Why You Shouldn’t Start a Business With an MVP
The MVP has a misuse problem.
In the startup ecosystem, Minimum Viable Product has become the default answer to almost every question about how to start. Not sure if your idea will work? Build an MVP. Not sure if people will pay? Build an MVP. Not sure if the problem is even real? Build an MVP.
For founders who have already confirmed their problem and their customer, an MVP is the right next step. For founders who haven’t — and most idea-stage founders haven’t — it’s the wrong question entirely. Building before you’ve validated the fundamentals doesn’t accelerate learning. It creates expensive noise.
What an MVP Actually Assumes
In the startup world, the MVP concept has a specific origin: the smallest thing you can build to test a hypothesis with real customers. The keyword is “test.” An MVP is a learning tool, and it only works when you already know what you’re testing and who you’re testing it for.
At the idea stage, most founders don’t have those two things settled. They have a general sense of the problem, a rough picture of who might have it, and a product concept they’ve been developing in their head. That’s not enough information to build from. It’s barely enough to start with.
An MVP built without a validated problem and a confirmed customer segment is not a minimum viable product. It’s a guess with a user interface.
What to Do Before You Build
The work that precedes an MVP — validating the problem, identifying the customer, testing the core assumptions — doesn’t require building anything. It requires talking to people, observing behavior, and being honest about what you’re finding.
The question to answer before you write a line of code or spend a dirham on development: does the problem you believe exists actually exist, for the specific people you think have it, in a way that makes them willing to change their behavior or spend money?
If you can answer yes with evidence — not instinct, evidence — you’re ready for an MVP. If you can’t, you’re not. That’s not a discouraging conclusion. It’s a useful one, because evidence is collectable without building anything at all.
Three Ways to Start Without Building
1. Problem Interviews
The highest-leverage thing an idea-stage founder can do is have 10 to 15 structured conversations with people they believe have the problem they’re solving. Not pitching the idea — listening. The goal is to understand whether the problem is real, how consequential it is, and what the person currently does about it.
A well-run problem interview answers three things: is the problem acknowledged as real? Is it frequent or high-stakes enough to warrant a solution? And is the person dissatisfied enough with existing options to consider switching? Those three answers determine whether you have a business premise worth building from.
2. A Single Landing Page
A one-page description of the problem you’re solving — not the product, the problem — with a way for interested people to express intent: an email sign-up, a waitlist, or a request for a demo. You’re not selling anything. You’re measuring whether the problem resonates enough for someone to take a low-commitment action.
This approach works particularly well for B2C concepts. A landing page in the right channels can generate signal within days, at near-zero cost.
3. A Manual Version
Some of the most effective early-stage validation involves delivering the solution by hand before automating it: offering the service manually, using existing tools, or personally performing what the product would eventually do. This tests whether the solution works and whether people will pay for it without committing to the build.
It also forces a level of customer proximity that early products often don’t achieve. Founders who have done the work manually tend to understand their customer’s problem in a way that makes the eventual product significantly better.
What Comes After
Once you’ve answered the core questions, an MVP makes sense. By that point, you’re not building to find out if the idea has merit. You’re building to test a specific product hypothesis with a customer you already understand. That’s a very different kind of build, and a much less risky one.
Mapping the assumptions clearly before you get there is where a Lean Canvas is useful; it gives you a one-page view of what you know, what you’re assuming, and where the riskiest gaps are.
The IDEA Program is structured around this sequence: Problem, then Customer, then Idea, then First Build. Each milestone is designed to answer the questions that make the next step less risky. 90% of founders who completed it launched a product in the market, because they were building from a validated premise, not a polished guess.
If you’re at the idea stage and not sure where to start, the IDEA Program gives the pre-build process a structure. You don’t need an MVP yet. You need a process.
