Startup MVP Workflow — Idea to Validated Landing Page
Validate before you build. One day, not one quarter.
6 steps≈ 8 hoursIntermediateFree
What you'll have at the end: A live landing page with real demand signal from paying-intent visitors, pricing hypothesis, and a go/no-go decision — in one focused day instead of a quarter of building.
Best model: ChatGPT (web) or PerplexityAlts: Claude
Copy-paste prompt
I'm considering a startup around [IDEA]. Target: [SEGMENT].
Research and tell me:
1. Is this problem being discussed on Reddit, Twitter/X, Hacker News, or ProductHunt in the last 90 days? Cite 5 specific threads/posts if yes.
2. Google Trends for the main keywords — rising, flat, or declining? Cite the exact search phrases.
3. Top 3 existing competitors + the ONE weakness each has (specific product feature, pricing, UX gap, geography)
4. The ONE segment where NO competitor serves well — be specific (not "SMBs", but "Shopify stores doing $10k-$50k/month in India selling digital products")
5. Realistic TAM for that segment, with the napkin math
If you can't find evidence of real demand, say so bluntly. Don't confabulate.
Expected outputValidation report + go/no-go on continuing to Step 2.
Step 2 · 45 min
Positioning + Name
A one-sentence positioning + 3 brand name candidates.
Best model: ChatGPT or ClaudeAlts: Gemini
Copy-paste prompt
Based on the validated opportunity, write 3 one-sentence positioning statements in this format:
"[Product] is the [category] that helps [specific segment] [achieve outcome] without [common objection]."
Then give me 10 brand name candidates. For each: domain availability check (just note which TLDs are usually free vs taken for that style of name), the vibe, and the risk (trademark adjacent? hard to spell? already a public brand?)
Avoid cliché startup names: no "ly" suffix, no portmanteaus of generic words, no "AI" in the name unless it's truly AI-core.
Expected output3 positioning options + 10 name candidates with risks.
Step 3 · 60 min
Landing Page Copy
Waitlist-ready landing page with ONE promise.
Best model: Claude 3.5 SonnetAlts: ChatGPT 4o
Copy-paste prompt
Write a waitlist landing page for [NAME]. Positioning: [PASTE POSITIONING]. Target: [SEGMENT].
Structure (keep tight — this is pre-MVP, don't fake features):
- HERO: headline (one bold promise), subhead (specific audience + outcome), waitlist email field + button
- "WHY THIS EXISTS" — 3 sentences on the problem in the founder's voice
- "HOW IT WORKS" — 3 steps describing the EXPERIENCE (not internal tech)
- "WHAT YOU'LL GET" — 3 bullet benefits phrased as outcomes
- Social proof placeholder — "RESEARCH NEEDED: 2 beta user quotes"
- CTA repeat with specific tease ("First 100 signups get X")
- Honest "not yet available" line — early access only
Keep it under 500 words total. No feature laundry. No "revolutionizing the industry" slop.
Best model: Framer AI or Lovable or CursorAlts: Vercel v0, Replit
Copy-paste prompt
Build me a single-page Framer / Vercel v0 / Lovable landing page using this copy: [PASTE COPY].
Design constraints:
- Minimalist, one-column, 1200px max-width
- One hero image or Lottie animation above the fold
- Email capture using Formspree, ConvertKit, or Vercel's own
- Mobile-first, loads under 1.5s on 4G
- No fancy animations beyond a subtle fade-in
- Favicon + Open Graph image (I'll give you the brand mark)
Deploy to a *.vercel.app or *.framer.ai subdomain. I'll point a real domain later.
Expected outputLive landing URL ready to collect emails.
Step 5 · 45 min
Pricing Experiment
Validate price before you build.
Best model: ChatGPT or Claude
Copy-paste prompt
My product is [PRODUCT] targeting [SEGMENT]. Based on competitor pricing (list what you found: [COMPETITORS]) and the segment's willingness-to-pay signals from Reddit/Twitter:
1. Suggest 3 pricing options: anchor low, anchor high, and the "most likely to convert" middle
2. Which pricing page element kills conversions most often for this segment? (fake urgency, hidden add-ons, feature-table bloat, etc.)
3. Design a "pay-what-you-want" or "20 beta spots at 50% off" early-access offer that would feel earned, not spammy
4. Write 2-sentence pricing-page copy for the chosen option
My landing page is live at [URL]. I need to drive 200-500 relevant visitors this week to validate signal.
Give me a 7-day distribution plan with these channels:
1. Reddit — list 3 subreddits where this fits, the EXACT title + first-line of a value-first post (not an ad)
2. Twitter/X — 3 tweet variants that would plausibly perform (not threads, just single posts)
3. ProductHunt — only if the MVP is ready; otherwise, how to warm up
4. Communities (Slack/Discord/Indie Hackers) — 3 specific ones with real engagement
5. One warm outreach to 10 ideal early users on LinkedIn
Then: metrics that mean "proceed to build" — specific conversion rate on waitlist, qualitative signal threshold, and time-to-first-10-emails.
Expected output7-day distribution plan + go/no-go thresholds.
Pro tips
▸If after 500 relevant visitors you have under 2% waitlist conversion, the problem isn't your copy — it's the premise. Pivot the segment, not the design.
▸The founder voice beats polished brand voice at this stage. Write like you're explaining it to a friend over coffee.
▸Charge a small deposit ($5-$10) during waitlist signup if you can — refundable. Converts "I like the idea" into "I'll actually pay".