Conducting experiments for achieving product market fit


What is product-market fit, and why it is important?

Product-market fit is a term used to describe the position at which there is broad market acceptance for a new service, product, or a redesigned version of them. It’s not easy to explain the term comprehensively, but the good news is I have crafted a post about it. Please read it here.

Product-market fit is important because until that moment the company can’t evaluate whether or not what the business is launching solves an existing problem that a large enough market face (if there is enough group of people with the specific pain, which the new product or service is about to remove). Without clarity on this, the company could continue funding building value that is not commercially viable. What is even worse, still, the firm could burn cash by hiring talents or investing in marketing, sales, or other functions that will not generate a significant ROI.

Three main reasons to go with a lean approach in business.

a) Reduced costs. Improved quality.

Lean experimenting is focused on removing non-value added costs. These costs drive up the price to the customer resulting in a loss of sales, as well as affect the bottom line net profit since costs that cannot be passed on to the customer add to the bottom line. By focusing on reducing this waste in new product launch or product redesign, one will see dramatic improvements in the cost of introducing new products.

b) Waste removal.

Usually, we can find several areas of waste, inventory, waiting time or stalling the customer service, logistic, information delivery, quality of service or maintenance, overproduction, process performance, and creativity (brains horsepower). By eliminating waste, a business has the opportunity to remove time spent on unnecessary tasks in launching new products. By reducing waste and errors, the company will improve quality, which will reduce the lead time by not having to remake parts or redo expensive and time-consuming work.

c) Growth unlocked faster.

The lean approach is more than a cost-savings exercise. It is also about possible revenue generation, which I like to call “growth unlocked.” Money and time that is preserved and potential recourses that are freed up can be utilized to bring other revenue-generating opportunities, such as creating a new product or additional services or selling inventions and concepts. The improved performance to the customer, such as better quality or faster, more reliable delivery, maybe a premium on the market that is worth a higher price.

If you want to leverage a lean approach in your new product development process, let’s have a chat about it. Please register for the discovery call here.

Steps required to start lean experimentation in achieving product-market fit.

Four different steps build up the lean experimentation process. In most cases, the business should pass all steps to make sure the concept was validated comprehensively in achieving product-market fit:

  1. Validate the problem. Is this a problem worth solving? What is the evidence end-users care about the problem? How can we run metric-driven decisions about the problem?
  2. Validate the market. What is the opportunity size? There are few billions of beds in the world and several hundreds of millions of travelers, so AirBnB’s concept made sense. Does yours?
  3. Validate the productThe problem might exist, but does the product answer them? What is the evidence? What are the numbers which can use as the thesis support? 
  4. Validate willingness to pay.  Will people be willing to give something for the product (email, login, money, data, passport number, car plate number, or finger scan)? 

Going through and tracking all the four points, it’s hard without a tool. Fortunately, there is a tool for this exercise; it’s called the Validation Board.

The Validation Board is a one-page tool, which is very helpful in conducting a lean experiments. It’s a terrific way to test early-stage ideas, products, and services which don’t get traction yet. Here is the Validation Board template:


Please have a look at how you can work on the specific elements of Validation Board:

a) Customer hypothesis part

This part is about forming a profile of your target customer. For the customer hypothesis, a customer is someone who gives you something in return (currency). The currency could be one or few of the following:

  • customer’s time, 
  • information about customers or their activities, 
  • money, 
  • or anything else of value. 

b) Problem hypothesis part

This is your speculation about a problem that your customer (previously identified) faces, and you can solve. Let me give you an example. The Problem Hypothesis could be that first-time moms want a convenient way to discretely get advice related to mental health as hey faces hard situations because of the first kid ruling their lives now.

Having gone through this with your customers, you can generate the following attributes of the problem hypothesis: 

  • All moms have daily concerns that they’d like advice on including Attitude, Behavior, Skill, Planning, Handling activities, Time Management
  • Moms who live in big cities might experience other issues as well (hint: traffic)
  • Moms who live in smaller town might experience something different (hint: access to the internet)

c) Assumptions building

These are assumptions about the customer and problem hypothesis that require running experiments to check if they are vital or not. 

  • New moms can’t get right, continuous help from parents 
  • New moms know that they need help
  • New moms are happy to pay for help
  • If new moms get help, they are willing to apply the suggested change quickly
  • New moms need everyday assistance to build new habits

d) Solution hypothesis as the pre-stage to lean experimentation

When a hypothetical customer and problem is well defined, it comes naturally to describe a theoretical solution. New moms might appreciate a web portal with doctors on-demand, a social network where other moms can jump on the call and share the advice. It can also be a private coaching or local country club with the appropriate classes. All of those are hypothetical solutions that should answer boldly to the problems, and the solutions should be easily testable; otherwise, the business is not able to implement a lean-approach.

e) The riskiest assumption in conducting experiments

Of the assumptions generated, one is the riskiest, and if it fails your test, then the whole idea needs to be revisited – the pivot. In one of my projects, we assume that people take many pictures of clothes during the day, and they can’t describe them, so it is tough to find them using a text search. We assumed those people would appreciate it if we grab their pictures and recognize them automatically. The riskiest assumptions were that people forget to check the images which they took during a day. We assumed they just forgot to do it, or they didn’t have time to do it.

Seven types of lean experiments explained with prons and cons described.

a) Failures scanning

Investigate the biggest failures in the specific industry and explain the reasons. What can you learn from Google+ or Amazon FirePhone? This method is about collecting facts. This experiment helps you to define problems, solutions, and possible ways of developing the product or services. Consider contacting the people behind these unsuccessful ideas to get the true story and learnings from them. For instance, if you are about to launch the product in the telemedicine industry, you want to check why Telemedicine kiosk-maker HealthSpot shuts down, what can you learn from them. You could even call their customers and run a survey on what happened (especially the company secured $45 million funding)? 


  • cheap and fast
  • simple to delegate or buy


  • general assumptions can be discovered
  • competitors can get the same insight

Markets: B2B, B2C, P2P,

Great to test: Problem, Assumptions, Market

b) Comprehension test

Evaluate whether customers understand a marketing message explaining the value proposition to eliminate a possible false negative bias in later tests where customers indicate they do not want the value proposition when they do not understand it. As an example, you will want to run a comprehension test before launching a web site for new moms to make sure they understand your offering. They might be many wording issues, shortcuts you accidentally took, or other issues. Your target group should understand the value proposition the way it should be grasped. 


  • cheap and fast
  • simple to delegate or buy
  • check the understanding in face to face conversation (ability to read body language)


  • fragmented feedback
  • subjective feedback
  • require several iterations

Markets: B2B, B2C, P2P,

Great to test: Problem, Business Model (distribution channels, value proposition)

c) Wizard of Oz testing / Concierge

A Wizard of Oz test is a human (wizard) behind a computer screen manually completing the tasks. The most famous example is Zappos, which started without any inventory. Founders posted shoes offering on the website, and when the order came, somebody had to run out to the store and purchased them, pack them, ship them and handle returns if necessary. This is an experimental alternative to figure out if you are approaching a market fit.

Similarly, we could image recognition service and ask people for uploading pictures, which a customer service team tag and recognize manually. This way, a business can quickly learn what kinds of photos are uploaded, what is the level of difficulty, how people upload pictures, what phones they use, etc. Managers can also test end-users tolerance by making intentional mistakes or twists. Afterward, the firm can give users the option to speak with a real human and measure how often they are happy or not.


  • direct interactions and direct information exchange
  • real data and real business cases uncovered
  • high chance of discovering not participated scenarios


  • expensive because of the amount of time required for end to end the experiment
  • risky for well-known brands
  • labor-intensive

Markets: B2B, B2C, P2P,

Great to test: Solution, Business Model (value proposition, revenue, costs, relationship, customer segments)

d) Qualitative interviews and contextual inquiry

By interviewing users and challenging open-ended questions about their current process for obtaining a specific result (fitness program), we can learn about customers’ needs. Managers can start to notice patterns around where the challenges exist and what kind of solution could remove the pain. Instead of pitching customers a specific answer, it’s possible to check if they’d use something, the firm tries to probe at their current behaviors and anticipate how those could inform habits they’ll be more likely to adapt.


  • high-quality information
  • ability to explore deep motivation
  • speed of set up


  • measurability 
  • ability to find patterns
  • recruiting or finding participants (currency: time)

Markets: B2B, B2C, P2P,

Great to test: Solution, Assumptions, Business Model (all elements of Business Model Canvas), Problem and Market Size

e) Explainer video and currency

The landing page is a terrific way of getting data, tracking behavior, and more. It allows the business to offer a new product concept on the web, test the product’s value proposition, and evaluate if the product idea is on the right track. After making sure that users understand the value proposition, the landing page can be used as a stand-alone page where the firm displays and explain the value proposition and aim to collect the currency. The best way to describe the value proposition is to use an explainer video. It is a short video that explains what the product or service does/deliver; An explainer video is not specifically:

  • viral video or meme
  • tutorial or How-To
  • a company overview
  • or promo.


  • visual
  • many metrics are given
  • focused on the specific topic


  • significant time from starting to finishing the experiment
  • require skills of at least a few people

Markets: B2B, B2C, P2P,

Great to test: Solution, Assumptions, Problem

If you want to leverage a lean approach in your new product development process, let’s have a chat about it. Please register for the discovery call here.

f) Smoke testing

The term smoke test comes from the computer programming world. A smoke test is a process of performing sets of experiments involving the “important functionality of a component or system, to ascertain if crucial functions of the software work correctly.”

Does everything work as the documentation define? Smoke tests are mainly intended to be run quickly and repeatedly during the software development stages. The same is with product launch and product development, but a little bit different.

In the world of business, smoke testing is one of the most cost-effective and productive ways to test newly formed products or services — resulting in tangible learnings and actionable insights. Smoke tests help companies answer the fundamental question, “Will people pay (exchange currency) for the company product?” 

The landing page is an excellent way of running a smoke test. The webpage can then be used to easily track and quantify early demand for the new product in actual dollar-terms, which in turn can validate (or disprove) the hypothesis and determine whether the company has a reasonable business case (or not). Any information collected from these first few highly motivated users can also be subsequently used to target them as the first customers. As these early adopters have already essentially “pre-bought” the company’s product.


  • measure the ability to get currency
  • focused on performing specific tasks
  • end to end (covers whole value chain)
  • finds flaws which can be removed by the next version


  • require time for planning
  • should be run smoothly (no lags)

Markets: B2B, B2C, P2P,

Great to test: Business Model (all Business Model Canvas elements)

g) Single feature MVP

Most companies create a complex prototype of the idea and put it in front of end-users to see how they respond. What is complicated, it’s hard to understand, use, and comment. A better option is to avoid mega-concepts and unclear conclusions by testing a single-feature minimum viable product (MVP). Some of the most successful concepts like DropBox, Google Notes, or Twitter, started with a single, simple feature. A single feature MVP focuses on one aspect of a broader concept.

Why do so? To have the statistical guarantee that it’s that specific feature being validated (and not other ancillary components of the offering). Single feature MVP doesn’t necessarily mean the delivery of a unique feature only. What it means is that the company focus is on best presenting one feature and that the business should plan and build the MVP around this specific feature. For instance, if the firm is about to launch a mobile wallet app and focuses on implementing the NFC payment feature, but the same time to develop the feature, it’s necessary to register the credit card too. Otherwise, the payment feature won’t make sense at all. The point is, there are a central feature and a few supporting features. One single feature MVP is not about many, main features as it brings too much complexity, remove prioritization, and make experimentation fat.


  • simplicity
  • focus
  • many data are given
  • speed
  • ability to be improved in the next iteration


  • requires multiple tests
  • adjust assumptions
  • usually, it’s not enough to get the market-fit, though it might look like

Markets: B2B, B2C, P2P,

Great to test: Solution, Assumptions, Business Model (especially pricing, revenue, value proposition, channels)

For your convenience, please download the PDF if you want to explore all lean experiments (link)

Supercharge Your Business with Data Monetization and AI

Subscribe to the newsletter for weekly power-packed emails containing AI-powered productivity tips, AI products, and valuable insights on data analytics and monetization strategies, ensuring you stay ahead in the evolving world of Data and AI.

'Ai-Enhanced Sales' training is now available!