story. You can derive more PMF signals from plotting retention curves for different cohorts.
Cohorts are a subset of users grouped by shared characteristics. In the case of startups the most typical grouping is by users who were all acquired in the same day/week/month, the frequencies identified in the previous step.
This will help you not just see a snapshot of your user’s retention, but the evolution of it as your product changes, different bets and initiatives are set in motion to optimize its value.
A signal of PMF is that each subsequent cohort retention curve improves over time, as shown in the next figure:
Engagement measures how often an active user uses your product. If someone is an active user, just how active are they? As you might expect there are different benchmarks for different use cases.
Once again, engagement should be focused on that key metric that most represents value to your company.
Here is what you should do:
- Using user-level event log, look at the last 28 days of data
- Count number of active days for each user (for that one key metric/event)
- Compile the counts into a histogram
The main value that you get by plotting a histogram is the opportunity to identify Power Users, those that most of your team members should talk to. These users have invaluable insights that can’t be overstated. This will enrich your understanding of the qualitative market signals.
- Collect historical, user-level event log (best case scenario on triggering that north star metric/event)
- Calculate and visualize weekly or monthly (or other frequency) cohort user retention curves
- Calculate and visualize L28 DAU histogram (you may choose a different length of time)
3. Qualitative and Quantitative Signals: PMF User Survey
Yes, surveys have a bad reputation. Many consider them to be extremely unreliable and biased. Our view is that a well structured survey and, more importantly, a meticulous analysis of its results can be an extremely powerful way to check PMF signals.
Sean Ellis Survey
Ask existing users of your product how they would feel if they could no longer use it.
Ellis says that, if at least 40% of users answered that they would be “very disappointed” without your product,” you have achieved PMF.
The other six complementary questions are skipped from the analysis many times, and this is a big mistake. The answers to those questions help us acquire the WHY behind the main answer.
Keep it simple at the beginning
Treat this framework like any other “bet” in your startup: make it your own. Rather than copying this word-for-word, it could be a reduced version or one with specific product questions. Ultimately we encourage you to understand its value and use it as further evidence as you build a case that you have achieved PMF.
If you want to take this survey further, we highly recommend that you read Rahul Vohra’s (founder and CEO of Superhuman) article: “How Superhuman Built an Engine to Find Product/Market Fit“. His company complements the survey with segmentation efforts that create a fantastic tool for product-led growth.
- SurveyMonkey Template (I would remove the first question if you already have instrumentation in place and you already know how these users were acquired)
Make it your own
If we had to share a definition for PMF given the framework we’ve just shared we would say:
Product-Market Fit as an impermanent state of being in which qualitative and quantitative signals provide strong evidence that your product/service is something that a critical mass of customers wants and needs.
This framework is a mental model, approach, and suggested tools to use as a jumping off point. The outcome is to develop an informed opinion on where their business currently sits on the PMF continuum. Every startup should adapt them to its specific needs.
In TheVentureCity accelerator program, we recommend startups to continuously evaluate their position on the PMF continuum. If the PMF signals are relatively weak, we recommend intentionally constraining growth and spending to allow organic behavior to show the actual value your customers glean from the product. After spending time and effort iterating the product to strengthen the PMF signals, it will then be the right time to focus on growth.