How to Validate Your App Idea Without Spending Money on Development

The Hierarchy of Evidence

Not all validation is equal. Ranking from weakest to strongest: (1) people saying they'd use it, (2) people signing up for a waitlist, (3) people paying for early access, (4) people using a manual version, (5) people paying for the real product and coming back. Most founders operate at level 1 or 2 and call it validation. Levels 4 and 5 are where real product decisions get made. Aim for level 3 at minimum before writing code.

• Weakest: 'My friends said it sounds cool'
• Weak: Survey with 50 responses showing interest
• Medium: Waitlist with 200 email signups
• Strong: 10 people paying $X for early access
• Strongest: 10 paying users who came back 3× in a week

The Fake Door Test

Build a landing page that describes your product as if it exists — features, pricing, CTA. Drive 500 people to it (Product Hunt, Reddit, Twitter, small ad spend). Measure: how many click the CTA? How many complete the signup/payment flow? If less than 2% of visitors convert to a signup, you either have a messaging problem or a product problem — either way, you've learned something for $200 in ad spend instead of $20,000 in development.

• Landing page: Framer, Webflow, or even Notion in public
• Traffic: Reddit (relevant subreddits), Twitter/X, small Meta ad
• Measure: email signups, CTA click rate, time on page
• Signal: >3% signup rate on cold traffic = strong early interest

The Concierge MVP

Deliver the outcome your app promises manually, for 5–10 real users, before building the software that automates it. If you're building a travel itinerary app: plan 5 trips manually for strangers via email. If you're building a job board: hand-curate 50 jobs and send them via Mailchimp. If you're building a marketplace: facilitate 3 transactions manually over WhatsApp. Concierge MVPs reveal which parts of the workflow users actually value (often not what you thought) and which parts they abandon.

How to Do Customer Discovery (Without Being Annoying)

Don't ask 'would you use my app?' Ask about past behavior: 'Tell me about the last time you had this problem. What did you do? How much time did it take? Did you pay for anything to solve it? How much?' These questions reveal actual behavior, not hypothetical intent. Book 10 thirty-minute calls with people in your target market. You can find them on LinkedIn, in Facebook groups, on Reddit, or through mutual contacts. Offer $20 Amazon gift cards if needed. Ten honest conversations are worth more than 1,000 survey responses.

• Ask about past behavior: 'the last time you…' not 'would you ever…'
• Look for: emotional language, time spent, money already paid
• Ten 30-minute calls = enough to spot patterns
• Record with Otter.ai or Granola — then review what you heard

Pre-Selling: The Gold Standard

If 10 people will pay you money for a product that doesn't exist yet, you have signal. Not commitment, not 'sounds great' — actual payment. 'I'm building [product] and launching in 90 days. I'm offering 10 founding member spots at $199 — half what it'll cost at launch. Would you buy in?' This is the highest-quality validation signal outside of a live product. Stripe lets you charge a card before a product exists. If you can't close 5 pre-sales at any price, the idea needs rethinking.

When You've Validated Enough

You've validated enough to start building when: you've had 10+ conversations with the target user and heard the same pain described unprompted, at least 3 people have paid or committed to pay, and you can describe the problem and solution in one sentence that users immediately recognize. You haven't validated the technology, the business model, or the exact feature set — those get validated in the build and launch phase. But you've confirmed the problem is real and worth solving. That's enough to start.