If you are just starting out with an idea, you will hear the same advice again and again: build an MVP. Build it lean. Build an MSP—a minimum sellable product. I give clients that advice too.
But an MVP is not a shortcut to success. Plenty of teams build one, launch it, and still end up with very few users. When I suggest looking at market response, analytics, and competitors, the usual reply is: “We already tried that. It didn’t work.”
The problem is rarely the idea of an MVP itself. More often, it is how teams use it.
What an MVP Is Actually For
Lean Startup defines a minimum viable product as the version of a new product that helps a team gain the most validated learning about customers with the least effort.
The important phrase is validated learning. An MVP is not simply the smallest product you can ship. It is an experiment designed to answer a specific business question: will people use this, pay for it, or come back for more?
That experiment can take many forms. It might be a landing page, a manual service presented as automated, or a basic prototype. What matters is that real customers can interact with it and that you can observe what they actually do—not just what they say they would do.
Source: Agile Alliance
Where Teams Go Wrong
In my work with entrepreneurs, I often see the same pattern: they build something, then add more features, then build even more on top of that. But that was never the point of the MVP.
An MVP should generate evidence. The data from that first version should tell you what to test next. Too often, teams skip that step and treat the MVP as a stripped-down final product rather than a learning tool.
Start by Validating the Core Idea
One client I worked with wanted to create a unique collection of useful content. The first thing to validate was not the technology—it was whether people actually valued the collection.
She did not need to build a full platform. She could have started with:
- A simple WordPress blog with the content organized by category
- A downloadable PDF containing the curated collection
- A small ad campaign to attract the right audience
- A basic funnel to measure sign-ups, downloads, and feedback
That would have answered the core questions: Are people interested? Do they find the content useful? Would they want more collections like it?
Only after proving that demand should she move to the next test: whether people wanted a personalized experience.
Use the Results to Decide What Comes Next
Validation is not about getting a perfect result on the first attempt. If four out of ten people find the content useful, that may be enough signal to keep exploring.
And if people do not respond well, that does not automatically mean you should shut the business down. It may mean you need to pivot—change the audience, the offer, the format, or the business model. In many cases, that pivot is what leads to a viable product.
The larger vision for this client included personalized content collections, certifications, skills training, and job opportunities. Each of those ideas could be tested separately. The mistake would have been building everything at once and then blaming marketing when customers did not arrive.
As the Agile Alliance puts it, teams often mistake an MVP for “the smallest amount of functionality they can deliver,” without making sure it is enough to learn whether the product is commercially viable.
A Practical Example
We have also shipped imperfect MVPs in under a week. The important part was not that they were polished; it was that we learned from them. In one case, those lessons helped a business move from a 10% decline in users to 155% quarter-over-quarter growth.
In another case, a client already had a fully developed product, but conversion was unexpectedly low. Our initial hypothesis was that one or both of these issues were getting in the way:
- The price was too high compared with competing offers.
- The application was difficult to use or too buggy.
Research showed that both were true. However, seven out of ten users were still able to complete the application process up to the payment stage. That told us the biggest issue was not the bugs—it was the offer.
Competitors were providing less value for almost nothing, while our client’s offer was priced too high for the market. Once we confirmed that hypothesis, we redesigned the offering, user journey, and pricing.
We launched the revised app within a week. Two days later, conversion had risen from 5% to 45%. The app was still not perfect, but we had solved the most important problem before spending time and money on a fully polished product. Improving the experience later pushed conversion and user growth even further.
What to Do Differently
A useful MVP requires research, strategy, and a clear plan for what you want to learn. Launching repeatedly without analysing the results will not get you very far.
Before building, define the assumption you want to test. After launching, study the response. Then make the next move based on evidence—not instinct, not feature pressure, and not the urge to keep building.
That is how an MVP becomes useful: not as a smaller product, but as a faster way to learn what the market actually wants.
