How to validate a business idea before you build it

Having an idea and knowing that it is worth building are two different things. The first can happen in an afternoon. The second usually takes evidence.

That is what idea validation is really for. It is not a ceremony you complete before starting a company, and it is not a way to prove that your idea is good. It is a way to discover which parts of what you believe are actually true. The earlier you do that, the cheaper it is to change your mind.

Start with the problem, not the solution

People often begin validation by describing the product they want to build. They ask whether someone would use an app that does this, whether a feature sounds useful, or whether a certain service seems interesting. The problem is that people are generally very generous with hypothetical ideas. Saying “that sounds useful” costs them nothing.

A stronger starting point is the problem underneath the idea.

Who experiences it? How often does it happen? What do they do today when it occurs? What is frustrating, expensive, slow or awkward about the current solution?

If the underlying problem is weak, improving the proposed solution will not make the opportunity stronger.

Validation is not asking people whether they like your idea

One of the easiest mistakes is to confuse encouragement with evidence. Friends may tell you the idea sounds great. Potential users may say they would probably try it. Someone may even tell you they would pay for it.

Those responses can feel reassuring, but they are predictions about future behavior. Validation becomes more useful when you look for what people already do.

Do they spend money trying to solve the problem? Have they built a workaround? Do they repeatedly complain about it? Are they actively searching for an alternative? Have they invested meaningful time in solving it themselves?

Behavior is usually stronger evidence than enthusiasm.

Write down what must be true

Every business idea contains assumptions. You may believe that a particular group has a problem, that the problem is important enough to act on, that your proposed approach is useful, that people will change their current behavior, or that they will pay enough for the solution to become sustainable.

Write those assumptions down before trying to validate the idea. Then ask a simple question:

Which assumption would hurt the idea most if it turned out to be wrong?

Start there.

Testing the easiest assumption can make you feel productive. Testing the most dangerous assumption can save you months.

Look for evidence in increasing levels of commitment

Not all validation signals are equal. A person saying “interesting” is a weak signal. Someone agreeing to another conversation is stronger. Someone giving you access to their workflow, introducing you to a colleague, joining a waiting list, testing a rough prototype, or paying for something is stronger still.

The useful progression is from words toward behavior.

You do not always need payment as the first test. The type of commitment should match the stage of the idea. But there should eventually be some cost to the person saying yes: time, effort, reputation, access, money or a change in behavior. Without that, it is difficult to distinguish genuine demand from politeness.

Talk to people without selling them the answer

Good validation conversations are surprisingly difficult because the person with the idea already knows what they hope to hear. Try not to lead the conversation toward your solution. Ask about what already happens.

Instead of asking:

“Would you use a tool that automatically does this?”

ask:

“What happened the last time you had this problem?”

Instead of:

“Would this feature be useful?”

ask:

“How do you handle this today?”

Instead of:

“Would you pay for this?”

ask:

“What have you already spent time or money on to deal with it?”

Specific stories about past behavior are usually more useful than predictions about an imaginary future.

Search for contradiction, not agreement

Validation should be capable of changing the idea. If every conversation is interpreted as proof that you are right, you are not really validating anything.

Look deliberately for people who see the problem differently. Ask what would make the idea unnecessary. Find out who does not experience the problem and why. Pay attention when someone’s behavior contradicts what they say.

A useful validation process does not only make an idea stronger. Sometimes it makes the idea smaller, stranger, more specific, or completely different. That is progress too.

The purpose of validation is not to protect an idea. It is to expose it to enough reality that it can change.

The right person may not look like your target user

There is another kind of validation that is easy to miss. Sometimes the useful perspective does not come from someone who looks exactly like the customer you imagined.

A person in another industry may have encountered the same underlying problem. A researcher may understand a mechanism behind it. Someone building a completely different product may have discovered an approach that changes how you think about your own.

This does not replace talking to real users. It complements it. Ideas often become more interesting when they encounter perspectives that were not part of the original plan.

Validate the next decision, not the entire future

You cannot validate a company before it exists. You cannot prove in advance that a product will succeed, that a market will grow, or that people will behave exactly as you expect once the product is real. What you can do is reduce uncertainty around the next important decision.

Is this problem real enough to keep exploring?

Is this audience the right one?

Is this assumption worth prototyping?

Is someone willing to commit more than a compliment?

Should you build the next version, change direction, or stop?

Validation works best when it helps you make the next decision with better evidence than you had before.

Where Overlap fits

Some ideas are easy to validate publicly. You can post them, ask a community, or contact people you already know. Others are less ready for an audience. They may be unfinished, difficult to explain, commercially sensitive, or simply early enough that you do not yet know who would be useful to speak with.

Overlap starts with the thought itself. You leave an idea, question or problem privately as a Drop. Other people cannot browse it. Overlap looks for meaningful intersections with other people's thinking, including connections that may not be obvious from job titles, industries or existing networks.

If something appears worth exploring, both people decide whether they want to connect and begin a conversation.

It is not a replacement for customer research, experiments or real-world evidence. It is another way an early idea can encounter a perspective it did not know to look for.