How to Validate a Startup Idea Before Building Anything
7 ways to validate a startup idea before writing code, from customer interviews to landing page tests. A step-by-step guide for first-time founders.
- LaunchMap editorial team
- Updated
- Published
- 17 min read
On this page
- What does "idea validation" actually mean?
- Why you should validate before building
- Is your business idea good? 10 honest questions
- Step 1: Define your assumptions
- Step 2: Research the market (without leaving your desk)
- Step 3: Talk to real people (customer discovery interviews)
- Step 4: Test demand with a landing page
- Step 5: Run a pre-sale or smoke test
- Step 6: Analyze competitors and find your edge
- Step 7: Build your validation scorecard
- Common idea validation mistakes to avoid
- How AI can speed up idea validation
- A 7-day plan to test your business idea
- When to drop or pivot an idea: three composite examples
- Conclusion
- Frequently asked questions
Test your idea in about 30 seconds
Map my ideaMany startups do not fail because the product was bad. They fail because too few people needed it, and the money ran out before the founders found out.
CB Insights' 2026 study of venture-backed startups that shut down since 2023 found that 70% ran out of capital, usually as the final blow, and 43% had poor product-market fit. Running out of money is often the symptom. Building something too few people need is a common cause, and it is the one you can test before you spend anything.
Idea validation is the process of testing whether your concept solves a real problem for a real audience before you invest serious time, money or code into it. It is the difference between a confident launch and an expensive guess. This guide shows how to validate a business idea or a startup idea in seven steps that work with zero budget, zero audience and zero technical skills, then turns them into a plan you can finish in a week. If you are still choosing what to test, our list of startup ideas pairs each idea with a first test under $100.
You can start the first step in about 30 seconds: the free business idea validator names your first customer, the three assumptions most likely to kill the idea with a cheap test for each, and real competitors from a live web search. No sign-up is needed, and saving the result turns it into a step-by-step launch route.
What does "idea validation" actually mean?
Startup idea validation is not about asking friends if your idea sounds cool. It's not a survey with leading questions. And it's definitely not "I Googled it and nothing like this exists."
True validation answers three questions:
- Problem validation: does this problem actually exist, and do people care enough to pay for a solution?
- Solution validation: does your proposed solution address the problem in a way that's better, faster, or cheaper than current alternatives?
- Market validation: is there a large enough group of people who have this problem and are willing to pay?
If you can answer "yes" to all three, you've got something worth building. If not, you've saved yourself months of wasted effort, which is just as valuable.
Why you should validate before building
Skipping validation is the most expensive shortcut a founder can take. Here's why pre-launch validation matters:
- It saves time. Building an MVP takes weeks or months. A validation sprint takes days.
- It saves money. You won't burn runway on features nobody wants.
- It reduces risk. You go into development with evidence, not assumptions.
- It sharpens your positioning. Talking to real users helps you find the exact words and angles that resonate.
- It builds early traction. Validation conversations often turn into your first waitlist signups, beta testers, or even paying customers.
Eric Ries's The Lean Startup popularized the build, measure, learn loop. You can run the measure and learn parts before you build anything at all, which is what the rest of this guide does.
Is your business idea good? 10 honest questions
A business idea is good when a specific group of people has the problem often, already pays to work around it, can be reached cheaply and will pay a price that covers your costs. Nobody can judge that from the idea alone, so the honest way to answer "is my business idea good?" is to count how many of the questions below you can answer with evidence from customers rather than with hope.
| Question | A good answer looks like | A warning sign |
|---|---|---|
| 1. Who exactly has the problem? | A group narrow enough to find this week, such as independent bookkeepers with 20 to 50 clients | "Everyone" or "small businesses" |
| 2. When did they last run into it? | Most people you ask describe a specific, recent time without prompting | They agree it sounds annoying but cannot name a single time |
| 3. What do they use today? | A workaround that costs time or money: a spreadsheet, a freelancer, a tool they complain about | Nothing at all, which usually means it does not hurt enough |
| 4. Have they paid to solve it? | They already spend money on the problem, so your price replaces an existing cost | They say they would definitely pay for something that does not exist yet |
| 5. Can you reach the first 10 customers without ads? | You can name the communities, lists or contacts where you will find them | The plan starts with going viral or a large ad budget |
| 6. Does the price cover your costs? | Price minus the cost of each sale leaves enough to cover fixed costs at a volume you can reach | You need thousands of customers at a low price just to break even |
| 7. Why would they switch to you? | A specific reason for a specific group: faster, cheaper, simpler or built for their niche | "Nobody else does this", or better design as the only reason |
| 8. Is the market big enough for your goal? | A bottom-up count of customers times a yearly price that covers the income you need at a small share | A top-down "$50 billion market" with no customer count behind it |
| 9. Can you test it within a week for under $100? | Interviews, a landing page or a preorder can show whether people commit | The only way to find out is to build for months |
| 10. Do you want to work on this problem for years? | You know the customers or the industry and would keep going after the first rejections | You picked it because the market looks hot |
Count the questions you can answer with something customers said or did, not with a guess. With 8 or more, move on to a preorder or a paid pilot. With 5 to 7, the idea is promising, and the gaps are the riskiest assumptions to test first. With fewer than 5, the idea is still a guess, so start with interviews before you spend money. These cut-offs are our rule of thumb, not a benchmark.
Step 1: Define your assumptions
Every startup idea is a bundle of assumptions. Before you test anything, write them down explicitly.
Ask yourself:
- Who has this problem? (Your ideal customer profile)
- What is the problem, in their words?
- How are they solving it today? (Existing alternatives)
- Why would they switch to your solution?
- How much would they pay?
Be specific. "Small business owners" is not an ideal customer profile. "Solo e-commerce founders doing $10K to $50K a month who spend 5 or more hours a week on inventory" is a customer you can actually find and talk to. Our guide to finding your ideal customer profile shows how to narrow it down.
Write down your riskiest assumption: the one thing that, if wrong, kills the whole idea. That is what you validate first.
Tip: write each assumption as a sentence that can be proven wrong, then rank them by the damage they do if they are wrong. A lean canvas puts all nine parts of an idea on one page, which makes the weakest one easy to spot.
Step 2: Research the market (without leaving your desk)
Before talking to anyone, do your homework. Desk research helps you understand the landscape and avoid obvious dead ends.
Search volume and demand signals
Use free tools to check whether people are actively searching for solutions to the problem you're solving:
- Google Trends: is interest growing or declining?
- Google Keyword Planner: what's the monthly search volume for related terms?
- AnswerThePublic: what questions do people ask about this topic?
- Reddit and Quora: are people complaining about this problem? How often?
If nobody is searching for it, that's not necessarily a dealbreaker, but it means you'll need to educate the market, which is much harder and more expensive.
Competitive analysis
Search for existing solutions. If there are zero competitors, be cautious: it might mean there's no demand. If there are many, look for gaps:
- What do customers complain about in reviews (G2, Capterra, Trustpilot)?
- What features are missing?
- What audience segments are underserved?
Your startup doesn't need to be first. It needs to be better for a specific group of people.
TAM, SAM, SOM
Estimate your market size:
- TAM (total addressable market): everyone who could theoretically use your product.
- SAM (serviceable addressable market): the segment you can realistically reach.
- SOM (serviceable obtainable market): the share of SAM you can realistically win in the next few years.
You do not need a 50-page report. A back-of-the-napkin estimate with real data points is enough to decide whether the opportunity is worth pursuing. The TAM SAM SOM calculator does the bottom-up math from a customer count and a yearly price, and the TAM SAM SOM examples show what finished estimates look like.
Step 3: Talk to real people (customer discovery interviews)
This is the most important step, and the one founders skip most often.
Customer discovery interviews are short (15-30 minute) conversations with people who match your ideal customer profile. The goal is not to pitch your idea. The goal is to understand their problem deeply.
How to find people to interview
- LinkedIn: search by job title, industry, or company size. Send a short, honest message.
- Online communities: reddit, Slack groups, Discord servers, Facebook groups.
- Your network: ask for warm introductions.
- Twitter/X: search for people complaining about the problem.
Aim for 10-15 interviews. That's usually enough to spot clear patterns.
What to ask
Follow the principle from Rob Fitzpatrick's book The Mom Test: ask questions that even your mom cannot lie to you about. The free customer interview questions generator writes a full set for your product and customer. The core ones:
- "Tell me about the last time you dealt with [problem]."
- "What did you do to solve it?"
- "What was the hardest part?"
- "Have you tried any tools or services? What did you like and dislike?"
- "If you could wave a magic wand and fix one thing about this process, what would it be?"
Do not ask: "Would you use an app that does X?" People say yes to be polite. It means nothing.
What to listen for
- Frequency: how often does this problem occur?
- Intensity: how painful is it? (Mild annoyance vs. hair-on-fire problem)
- Willingness to pay: have they already spent money trying to solve it?
- Current workarounds: what do they do today? (Spreadsheets, manual processes, competitor tools)
If 7 out of 10 people describe the same problem with genuine frustration, you're onto something.
Step 4: Test demand with a landing page
Conversations tell you if the problem is real. A landing page tells you if your positioning and messaging can attract people at scale.
What your validation landing page needs
Keep it simple. One page, one message, one call to action:
- Headline: clearly state the problem you solve and for whom.
- Subheadline: describe the solution in one sentence.
- 3-4 bullet points: key benefits (not features).
- Social proof: even early-stage quotes from interviews work.
- CTA: "Join the waitlist," "Get early access," or "Try free."
- Email capture form: this is your validation metric.
How to drive traffic
You do not need a big budget. Free and low-cost options, compared in more depth in our guide to marketing channels for early-stage startups:
- Post in relevant Reddit communities (add genuine value, do not spam).
- Share in niche Slack and Discord groups.
- Run a small experiment on X or LinkedIn with a compelling post.
- Run $50 to $100 of Google or Meta ads targeting your ICP.
What counts as validation?
There is no standard number. For context, Unbounce reports a median conversion rate of 6.6% across 41,000 landing pages of all kinds. These are rough rules of thumb that many founders use for a validation page, not industry benchmarks:
- Email signup rate above 5-10%: your messaging resonates.
- Below 2%: your positioning needs work, or the audience isn't right.
- People reply to your confirmation email asking when you launch: strong signal.
- People share the page organically: very strong signal.
Step 5: Run a pre-sale or smoke test
The strongest form of validation is when someone pays you money.
Before building your product, you can:
- Sell a pilot or consulting engagement: deliver the outcome manually for 3-5 customers. You'll learn exactly what they need.
- Pre-sell with a discount: "Get lifetime access for $49 if you sign up before launch." Real dollars beat survey answers.
- Create a "Wizard of Oz" MVP: the frontend looks automated, but you do the work behind the scenes. Our MVP roadmap guide covers when this beats building.
- Offer a free tool or template: give away a simplified version of your value. If people use it, there is demand for the full product.
For more ways to test before you build, our minimum viable product examples collect 24 first versions, from Kickstarter pre-orders to a 10-day pop-up grocery store.
If you can get 5 to 10 people to pay, even a small amount, before you write a single line of code, you have validated both the problem and the willingness to pay. Before you set the price, check that it covers your costs: the break-even calculator shows how many sales a month a price needs, and the unit economics calculator shows whether a customer is worth more than it costs to win them. Our guide to the LTV to CAC ratio explains how much more is enough. To pick the pre-sale price and decide how many buyers count as a pass, see how to test your first price.
Step 6: Analyze competitors and find your edge
By now, you have real data. It's time to sharpen your competitive positioning.
Map out 3-5 direct and indirect competitors. For each one, note:
- Who they target (and who they ignore).
- Their pricing model.
- Their biggest strengths and weaknesses (based on reviews).
- What customers wish they did differently.
Then define your unique angle:
- Niche down: serve a specific audience better than anyone else.
- Simplify: if competitors are complex, be the easy option.
- Price differently: freemium, usage-based, flat-rate.
- Offer a better experience: speed, design, support.
Your competitive edge doesn't have to be a feature. It can be a focus, a price point, or even a brand voice.
Step 7: Build your validation scorecard
After completing these steps, score your idea across five dimensions:
| Dimension | Question | Score (1-5) |
|---|---|---|
| Problem | Do people describe this as a real, frequent, painful problem? | ___ |
| Demand | Did your landing page / pre-sale attract real interest? | ___ |
| Willingness to Pay | Have people paid (or clearly stated they would pay) for this? | ___ |
| Market Size | Is the addressable market large enough to build a business? | ___ |
| Competitive Gap | Is there a clear opening that competitors aren't filling? | ___ |
20-25: Strong validation. Build with confidence. 15-19: Promising, but refine your positioning or ICP. 10-14: Weak signal. Pivot or dig deeper before building. Below 10: Go back to step one with a new angle.
Common idea validation mistakes to avoid
Even experienced founders fall into these traps:
- Confirmation bias: only talking to people who agree with you. Seek out skeptics.
- Feature obsession: validating a feature list instead of a core problem.
- Overthinking TAM: a $100B market means nothing if you can't reach anyone in it.
- Skipping willingness to pay: "I'd definitely use that" is not the same as "Here's my credit card."
- Validating with the wrong audience: your ICP matters. Feedback from the wrong people is noise.
- Analysis paralysis: validation should take days, not months. Set a deadline and commit.
How AI can speed up idea validation
AI cannot validate an idea: only customers can, by paying or committing. What AI does well is the desk work around the tests: finding the assumptions that matter, researching competitors and writing interview scripts, so your days go into conversations instead of preparation.
LaunchMap is built around that split. It turns an idea into a five-stage route from problem to growth, with a measurable gate at the end of each stage:
- Riskiest assumptions first: each one comes with a cheap test you can run this week.
- Market research with sources: real competitors, prices and communities, each linked to the page it came from.
- Missions with ready-made materials: interview scripts, recruiting messages and landing page copy for the current stage.
- A route that recalculates: log what happened in an interview or a test, and it updates your assumptions, checks the gate and changes the next missions.
This does not replace talking to customers. Nothing does. It gives you a sharper starting point and keeps every result in one place. You can map your idea for free, and the plans page lists what Pro adds.
A 7-day plan to test your business idea
You can test a business idea in seven days: assumptions and recruiting on day 1, desk research on day 2, customer interviews on days 3 to 5, a landing page and a request for money by day 6, and a decision on day 7. Each day ends with a pass line (our rule of thumb, not a benchmark), which makes the plan a startup validation checklist as well: if you miss a pass line, fix that part before you spend money on the next one. If you only have evenings, give each day two evenings and finish in two weeks.
Day 1: Assumptions and recruiting
- Write down your idea in 2 to 3 sentences
- List your top 5 assumptions and mark the riskiest one
- Define your ideal customer profile (be specific)
- Get a free snapshot of the riskiest assumptions and competitors from the idea validator
- Message 20 or more people who match the profile and ask for 15 minutes about the problem, not about your idea
Pass line: the riskiest assumption is written as a sentence that can be proven wrong, and the interview requests are out.
Day 2: Desk research
- Check search volume and trends for your problem space
- Read 20 or more reviews of competitor products
- Identify 3 to 5 communities where your ICP hangs out
- Keep recruiting until at least 10 interviews are booked
Pass line: you can name what customers use today, what they complain about and where they gather.
Days 3 to 5: Customer interviews
- Run 10 to 15 interviews about the last time the problem happened
- Document patterns: top problems, current solutions, money already spent
- On the evening of day 5, put up a one-page site with the problem, your promise and a waitlist
Pass line: at least 7 of 10 people describe the same problem without prompting, and some of them already spend time or money on a workaround.
Day 6: Demand and commitment
- Share the page in 3 communities or spend $50 to $100 on ads aimed at your ICP
- Track signup rates and qualitative feedback
- Ask the interviewees with the worst version of the problem for a deposit, a preorder or a paid pilot
Pass line: more than 5 percent of targeted visitors join the waitlist, and at least one person commits money.
Day 7: Decision
- Complete your validation scorecard
- Decide: build, pivot or drop
- If "build", define your MVP scope and start, and estimate the money it needs with the startup cost calculator
Pass line: a written decision with the evidence behind it, and the next test booked if the answer is pivot.
When to drop or pivot an idea: three composite examples
Drop an idea when its riskiest assumption fails with the right customers, and pivot when the problem holds up but your customer, offer or price does not. All three examples are composites: each combines situations that come up often in early validation into one story, and the numbers are illustrations, not data from a real company.
1. Dropped: camping gear rental between neighbors. The riskiest assumption was that people who camp a few times a year would rather rent gear than buy or borrow it. The gate, written down before the test, was 5 paid deposits within two weeks. Twelve interviews showed that most of these campers borrowed gear from friends for free, and the landing page collected 40 signups but no deposits. The problem was real, but a free alternative solved it well enough, so the founder dropped the idea and kept the list of campers for a future product.
2. Pivoted to a new customer: chasing late invoices. The plan was a service that chases unpaid invoices for freelance designers. Freelancers described late payments, but none of them agreed to pay for a pilot. Three of the interviewees ran small agencies with 10 to 30 clients, and all three already paid an assistant to chase invoices. The founder switched to agencies, set a new gate of 3 paid pilots from 10 agency conversations, and reached it in two weeks.
3. Pivoted the offer: a weekly bread subscription. A home baker tested a weekly sourdough subscription in a mid-sized town against a gate of 15 subscribers. Tastings went well, but only 2 of 30 interested buyers wanted a weekly commitment. Most of them asked to order for weekends and occasions instead, so the baker replaced the subscription with Friday preorders and pickup. The new gate was 20 preordered loaves for the first Friday, and they sold out two days early.
In all three cases the decision took days, not months, because the pass line existed before the results came in. Without one, mixed results tend to look good enough to keep going. If the answer is drop, start the next idea from what you are good at: the free business idea generator suggests ideas matched to your skills, budget and time, and our list of startup ideas is another place to look.
Conclusion
Validating a startup idea is not about removing all risk. That is impossible. It is about removing the unnecessary risk that comes from building on untested assumptions.
Good founders fall in love with the problem rather than the idea, and they let evidence guide them toward the solution. You do not need a product to start. You need clarity on who you serve, what problem you solve and whether anyone will pay for it. If you are weighing how to fund what comes next, our comparison of bootstrapping and fundraising is a good next read.
Frequently asked questions
How long does it take to validate a startup idea?
A week is enough for a first decision if you work on it every day: two days for assumptions and desk research, three days of customer interviews, a day for a landing page and a request for a deposit or preorder, and a day to decide. If you only have evenings, plan for two weeks. If it takes longer than a month, the test is usually too broad or the decision keeps being postponed.
How do you know if a business idea is good?
Look for evidence, not opinions: a specific group of people has the problem often, already spends time or money on a workaround, can be reached without a big ad budget and will pay a price that covers your costs. If customers' behavior backs most of those points, the idea is worth a preorder test. If only your own enthusiasm backs them, start with interviews.
What is the cheapest way to validate a business idea?
Talking to potential customers costs nothing but time. After that, a one-page site with a waitlist, plus posts in the communities where your customers gather or $50 to $100 of ads, gives you a demand signal for the price of a dinner.
What signup rate counts as validation for a landing page?
There is no standard number. Many founders treat 5 to 10 percent of targeted visitors joining a waitlist as a good early signal and under 2 percent as a sign that the message or the audience is wrong. A deposit or a preorder is stronger evidence than any signup rate.
Can you validate a startup idea without building a product?
Yes. Interviews, a landing page, a preorder and a pilot you deliver by hand all test the idea before any code exists. Delivering the result manually behind a simple website is often called a Wizard of Oz test.
What is the difference between problem validation and market validation?
Problem validation checks that people have the problem and care enough to act on it. Market validation checks that enough of those people exist and will pay a price that can sustain a business. The second only means something once the first is done.
What should you do if your idea fails validation?
Look at which assumption failed. If the problem is real but the price is not, change the offer or the customer group. If nobody has the problem, drop the idea and keep what you learned about the customers, which often points to a better problem.
Written by the LaunchMap editorial team. We research each guide from primary sources and check every number before it goes live. Sources last checked on November 10, 2026.
Sources (4)
Updated November 10, 2026: added a 10-question checklist for judging whether a business idea is good, turned the action plan into a 7-day plan with a pass line for each day, and added three composite examples of drop and pivot decisions.