How to Validate a Startup Idea in 7 Days (Before You Write Any Code)

To validate a startup idea, you need evidence of four things: a real problem, a market that is not already solved, people willing to act, and people willing to pay. You can collect early evidence for all four in about seven days without writing a line of code. This guide walks through that week day by day, with what to do, what to look for, and when to stop.

The goal is not certainty. Nothing short of a launched product gives you that. The goal is to avoid the most expensive mistake in startups, which is spending months and tens of thousands of dollars building something nobody needed.

What does it mean to validate a startup idea?

Validation means replacing opinions with evidence. "My friends love it" is an opinion. "Eight of twelve people I interviewed already pay for a workaround" is evidence.

A useful way to think about it is as a chain of four questions. Each one only matters if the one before it holds up:

1. Problem. Do specific people have this problem often, and does it hurt?
2. Market. What do they use today, and why is it not good enough?
3. Demand. Will strangers take an action, like signing up, when they see your solution described?
4. Value. Will they pay, and roughly how much?

The week below tests them in that order.

Day 1: How do you define the idea so it can be tested?

Write three sentences, and keep them where you will see them all week:

• Who: the exact person with the problem. "Small business owners" is too broad. "Owners of independent dental clinics with two to five chairs" is testable.
• Problem: what goes wrong for them, in their words, not yours.
• Bet: "We believe [who] will [do something] because [reason]."

Then write your decision criteria before you collect any data. For example: "I continue if at least 6 of 12 interviewees describe the problem without prompting, and at least 5% of landing page visitors join the waitlist." Writing this first matters. Once results come in, it is very easy to move the goalposts.

Day 2 and 3: How do you run problem interviews that tell the truth?

Find 10 to 15 people who match your "who" sentence. LinkedIn, niche communities, Slack groups, Reddit, and your own network all work. Offer a 20 minute call, not a pitch.

The rules that make interviews useful:

• Ask about the past, not the future. "Tell me about the last time this happened" beats "would you use an app that...?" People are generous about hypothetical futures and honest about what they actually did.
• Ask what they already tried. If they have never tried to solve it, it may not hurt enough to pay for.
• Ask what it cost them. Time, money, missed revenue, stress. Pain you can quantify is pain people pay to remove.
• Do not pitch. The moment you describe your solution, the interview becomes politeness.

After each call, write down the exact phrases people used. You will reuse them on your landing page on day 5.

Day 4: How do you check the competition without getting discouraged?

List every way people solve this problem today, including the non-obvious ones: spreadsheets, an assistant, an agency, doing nothing. Then, for each direct competitor, note:

Competitors are not bad news. They are proof that people spend money on the problem. The bad news is a market with no spending at all.

This is the step where a research tool saves the most time. Truewick, the AI market research tool we built at Sprout, scores an idea across ten dimensions, benchmarks it against 6,000+ Y Combinator companies, and gathers cited evidence like real complaint quotes and competitor pricing. It is free to start, and it turns a day of tab-hopping into a few minutes, so you can spend the rest of the day reading what it found.

Day 5: How do you build a smoke test landing page?

A smoke test is a simple page that describes your solution as if it exists, with one clear action: join the waitlist, request access, or book a demo. You are measuring whether strangers care enough to act.

Keep it to one screen of content:

• A headline in your customers' words from the interviews.
• Three bullets on what the product does for them.
• One call to action. Not two.

Any no-code page builder works. Connect a simple form and basic analytics so you can count visitors and sign-ups.

Day 6: How do you get traffic and test willingness to pay?

Send 100 to 300 targeted visitors to the page. Post in the communities where you found your interviewees, run a small ad campaign aimed at your "who", or message people directly. Targeting matters more than volume. Fifty of the right people tell you more than five thousand of the wrong ones.

To test willingness to pay, add a pricing step. Two simple ways:

• Show the price on the page. If sign-ups hold up when the price is visible, that is a strong signal.
• Ask for a small pre-order or a paid pilot. Even a refundable deposit separates real buyers from polite ones.

There is no universal "good" conversion rate. It depends on how targeted your traffic is and how big the ask is. That is why you set your own threshold on day 1, before you saw any numbers, and judge against that.

Day 7: How do you decide whether to build?

Go back to the criteria you wrote on day 1 and score honestly. There are three outcomes:

• Go. The problem is real, the market has a gap, and people acted. Move to a tightly scoped MVP. Our guide on how to build an MVP covers the next steps.
• Change. The problem is real but your angle is wrong. Adjust the "who" or the solution and run the week again.
• Stop. People did not recognize the problem or would not act. That is a successful week. You just saved months.

What counts as strong evidence, and what does not?

Not all signals are equal. A useful habit is to sort everything you hear into strong and weak evidence as you go, so the day 7 decision is not swayed by the most memorable conversation.

The pattern is simple. Strong evidence costs the other person something: time, effort, money, or their own words about a real past event. Weak evidence costs them nothing, which is why it is easy to collect and easy to misread.

If most of what you have is in the left column, the honest answer is "not yet", even if it feels like momentum. Run the week again with a narrower "who" or a sharper problem before you spend on building.

What are the most common validation mistakes?

• Asking friends and family. They are the least reliable sample you have.
• Pitching in interviews. You learn whether people are polite, not whether they have the problem.
• Counting compliments. "This is cool" is not a signal. Sign-ups, deposits and referrals are.
• Skipping the price. Interest at a price of zero tells you very little.
• Validating for months. A week of focused work beats a quarter of "research". The goal is a decision.

What comes after validation?

Once your idea passes, the next risk is building too much. Scope the smallest product that tests your bet, and plan for six to twelve weeks of build time. If you want a feel for budget, read how much it costs to build an MVP.

At Sprout, we start every client engagement with this kind of validation, using Truewick, so the first sprint goes into the version of the idea most likely to work. If your week says go, book a free strategy call and we will scope it with you. MVP builds start from $12,000, and our pricing is published.