Most startup ideas die for one reason: nobody needed the product the founder built. Not because execution was wrong, not because distribution failed, not because the team was weak. Because the problem was not real enough, or not big enough, or not painful enough for paying customers.
Validating before you build sounds obvious. Almost nobody does it properly. This is the 2026 playbook for doing it in 2 weeks, cheaply, honestly.
We run BetterLaunch.co, see ~200 indie products launch per month, and our pattern-match is clear: the founders who talked to 20 target users before building have roughly 3x the 12-month revenue of founders who built first and talked later.
#TL;DR
- Idea validation = confirming real target users have the problem, and would pay for a solution, before you build.
- The minimum 2-week validation sprint: 20 user interviews + 1 landing page + 1 paid-ad test + Mom Test-style questioning.
- Positive signals: users describe the pain unprompted, existing inadequate solutions exist, they express willingness to pay.
- Negative signals: "that's a cool idea" with no specifics, hypothetical interest, no current workaround.
- Skip validation only if you personally live the pain daily (you are your own target user).
- [BetterLaunch](https://betterlaunch.co/submit) is where to list your validated product after launch.
#Why 90% of startup ideas fail validation (and that's a feature)
A startup idea is a hypothesis. The validation sprint is the experiment.
Hypotheses that fail validation are not wasted effort; they save you 12 months of building the wrong thing. Founders who treat validation as a gate, not a formality, save years of life.
Common failure modes:
- "This idea is so obvious, someone must have built it." Usually: they did, and nobody bought.
- "Everyone I talked to loved the idea." Friends, family, and fellow founders give you politeness, not truth.
- "I know my market." Until you verify with strangers, you don't.
- "We'll figure out pricing later." If nobody will pay, there is no business.
- "The product will sell itself." Almost nothing does.
#The 2-week validation sprint
One full-time week or two part-time weeks. Structure:
#Week 1 days 1-2: define the hypothesis
Write down, in one page:
- Who is the target user (as specific as possible).
- What problem they have.
- How they currently solve it (or fail to).
- What your solution would be.
- Why they would pay for it.
- How much they might pay.
Each of the 6 should be written with specific, testable claims. "SaaS founders need better marketing" is too vague. "Solo SaaS founders at $1K-$10K MRR want to rank for 3-5 keywords in Google to drive organic signups, and currently hire freelancers at $500/post to do it" is testable.
#Week 1 days 3-7: 20 customer interviews
Book 20 calls with target users. Each 20 to 45 minutes. Use the Mom Test framework from Rob Fitzpatrick's book:
- Ask about past behavior, not future intentions. "Tell me about the last time you ran into this problem" beats "Would you use a product that solved this?"
- Ask open questions. "What do you do when X happens?" not "Would you like Y?"
- Avoid pitching. Do not describe your solution until late in the conversation; their opinion before you pitch is the honest one.
- Follow the pain. If the user shrugs, the pain is too small. If they describe a vivid workaround, you are onto something.
What to listen for:
- Gold: "Oh yes, I spent 6 hours last Tuesday fixing this by hand. It drives me insane."
- Yellow: "Yeah, that happens sometimes. Not a huge deal."
- Red: "I guess that could be an issue. I don't really think about it."
After 20 interviews, tally how many hit "Gold." If under 5, the pain is not urgent enough. If 10+ hit Gold, move to Week 2.
Where to find 20 interviewees:
- Your existing network.
- Niche subreddits (ask permission or DM users).
- Indie Hackers forum.
- LinkedIn (cold DMs targeting specific titles).
- Warm intros from anyone who knows the space.
#Week 2 days 8-10: landing page test
Build a simple landing page with:
- Clear headline of the promise.
- Sub-headline of the outcome.
- 30-second demo video or 3 screenshots (can be mockups).
- Email waitlist signup.
- A simple "Would you pay $X/month?" question below the fold.
Tools: Carrd, Framer, or a Next.js template. Budget: $0 to $20.
#Week 2 days 11-12: paid-ad test
Run $50 to $200 of paid ads on Meta, Google, or LinkedIn targeting your exact ICP. Drive traffic to the landing page.
Key metrics:
- Landing page → waitlist signup: 5%+ is strong, under 2% is weak.
- Cost per signup: under $5 is excellent, $5 to $15 is okay, over $20 is concerning.
- Qualitative comments: people saying "I need this" in ad comments is a strong signal.
Tools: set up Meta Ads or Google Ads with a small budget. Track conversions via the landing page.
#Week 2 days 13-14: synthesis and decision
Compare what you learned against your original hypothesis.
- Validated (build): 10+ Gold-level interview signals + 5%+ landing page conversion + willingness-to-pay confirmed.
- Semi-validated (narrow and retest): some signals strong, others weak. Often means the ICP is wrong; narrow to the Gold segment and re-test.
- Invalidated (kill it): weak signals across the board. Move to a new idea.
Do not rationalize weak signals into a "go." The sunk cost is minimal at this point; restarting a 2-week sprint is infinitely cheaper than 12 months of wrong building.
#The Mom Test framework (in 5 bullets)
Rob Fitzpatrick's 2014 book is still the best short text on validation. The 5 core rules:
- Talk about their life, not your idea. Understand their reality.
- Ask about specific past behavior. "Last time you did X, what happened?"
- Listen; don't pitch. Pitching triggers politeness; listening reveals truth.
- Look for commitments. Do they offer to introduce you to someone, make intros, or pay a deposit? Signals.
- Ignore compliments. "Cool idea" is noise. "I would pay $100 for this today" is signal.
One-sentence summary: if you are the one talking in a validation interview, you are doing it wrong.
#Red flags that a startup idea is not validated
- "Everyone would want this." Not a market; "everyone" is not a customer.
- No existing workaround. If no one currently solves this problem, it is usually not a real problem.
- Target users can articulate neither the pain nor the cost of not solving it.
- Willingness-to-pay is hypothetical. "I might pay $10" is different from "here's my credit card."
- Every compliment, no criticism. Real feedback includes objections.
- No urgency. Interviewees shrug, say "someday," or defer.
If you see 3+ of these, hold off on building.
#Positive signals that an idea is validated
- Users describe the pain unprompted. Before you pitch, they tell you about the problem.
- Specific current workarounds exist. Spreadsheets, manual work, glue scripts, other tools stitched together.
- Unsolicited features requests. Before your product exists, they tell you what they want.
- Willingness to sign up, subscribe, pre-order, or pay a deposit.
- Referrals. Interviewees connect you to other target users who have the same problem.
- Competitive activity. Other tools attempting to solve this (even imperfectly) signals a real market.
3+ of these = proceed with cautious optimism.
#The landing page test in detail
The landing page test is a faster proxy for validation when user interviews are hard to schedule.
What to include:
- Headline that promises the outcome.
- Sub-head that clarifies who this is for.
- 3 to 5 bullet benefits.
- A mockup or short video.
- An email waitlist.
- Optional: a "reserve your spot" CTA with credit-card collection (doesn't charge; measures intent).
Traffic to drive:
- $50-$200 of Meta or Google ads targeting ICP.
- 1 post on relevant subreddits.
- 1 post on X/Twitter with BIP thread.
- 1 post on Indie Hackers or similar community.
Signals to measure:
- Visit → signup rate (target 5%+).
- Cost per signup (target under $10).
- Qualitative comments and DMs.
Cost: $50 to $300. Time: 3 to 5 days. Worth it.
#Validation tools and stack
Landing page builders (fast):
- Carrd ($19/year, 30 min to ship).
- Framer (free tier, designer-friendly).
- Next.js template like ShipFast (if you code).
Scheduling interviews:
- Cal.com or Savvycal.
- Google Calendar + a simple scheduling link.
Recording interviews:
- Google Meet (free).
- Fireflies or Otter for transcription.
Surveys (if you can't get enough interviews):
- Typeform or Tally (free tier).
- Note: surveys are weaker signal than interviews; never rely on them alone.
Ad platforms:
- Meta (Facebook/Instagram) Ads for consumer products.
- Google Ads for intent-based queries.
- LinkedIn Ads for B2B (more expensive).
- Reddit Ads for niche-specific.
#When to skip formal validation
Three scenarios where the 2-week sprint is overkill:
- You personally live the problem daily. You are the target user; validation against yourself is the start.
- You have an existing audience that has directly asked for this. 500 people emailing "I'd buy this if you built it" is a validated demand signal.
- The build time is genuinely trivial (weekend project). If you can ship the MVP in 4 days, "build and see" is a valid alternative to formal validation.
Outside these cases, do the validation sprint.
#Common validation mistakes
- Asking friends and family. They'll tell you what you want to hear.
- Leading questions. "Would you pay for X?" is an invitation to say yes.
- Pitching first, listening second. Order matters. Listen first.
- Stopping after 5 interviews. The pattern emerges at 15 to 20.
- Treating compliments as signal. Compliments are social politeness, not demand.
- Ignoring negative feedback. Critical signals are often more useful than positive ones.
- Not writing down answers. Reviewing notes after 20 interviews reveals patterns you missed live.
- Validating the wrong ICP. If pain is low in the group you interviewed, try a narrower group.
#FAQ
How long does startup idea validation take? 2 weeks full-time or 4 weeks part-time is the typical range for a reasonable pre-build validation sprint.
How many user interviews do I need? 20 interviews is the minimum for meaningful pattern recognition. Founders often want to stop at 5; resist.
What's the Mom Test? A framework by Rob Fitzpatrick for conducting validation conversations. Ask about past behavior, not future intentions; avoid pitching; listen for specific pain and workarounds.
How much money should I spend on validation? $0 to $300 is typical. Most cost is landing page domain + ad test. Interviews are free.
Can I skip validation if I'm sure about my idea? Yes if you are the target user, have an existing audience asking for it, or can build an MVP in a weekend. Otherwise, don't.
What if I can't find 20 people to interview? Your ICP is either too hard to reach (consider pivoting) or you haven't looked broadly enough. Try Reddit DMs, LinkedIn cold outreach, niche Slacks, warm intros, and paid recruiting ($50 to $200 on services like UserInterviews.com).
When should I charge during validation? During the landing page test, asking "would you pre-order for $X" or setting up a deposit page (Stripe's "save payment method") is a strong signal. Actually charging typically waits until post-build, but willingness signals can be tested pre-build.
How do I know when the validation is "enough"? When you have 10+ Gold-level interview signals, landing page converting 5%+, and 3+ users willing to take a paid action (pre-order, deposit, committed intro).
#Summary
Idea validation is the cheapest insurance you can buy against wasting 12 months on the wrong product. 2 weeks, $0 to $300, 20 user interviews, 1 landing page, 1 ad test. Run the sprint. Trust the signal.
Once validated, ship fast and list on BetterLaunch for a DR 47 dofollow editorial listing and your first wave of real indie-founder traffic.
List your product on BetterLaunch →
