Basecamp has A/B tested their landing page continuously since 2004. After hundreds of variations, the winning patterns are specific: headline states the outcome, social proof appears above the fold, CTA uses action language, the page loads in under 2 seconds. Most indie developers skip the testing and guess. Reverse thinking for your landing page produces a structured first version faster than guessing and closer to what testing eventually reveals.

The blank landing page problem

You've built the product. Now you need a landing page. You open v0 or a Tailwind template and stare at the hero section. What should the headline say? How much social proof? One CTA or two? The blank page produces paralysis because every choice feels arbitrary without data.

The reverse approach eliminates paralysis by starting with failure. Describing the worst landing page requires no design skill and no testing data — you already know what a bad page looks like because you've bounced from hundreds of them.

Design the worst landing page

Spend three minutes. Be specific. The worst landing page: headline is the company name ("Welcome to Acme"). Hero image is a generic stock photo of a person at a laptop. Five paragraphs of text before any visual demonstration. CTA says "Submit" in gray on white. No social proof. Three navigation links that lead away from the page. Pop-up appears after 2 seconds asking for an email.

Flip every element

"Company name headline" → headline states the user's problem or outcome. "Stock photo" → screenshot or video of the product working. "Five paragraphs of text" → one sentence plus a demo. "Submit in gray" → high-contrast CTA with specific action ("See it work" or "Start free"). "No social proof" → three user quotes or a usage number above the fold. "Three nav links" → zero navigation (the page has one job). "Pop-up at 2 seconds" → no pop-up (or exit-intent only).

Each reversal is a concrete design decision. You now have a spec, not a blank page.

Example: a SaaS landing page reversed

Worst version: targets "everyone," lists 12 features in a grid, pricing requires clicking to a separate page, testimonials are anonymous ("Great product! — J.S."), page weight is 4MB.

Reverse thinking landing page version: targets a specific persona ("for freelance designers billing £3K-10K/month"), shows one primary feature with a working demo, pricing visible on the same page, testimonials include full names and companies, page weight under 500KB. This reversed spec matches what Stripe, Linear, and Vercel actually ship — specific, fast, focused.

When to use reverse thinking for your landing page

Use the reverse thinking landing page method before your first page (replaces guessing with structured decisions). Use it before any A/B test (the worst version reveals what specific element to test). Use it when conversion drops (the worst version highlights which current element has drifted toward bad).

Sparks trains reverse thinking across Chapter 3 with exercises mapped to product and marketing scenarios. AI scores each response on specificity — vague failures produce vague reversals, so the scoring pushes toward concrete elements that produce actionable design specs.