The Elusive Art and Science of Finding Product/Market Fit — Part III
Written by Harpal Singh; Last updated: December 11, 2022
8. Product/Market Fit Process
9. P/MF Dashboard
10. How to measure P/MF?
Acknowledgements
8. Product/Market Fit Process
The Part-I of this guide covered the startup lifecycle from Problem/Solution Fit to the Product/Market Fit and Scalability stage. In Part-II, we discussed P/MF Diagnosis steps to identify the root cause of not having P/MF. You should follow the P/MF Flywheel process described in this part when the root cause is either of these three:
You aren't solving a hair-on-fire problem for the customer.
There is no market or not big enough market for your solution.
You don't know who the real customer is and how to acquire them.
The path to Product/Market Fit isn't linear but there is a certain order for doing things like finding the market before building the product. Although a number of P/MF activities happen in parallel, we are looking to find answers first to things that matter most.
P/MF Flywheel
The P/MF Flywheel is a depiction of the P/MF process that encapsulates the discovery of four fits (product, channel, market and business model) together.
👉 The key aspect of the flywheel is to focus narrowly on a specific set of users to solve a hair-on-fire problem for them.
By building something they love rather than merely like, you can set out to find more users like them.
📌 It's better to build something that a small number of users love than a large number of users like. — Sam Altman
Before we delve into the what and how of each step of the P/MF Flywheel, please note:
The first three steps of the flywheel are about deeper problem discovery building upon the Problem/Solution Fit you already have. You don't need to build an MVP or add features to your product for the discovery part.
You need to have some idea of the market size before you begin this progress. The validation and analysis of the Total Addressable Market (TAM) comes after you have figured out which customers are you solving what problems for.
8a. Define Segmentation
The first step of the P/MF Flywheel is to divide the existing users of your product into segments. You can segment users by:
Behaviour (intent, usage, engagement) e.g. content creators using video and slides.
Demography (age, income, social status) e.g. high-school teachers.
Industry (sector) e.g. insurance providers.
Geography (location, country, language) e.g. French-speaking Canadians.
👉 The best way to define a segment for your product is to figure out a group of users likely to want the same thing who can be found through the same channel.
At Automata, we debated between segmenting by use case/behaviour i.e. what customers were using the robot for e.g. product testing, sorting, inspection etc. and industry e.g. SMEs, eCommerce warehouses, restaurants, car manufacturers etc. The distribution channel and approach for finding these groups would be very different.
Not enough users?
If you don't have enough users for your B2C product or only a handful of pilot customers for your B2B product, you can form a hypothesis of your customer segments based on your understanding of the market and start from there.
Not enough data?
You may have enough users but not have all the information about them to clearly define segments. This gap can be filled through a customer survey or using Predictive Analytics tools that build behavioural user profiles from product usage.
🎬 VideoCo Segments
Segmentation by Behaviour — Small Tech Teams, Content Creators and Educators/Teachers.
8b. Shortlist Lead Segment
After defining a handful of segments, the next step is to shortlist a lead segment that's most likely to have a hair-on-fire problem your product can solve.
👉 The segment should be as narrow as possible otherwise you will struggle to find a coherent group of users who need your product urgently and ready to pay for it.
During the Problem/Solution Fit stage, the customer segment is broad as you are trying to put your product in front of various types of users to see where it sticks the most. The objective is to find a set of users who show some signs of need and excitement for your product.
For VideoCo example, in the Problem/Solution Fit stage, your users could be anyone using video for calls like teachers, friends, lawyers or professionals from any industry. It doesn't mean they are the right users long-term for your product.
Achieving the Problem/Solution Fit can provide you enough data and insights to form an informed hypothesis about who the right customer could be. To find the Product/Market Fit, you must focus narrowly on a specific lead segment to find the right customer for your product.
🎬 VideoCo Lead Segment
Small technology teams who use video heavily to collaborate amongst themselves.
From past projects, I've seen 2-3 segments accounting for the majority of product usage with a long-tail of totally random users making up the rest. The lead segment is generally one of these 2-3 segments.
No supporting data?
If you don't have enough data to create segments in the first place, it's unlikely you have data to pick a lead segment. In that case, you can pick a segment from the customer surveys and interviews conducted during the P/MF Diagnosis. Make sure to ask demographic questions in the survey for segmentation.
No matter how you pick your lead segment, you need a way to reach out to prospects and customers from the shortlisted segment to engage them throughout the P/MF process.
8c. Hair-on-fire problem and Value Proposition
This is the longest and hardest step of the flywheel and also where teams give up on the P/MF process after hardships. The teams then go back to making something nice that people mildly want (but don't need) resulting in a mediocre or a failed product.
Problem-centric Interviews
The discovery of hair-on-fire problem starts with interviewing prospects (non-customers) who have never heard of or used your product and customers who are using your product and/or paying for it. The ratio of prospects to customers is 80/20.
👉 You are interviewing prospects and customers from the narrowly shortlisted lead customer segment to discover a narrow painful hair-on-fire problem they want resolved in the context of your product.
Having hour-long one-to-one interviews with eight prospects and two customers from the lead segment will be enough to discover if they have the type of problems you are after.
These interviews are NOT about your product or its features at this point. Where possible, the whole hour should focus on the prospect and customer's world without discussing your product. You can learn how to conduct effective discovery interviews by asking the right questions from The Mom Test book by Rob Fitzpatrick.
Problem — Impact — Environment
👉 The hair-on-fire problem is determined by the impact of living with those problems in an existing environment.
When either of the variable changes, Impact or Environment, the Intensity i.e. level of pain of the problem changes. The table below shows an example of how a problem intensity goes from Normal to Critical with a change in the Environment (Covid-19).
| Environment | Hair-on-fire Problem | Impact | Intensity |
|---|---|---|---|
| Pre Covid-19 | The fast-pace startup life takes a toll on mental health for some individuals occasionally. | Sick leaves delay projects and work quality is impacted. | Normal — Startups can get away by not focusing on their employees mental health. |
| During Covid-19 | Covid-19 increased the loneliness and mental health challenges for many individuals with no work/life boundaries. | Severe dip in motivation across the board after the initial remote work excitement. Multiple project delays and major loss in revenue during Covid-19. | Critical — Startups must take care of their employees mental health to avoid high churn and loss of reputation. |
The change in environment doesn't have to be as drastic as Covid-19. It could be:
Your B2B customers facing higher demands from their customers like faster delivery times,
Your user's change in circumstance like a job change or getting married, or
Change in regulations like GDPR.
Such changes can affect the importance of the problem for your customers. The intensity of pain determines your customer's willingness and urgency to spend money to relieve that pain.
👉 It's tempting to settle on a problem that's neither too important nor too intense. It won't cause people to switch the product, change their behaviour and pay for it.
You will eventually solve a massive problem for a variety of segments but the journey to that point would take solving narrow problems really well for a narrow set of audiences.
Hair-on-fire Problem Won't Reveal Itself
Customer interviews rarely reveal the hair-on-fire problems directly. Why? Because customers don't think that way. Have you heard the infamous example “The customer wants to hang a picture, not put a ¾ inch hole in the wall”?
Customers think in the form of goals they are trying to accomplish or jobs they are trying to do. They either hire or pay for your product to get that job done. You can use the following two methods to dig deeper during the interviews and post-interview analysis to discover the right problem:
1. Jobs To Be Done (JTBD)
Jobs-to-be-done is an effective framework to understand the current and future state of customers, and the change they need to make to achieve that future state. The milkshake example by Clayton Christensen explains this concept well. A common belief is new products and services win in the marketplace when they help customers get a job done better, faster and/or cheaper.
2. Problem Reframing
Design Thinking provides us many ways to reframe problems by breaking it down into smaller problems like using Five Why's methodology, conducting a frame-storming session (aka brainstorming for framing) or zooming out to look at the big picture. Problem reframing is about looking at the problem from a different point of view.
🎬 VideoCo Problem Reframing Example
Problem — It takes time for new users to download VideoCo desktop app to get started.
Reframed Problem — It's embarrassing and stressful for users to complete the onboarding process at beginning of their first VideoCo call. This makes them less likely to use our product again.
In this example, the problem is the same i.e. getting started. The reframed problem highlights the motivations and emotions, which is a make or break situation in this case. In the initial problem, you might try to solve the problem by optimising the download and installation speed whereas, in the reframed problem, you would need to look at the whole journey to address it.
👉 Often you will find technical problems are actually process or human problems.
Value Proposition
The discovery of the hair-on-fire problem and a deep understanding of your customer's problem space will help you in crafting a strong value proposition. This proposition can be translated to brand messaging and solutions.
👉 The objective here is to build prototypes, MVP or product features to find a solution that addresses the hair-on-fire problem for the shortlisted lead customer segment.
You can follow the Build-Measure-Learn Lean Startup process to iterate your solution in biweekly cycles or by using weekly design sprints.
You should build only what's necessary to validate the solution and nothing more. For software startups, this could be low-fidelity prototypes and for hardware startups, this could be a Concierge or Wizard of Oz MVP. B2B enterprise startups can use piecemeal MVP by piecing together existing services to deliver and validate the value proposition.
The Build-Measure-Learn loop is often mistaken as the holy grail for finding Product/Market Fit by constantly iterating ideas with a bunch of users. It feels right but is inadequate without other elements like distribution channel, business model, team, positioning etc.
10x Solution
Discovering the hair-on-fire problem is merely a milestone in the long and arduous journey to P/MF. Finding a great solution and executing it well is equally important.
👉 You have to aim for a solution that's at least ten times (10x) better than the closest alternative for users to change their behaviour and get over the financial and emotional cost of switching.
Google Photos is an excellent example of a 10x solution. It solves the hair-on-fire problem for an average non-tech person to store, backup, share and search photos with ease. Hundreds of photo apps have existed before Google Photos but none of them solved the problem so well. The 10x solution made it the third-most used Google product after Search and Gmail hitting a billion users in just four years of its launch.
8d. Validate Market Size
It's likely you already have an idea about the market size when you built your initial MVP to find the Problem/Solution Fit. As you go through the process of finding and solving the hair-on-fire problem for a segment, you could end up targeting a completely different market or a different set of users in the selected market. This warrants a need to validate the market size and make sure there are enough people with a need for your product. This often translates to building or revising your business model, unit economics and financial projections.
👉 A small market with a great product isn't going to help you build a highly scalable business. You must find a large enough market for your product.
8e. Validate Acquisition Channel
👉 Validating a channel means you are able to acquire new customers in a repeatable way from the same channel who have the same hair-on-fire problem you are solving.
Startups are frequently opting to focus on Product-Led Growth (PLG) after they get the key channel working to break away from the long-term dependency on a channel.
Startups fail even after raising mega-rounds of investment and building a great product because they never find the market and channel to acquire customers. “Build it and they will come” is a fallacy I learnt way too late in my career.
👉 Products are made for channels rather than the other way around.
For walled-garden channels like SEO, Social Media, Email, Events and App Store, the owners of the channel i.e. Google, Facebook, Outlook, Event organiser and Apple App Store respectively define rules of the game. Your product must fit to suit these channels.
For relationship-based channels like strategic partnerships and direct sales, your product must meet the partnership criteria. It’s not uncommon for B2B startups to build features to get a partner or customer on board despite knowing that the feature won’t be used by the end-users.
One Channel at a Time
Peter Thiel in ‘Zero to One’ says most startups get zero distribution channels to work and strongly advises against the kitchen-sink approach of trying various channels at once. He further suggests if you can get just one distribution channel to work, you have a great business. I couldn't agree more.
I haven't come across a single startup that follows this as it's so counter-intuitive. It's quite uncomfortable for marketing teams to bet all their efforts to validate a single acquisition channel before moving on to the next one. Testing a channel could take months e.g. if you decide on SEO as your key acquisition channel, it will take months to create high-quality content with a sound SEO strategy to attract the right audience.
This doesn't mean you abandon your entire channel mix as maintaining some of them is essential for building and growing the brand.
👉 The key is to spend the majority of your marketing efforts and resources to find an acquisition channel that works instead of spreading yourself thin across a huge channel mix.
You can use paid media e.g. Facebook ads to acquire some users but it's highly unlikely to become a sustainable and repeatable source of your target audience to achieve P/MF.
You can't growth-hack your way to Product/Market Fit
Growth hacking experiments like ebooks, referrals, discounts, virtual queues etc. can give you short bursts of users and even generate huge upfront interest. However, focusing on growth hacks and tactics at this stage can create a false indication of progress towards Product/Market Fit.
Growth hacks can be helpful until you are experimenting to shortlist a channel to focus your efforts on for customer acquisition. Without P/MF, focusing on growth hacks is going to take you away from your goal.
The Leaky Flywheel
The idea of the flywheel is to discover a repeatable process without any leaks. You would need to start the process again when a leak happens. Let's look at two examples.
No hair-on-fire problem discovered
You could spend a few months defining segmentation, shortlisting lead segment, conducting interviews, gathering data and still not come across a hair-on-fire problem for the existing problem/solution space. It could be the shortlisted segment doesn't really have an acute problem, or you might not be digging deep enough. There is no point in continuing to develop a solution if you haven't discovered the hair-on-fire problem. In that case, you should pick a different segment and start again.
👉 The process of not finding a problem can be incredibly frustrating and demotivating. Whatever the case, don't settle for mediocrity.
It will become an uphill battle to build a great business if you settle on a nice-to-have product.
Market isn't big enough
It's possible to find a problem and build a solution for a set of users only to discover there isn't a big enough market for your product. In most cases, you could tweak the positioning, customer segment or value proposition to get on the right path without making considerable changes to your product.
9. P/MF Dashboard
You would need to go through the P/MF Flywheel a number of times to test various segments, problems and channels. You can track this process in a P/MF Dashboard as shown in the diagram below:
The full dashboard after going through multiple cycles would look something like this:
The dashboard shows the validation of multiple problem hypothesis (P1, P2, P3) for the shortlisted lead customer segment (A). These could be the problems discovered through customer discovery interviews or problem hypothesis formed for validation. As the hypothesis failed for segment (A), we pick a new lead segment (B) to go through the new problem (P4) validation finally discovering the hair-on-fire problem (HoF P).
After discovering the HoF problem, we build and iterate value proposition (MVP1 & MVP2) by putting it in front of an audience targeted via the Channel (C1).
Here is an example P/MF Dashboard for our hypothetical video conferencing startup, 🎬 VideoCo.
| Lead Customer Segment | Problem Hypothesis | Existing Solution | Hypothesis Validation |
|---|---|---|---|
| A — Small tech teams | P1 — People want to stay connected all the time and not feel lonely when working from home. | VideoCo Always On video conference feature fulfils this need. | Failed. Most prospects didn't like the idea of Always On feature. Zoom fatigue phenomena made it worse. |
| A — Small tech teams | P2 — Video calls take toll on mental health for people especially in different timezones. | VideoCo floating window functionality keeps video in the back and brings up only when desired. Slight product tweak required for hypothesis validation. | Failed. Not an acute problem. The solution doesn't address the mental health problem fully, although most agreed it helps in easing the pain and improving their situation. |
| A — Small tech teams | P3 — People struggle with staying productive at home. | Tweak the product to match people with strangers who always stay mute to keep accountability. | Failed. Important but not an intense problem. People jumped on this idea as a novel experience and loved it but don't see this as a long-term solution. |
| B — Content creators | P4 — It's incredibly frustrating for content creators to stitch a video and presentation together. They have to pay freelancers and it takes a lot of time.
The experience sucks for the end users too with the constant switching between video and presentation. |
Product does't have all the functionality yet. | Hair-on-fire problem discovered. |
All the problems discovered from customer discovery sessions can be mapped on a 2x2 matrix of Importance vs Intensity. The hair-on-fire problem is the one on top right i.e. most important with high intensity. Example matrix for VideoCo shown below:
| Hair-on-fire problem | MVP (Build-Measure-Learn) | Channel |
|---|---|---|
| Explainer video creation takes a lot of time and money to produce average results | MVP 1 — Superimpose video thumbnail on top of slide deck for hypothesis validation and allow content creators to upload and share video with ease. | C1 — Virality: Content creators share their content and a small proportion of viewers sign up for VideoCo as creators. |
| Same as above | MVP 2 — Allow controls to switch between presentation, full screen video and video thumbnail. | C1 — Early signs of success. Optimise sign up flow for viewers. |
Timelines
👉 It takes a couple of years to find the Product/Market Fit for startups and a few months to discover the hair-on-fire problem and the channel to acquire customers with that problem.
The timeline is directly proportional to the number of times you have to go through the P/MF flywheel. The best way to optimise the overall time is to:
a) Focus on P/MF on an ongoing basis rather than random ad-hoc efforts, and
b) Have a dedicated team like P/MF Council constantly driving it forward.
P/MF Council Meetups
The P/MF Council has representation from Product, Marketing, Design/Research, Sales (for B2B) and Founding Team, as discussed in Part II. The council meets on a weekly basis to discuss the P/MF Flywheel progress represented through P/MF Dashboard. The high-level responsibility of council members are:
Product — To support discovery, lead the P/MF council, prioritise and manage the P/MF process.
Designer / Researcher — To bring customer insights and profiles to the council through customer discovery interviews.
Marketing — To find and validate the acquisition channel based on customer insights.
Sales — To utiltise the insights for identifying leads and generating sales.
Co-founder — To make sure business model works and riskiest assumptions are tackled upfront.
Where possible, head of engineering and all council members should be directly involved in customer discovery interviews. The council should update the entire company on the P/MF progress on a monthly basis.
👉 The insights from the P/MF Flywheel process should inform your value proposition, solution, product roadmap and ideal customer profile after discovering the hair-on-fire problem and the right customer.
10. Measuring Product/ Market Fit
There are two core principles behind my recommendation for measuring Product/Market Fit:
Rely on what people do, rather than what people say i.e. action not intention.
P/MF can't be reduced to a single metric.
How NOT to measure the Product/Market Fit?
Before we delve into measuring P/MF, let's look at how NOT to measure it based on the principles described above.
Surveys and Interviews — I strongly advise against surveys as a single metric to measure P/MF on an ongoing basis. Surveys are tricky. People say things they don't mean. Relying on people to predict how they will behave in the future is a recipe for disaster. Surveys and interviews are susceptible to interpretation, environment and emotions, and can be easily manipulated to get the answers you want.
40% Disappointment Survey — This can be helpful in the diagnosis stage as an indicator but not enough to prove P/MF especially when your user base is too small or not properly segmented.
Net Promoter Score (NPS) — NPS measures intention, not behaviour. For any given business, you will find customers who actively promote and criticise the business at the same time. I’ve personally witnessed failed businesses with excellent NPS scores.
Revenue — Millions in revenue doesn't mean much if everything else isn't working. You can spend a ton of money on growth and customer acquisition to generate desired revenue at a massive loss.
Valuation — It's easy to look at extremely high valuations like $48B for WeWork and conclude it's a great business. Relying on investors valuation for P/MF is a mistake.
How to measure the Product/Market Fit?
You can measure P/MF through a working P/MF Flywheel and a KPI for User Retention, Growth and Business Model each.
A Working P/MF Flywheel +
Retention, Growth & Business Model KPIs👉 A working flywheel means you have figured out a way to solve a specific hair-on-fire problem for a narrow set of audience in a large market and found a repeatable way to acquire and retain those customers.
There is no hard figure for the size of this segment to make the flywheel work. As a rule of thumb, I suggest at least 50K users for a B2C consumer startup and at least 50 paying customers for a B2B enterprise startup. These users/customers must be from the same audience segment with similar needs and way of finding you, not the total number of users/customers of your product.
Here is an example of a completed flywheel for our hypothetical startup, VideoCo:
We established that Product/Market fit is not a discrete big bang event in earlier parts of this guide.
👉 A working P/MF flywheel indicates you have moved from the Problem/Solution Fit to the Product/Market Fit zone, so to speak, but there is still plenty of work to be done to ensure you have solid Product/Market Fit.
Retention — Growth — Business Model
The KPIs for retention, growth and business model will vary by your product and business model. All these KPIs should be measured for the defined segment (not all users) to have quality P/MF data and to avoid inflated numbers and false positives.
Retention
Retention is the single biggest quantitative indicator of P/MF. It is measured by the number of active users and encapsulates product usage and engagement by definition. Retention is calculated on a weekly or monthly basis.
👉 A product with 60% or above retention over 3-6 months has a strong indication of P/MF.
You can expect a fairly deep drop in retention within the first weeks of acquiring new users. The more specific the segment, the higher the retention. However, the retention curve must flatten overtime for the shortlisted customer segment, as shown below, to ensure stickiness.
Retention Quality
You have to be honest with yourself when calculating retention by making sure the right customer segment is included and retained users are actually using the product.
It's fairly common for companies to keep paying for a product or service sometimes for years without getting the value (false positive). This is true for some individuals as well where they pay for SaaS products every month and rarely use it to its potential. SaaS industry thrives on this loophole.
📕 Retention Quality Case Study — A large business I worked for paid £60K annually for a SaaS marketing campaigns tool for years without using it because no one in the marketing or tech team could free up their time to commit to it. Then we paid £15K on top of it to have a dedicated account manager who used the product for basic campaigns on our behalf. We kept paying for that product and service for years as the switching cost and time commitment was too high. There was an initial sunk cost of implementing the product in the first place too. Yet, from analytics, ours would have been a great retention success story for that SaaS business.
Growth
👉 The Growth KPI is the key traction metric specific to your product and business. It could be new users, monthly active users, paid customers, subscriptions etc.
The ‘new users’ growth metric means acquiring users from the same audience segment through a specific channel shortlisted and validated as part of the P/MF flywheel. It doesn't mean acquiring users from any channel by any means using growth tactics or otherwise.
What does this growth translate to?
For SEO as an acquisition channel, it means more content production and increased traffic every month.
For Events and Conferences channel, it means an increase in quality and quantity of leads and events.
For API channel, it means an increase in integrations and API customers.
For Strategic Partnerships channel, it means more partnerships and consistent leads from existing partners.
For Direct Sales, it means more sales generated every month by growing sales team.
You get the idea. Growth has to come from the same channel for the flywheel to work and to achieve P/MF. The only exception to this is acquiring users via Word of Mouth (WoM), in which case you can add those numbers to the channel, if you are able to calculate WoM.
👉 3-5% compound monthly growth is an excellent indication of P/MF.
The compound monthly increase is exponential. Here's an illustration for ‘paid subscriptions’ growth metric. If your subscriptions are growing at 5% MoM with a MRR (Monthly Recurring Revenue) of £100K, your projected monthly and yearly revenue will be £240K and £2.9M respectively in just 18 months.
Acquisition and Retention Loop
The best growth channel is the one native to your product represented through Acquisition and Retention loop, also referred to as a viral loop. Loops are more effective than top-down funnels as the result of an action increases the overall value by bringing new users or retaining existing ones. They are great to build competitive moats too.
Here is an example loop for our hypothetical startup, 🎬 VideoCo, where the action of creating video content brings new users who watch that content:
Business Model
The P/MF KPI for business model depends on how you generate revenue. Some of the common KPIs include LTV/CAC (Lifetime Value/Customer Acquisition Cost), Unit Economics, Profitability, Average Order Value, MRR/ARR (Monthly/Annual Run Rate) and Churn Rate etc.
👉 LTV/CAC works for the majority of startups, and if your LTV is 3-5x CAC, you are heading towards building a great business.
In a nutshell, you can use these five questions to measure P/MF covering P/MF Flywheel, retention, growth and business model KPIs:
📌 Questions to Measure P/MF
1. Are you solving a hair-on-fire problem for a clearly defined user segment in a large market?
2. Are you acquiring new users in a repeatable way through a key acquisition channel?
3. Are you retaining more than half of your new and existing users?
4. Are you growing at 3-5% monthly compound rate?
5. Are you profitable? Or Is your business model working i.e. LTV > CAC?
If the answer to any of the above is no, you don’t have P/MF.
P/MF KPIs
The Retention, Growth and Business Model KPIs must work together to indicate a strong Product/Market Fit. Falter on one and you are looking at a smokescreen i.e.
Even if you have a strong 90% retention of existing customers but your CAC is extremely high and LTV is equal or less than CAC, you don't have P/MF.
Even if you are growing virally through Word of Mouth at 10% MoM but not retaining a solid 60% of those users, you don't have P/MF.
Even if you are generating a profit by keeping costs low but not adding enough new users and/or retaining a good portion of existing users, you don't have P/MF.
This is the reason we can't reduce P/MF to a single metric like revenue or retention.
Strengthening P/MF
Once you get the P/MF Flywheel working and the three KPIs are going in the right direction, you have built a product that's resonating with the early majority (compared to early adopters in the Problem/Solution Fit stage). There would be a handful of core features driving retention and growth among the early majority.
👉 Just when you thought you hit gold, it's also the most delicate time when you can easily lose P/MF.
You can easily lose P/MF due to always evolving and changing market forces, competitors and the importance and intensity of the problem you are solving for your customers. Covid-19 is an extreme, albeit good, example to highlight how many businesses were sadly rendered useless overnight.
Here are some actions you can take to strengthen P/MF and avoid falling back:
a. Multiple Segments — Make the P/MF Flywheel work for another segment by repeating the process. You might be able to acquire this new audience from the same acquisition channel. Ideally, you need to have 2-3 segments locked in with consistent growth to ensure a strong P/MF.
b. Adjacent problems — We focused on solving the hair-on-fire problem for a specific segment to make the P/MF Flywheel work. You can now move on to solving other problems you discovered during the process for an existing segment, or a hair-on-fire problem for the new segment.
P/MF and Scalability
The simple one-line advice is don't scale before you have Product/Market Fit. It's tempting to raise boatloads of cash, hire people in troves and throw money at marketing at the first sign of traction. You can use these two signals to evaluate whether the timing is right:
a) You are delivering repeatable value to customer and business highlighted through retention, growth and business model KPIs.
b) A good portion of your users are coming through Word of Mouth and referrals.
Conclusion
I'll leave you with this simple and profound quote by Sam Altman:
📌 All companies that grow really big do so in only one way — people recommend the product or service to other people.
Acknowledgements 🙇♂️
I’m incredibly thankful to Lisa deBettencourt, Oren Greenberg, Nick Pattinson and Rob Fitzpatrick for providing extensive feedback and inputs to this guide. Their feedback improved the quality of this guide drastically.
Many thanks to Jason Mesut, Tammer Mahdy, Aggelos Mouzakitis and Manfredi Sassoli de Bianchi for reading the draft and providing their comments as well.
Last but not the least, thanks to the startup founders, product and technology leaders who gave their time on virtual calls over the last couple of months to discuss their views on this topic. It would be too many names to list here. Rest assured, I’ve deep gratitude for everyone’s contributions.
📜
Key Takeaways (Part III)
It takes a couple of years to find the Product/Market Fit.
Focus narrowly on a specific set of users to solve a hair-on-fire problem for them.
P/MF can't be reduced to a single metric like NPS. Measure actions not intentions.
Use P/MF Dashboard to track and test the problem, solution and channel hypothesis.
A working P/MF Flywheel backed by Retention, Growth & Business Model KPIs is a measure of P/MF.