Idea validation tactics

'Proof of pay': ship a real app with a simple onboarding and a hard paywall showing one offer, submit every price you want to test up front, and run Apple Search Ads at the same daily spend (~$1K/day to learn fast; far less works over a longer period). Change only the lifetime price, walking it up: $4.99, $9.99, $29.99, $49.99. Read results directionally (day-one revenue per user), not for statistical significance. If you can't get past ~$4.99, drop or change the concept.

Define a 'commitment metric' for yourself and for users. For yourself: a specific date and a number (customers, clients, etc.) you must hit by then, so validation has a fixed time box. For users: the signal doesn't need to be payment, but must represent real usage; for Once it was event hosts agreeing to use the product at an actual dated event in front of friends and family, treated as nearly a proxy for payment. Their bar: 10 events with a real date and commitment before writing a single line of code.

Used ChatGPT to make a one-page HTML site (runify.xyz) with a one-sentence pitch and rank images. Two options: pay $5 via a Stripe link to become an 'early adopter' (lifetime subscription once live) or join the email waitlist. All Reels traffic pointed there.

After reading reviews where users asked to pay to remove ads, tested a one-time purchase of 30 ad-free days at $3, $5 and $10 instead of a subscription.

Took an unused poster design made for a client, put it on a poster mockup (posters were easy to source fast, which matters with TikTok Shop's two-day delivery), listed it on TikTok Shop and made one video with AI (ElevenLabs voice + Midjourney images). When it got 3 sales, he left a $100 deposit with a local print shop so they would print on each call and bill his card afterwards.

Before any code: cold-reached Hollywood speech/accent coaches (who work with actors) for differentiation. First coach, Ron, recorded lesson videos; ~12 beta users were manually sent the videos by email, emailed back speech recordings, and the coach gave asynchronous feedback. Used the pilot to pitch investors.

Over one weekend, put up a website with a 3D wearable mock-up bought on Fiverr and a one-line promise ('Pacing, made simple'), honest that the product didn't exist yet. The CTA went to a Typeform asking 'If we built this product, would you be interested in coming along for the journey with us?'

Instead of building the wearable first, built a free app. Took 100 people from the waitlist ('Founding 100'), ran a Facebook group with them and got on calls constantly. Learned users juggled 3-4 apps (symptom tracking, community, research) and combined them into one experience.

Found a keyword (two variants, one word and two words, each ~30 search volume) that no competitor used in its title, based on an owned app ranking #1 for similar-volume keywords at 100-150 downloads/day. Asked a client with a new app idea to name the app exactly that keyword. Launched with no free trial.

Noticed that under height videos on TikTok, people post their age, height and parents' heights, wanting a prediction. Made reply-to-comment videos predicting those commenters' heights. Inbound DMs requesting predictions served as proof that people care.

VSC framework, run on TikTok before deciding what to build or scale. V (viral): the format or pain point has hit 100K+ views on several similar posts from accounts under 5,000 followers, so the content, not the account's reach, drove views. S (scalable): you can make 100 of them without burning out, using AI video/slideshows or cheap UGC. C (convertible): read the comments on posts that promote a product; if people ask 'what's the app? where can I find this?' it converts; 'lol', disagreement or controversy means it probably won't. After posting, keep scanning comments: a viral post with no product mentions is a sign of low conversion.

Built V1 as a web app in about a week with AI coding (Supabase, Vercel, Mailgun, Stripe, OpenAI API, Tavily search API, Claude Code), charged from the start with Stripe, and skipped polish like logos. Web revenue and feedback came first; a few months later he built a native Swift iOS app (Vercel backend, Supabase, RevenueCat, Superwall paywall A/B tests).