A startup idea can feel convincing long before there is any evidence that it will work. You notice a problem, imagine a solution, and immediately start thinking about the product, the name, the website, and everything the business could eventually become.

That excitement is useful, but it can also make founders skip one of the most important stages of building a company: validation.

Validation means finding evidence that a real group of people has the problem you want to solve and cares enough about solving it to take action. It happens before you invest heavily in development, hiring, marketing, or infrastructure.

You do not need absolute certainty before building something. Startups are inherently uncertain. But you can reduce some of the biggest risks before spending months creating a product nobody wants.

Start with the problem, not the product

Before asking whether your solution is good, ask whether the problem is important.

Imagine you have an idea for software that automatically organizes meeting notes. It would be easy to immediately start thinking about transcription technology, integrations, dashboards, and AI features.

Instead, investigate how people currently deal with meeting notes.

Do they regularly lose important information? How much time do they spend organizing notes? Have they already tried other tools? What frustrates them about those products? What happens when notes are poorly organized?

These questions reveal the intensity of the problem.

There is a major difference between someone saying, “That is annoying sometimes,” and saying, “We waste five hours every week dealing with this.”

The second response points toward a much stronger potential market.

Founders should therefore define the problem as specifically as possible. Instead of saying, “Companies struggle with productivity,” identify exactly who struggles, what they are trying to accomplish, what prevents them from doing it, and what the consequences are.

The narrower the initial problem, the easier it becomes to test.

Talk to the people who might actually use it

Customer interviews are one of the simplest validation tools available, yet they are surprisingly easy to do badly.

The goal is not to convince people that your idea is brilliant. It is to understand how they behave today.

Instead of explaining the product and asking, “Would you use this?”, ask potential customers to describe the last time they experienced the problem.

What happened? What did they do? What solution did they use? How much did it cost? What did they dislike about it?

Specific stories are more useful than hypothetical opinions.

Someone might sincerely tell you they would pay €20 per month for your future product. That does not necessarily mean they will. But if they already spend €100 per month solving the same problem badly, that tells you something much more meaningful.

After several conversations, patterns should begin appearing. The same frustrations may come up repeatedly. Certain customer groups may care much more than others. You may even discover that the original problem is less important than another problem customers keep mentioning.

That is exactly what validation is supposed to reveal.

Research what customers already use

Competition is not automatically bad news.

In fact, finding competitors can validate an important assumption: people already spend money or time trying to solve the problem.

Study the alternatives carefully. Look at their pricing, positioning, features, customer reviews, target audiences, and business models. Pay particular attention to what customers complain about.

But do not limit competitive research to companies selling similar products.

Sometimes the biggest competitor is a spreadsheet, a WhatsApp group, a notebook, an internal employee, or simply doing nothing.

Your startup has to compete against whatever people currently do.

The question is therefore not just, “Who else sells this?” It is, “Why would someone change their existing behavior to use our solution?”

If you cannot answer that clearly, more validation is probably necessary.

Test demand before building the full product

At some point, conversations need to turn into actions.

One useful approach is creating a simple landing page explaining the problem, proposed solution, and value of the product. Visitors might be invited to join a waiting list, request early access, book a demonstration, or register their interest.

This does not prove that the business will succeed, but it provides stronger evidence than verbal enthusiasm.

Founders can go further.

For a B2B product, you might approach potential customers and try to secure pilot agreements before building the complete platform. For a service, you could manually deliver the result before developing software to automate it. For a consumer product, you might create a prototype and observe whether people actually use it.

The objective is to simulate the real customer decision as closely as possible without building the entire company first.

Ask whether people will pay

Payment is one of the strongest forms of validation because it introduces a real trade-off.

Someone saying “I love this” costs them nothing. Taking out a credit card is different.

Depending on the product, founders might test paid pilots, deposits, pre-orders, letters of intent, or early subscriptions. Not every startup can charge customers before development, and founders should be transparent about what exists and what is still being built.

Even so, conversations about price are extremely valuable.

If potential customers consistently like the product until money enters the conversation, there may be a gap between perceived usefulness and actual willingness to pay.

Pricing discussions can also reveal how customers think about value. A product that saves a company €50,000 per year can potentially support very different pricing from one that saves an individual ten minutes each month.

Build the smallest experiment possible

Validation does not require creating a polished minimum viable product immediately.

Sometimes the first experiment can be remarkably simple.

If you want to build a marketplace, manually connect ten buyers with ten sellers. If you want to automate a business process, perform the process manually for several customers. If you want to create an app, build a clickable prototype before developing the underlying system.

The principle is simple: test the riskiest assumption with the least expensive experiment.

If your biggest uncertainty is whether customers have the problem, conduct interviews. If it is whether they will sign up, test a landing page. If it is whether they will pay, try selling. If it is whether they can use the product successfully, build a prototype.

Each experiment should answer a specific question.

Know what evidence is strong enough to move forward

Validation is rarely a dramatic moment when an idea suddenly becomes “proven.” Evidence accumulates.

Ten enthusiastic conversations are encouraging. Ten people joining a waiting list are stronger. Ten people repeatedly using a prototype are stronger again. Ten customers paying for it provide even more meaningful evidence.

The goal is not to eliminate uncertainty. That is impossible.

The goal is to replace assumptions with progressively stronger evidence.

Sometimes that evidence tells you to continue. Sometimes it tells you to change the target customer, simplify the product, adjust the pricing, or rethink the idea completely.

That is not wasted effort. It is exactly why validation happens before major investment.

A few weeks spent discovering that an idea needs to change can save months of development and a significant amount of money.

The best time to discover that customers do not want your product is before you build it. The second-best time is as early as possible.