Every PageLint recommendation traces to a primary source with a confidence rating. Not “best practices” — verifiable claims with named books, peer-reviewed studies, and original eye-tracking research.
Can a first-time visitor understand what you do, who it's for, and why they should care — in 5 seconds?
One H1, visually unmistakable. Multiple competing headlines split attention and dilute message.
A first-time reader must understand the product from H1 + subhead alone, without scrolling.
Banned: leverage, synergize, unlock, transform, revolutionize, disrupt, empower, seamless, intelligent, robust, world-class. Replace with plain words.
Don't open with "Don't," "No more," "Stop." Negation forces readers to think the unwanted thought first.
Plus 4 more clarity checks in every audit →
Does the page meet visitors where they are — by audience, awareness state, and the language they actually use?
Target customer named or implied: role, company size, industry, situation. "For everyone" converts no one.
Lead with PROBLEM for unaware traffic. Lead with MECHANISM for solution-aware. Lead with NAME + DEAL for product-aware. Mismatch is the most expensive copy mistake.
At least one phrase verifiable in real customer language — Reddit, G2, Trustpilot — not internal product naming.
Plus 2 more relevance checks in every audit →
Is the value proposition specific, quantified, focused on one promise, and clearly different from the category leader?
Number, percentage, named source, quantitative outcome — not pure adjective.
Distinct value props in hero ≤2. Three+ = committee anti-pattern (everyone got their feature in).
Page says how it's better than the alternative — named or implied. "We're different" is not differentiation.
Headline mentions the outcome the visitor experiences, not an internal feature.
Plus 4 more value checks in every audit →
Does the page channel an existing mass desire and direct attention to its most favorable dimension — or try (and fail) to create new desire?
Connects to existing desire (gain, fear, status, comfort). Schwartz: copy can't create desire — only channel it.
Above-fold prominently surfaces what's strongest about the offering. What's focal becomes causal.
Hero/early body uses "you" or role-specific references. Self-relevance amplifies attention.
Real deadline, real inventory, real cohort cap. "Limited time" that runs forever is fake scarcity.
Plus 2 more motivation checks in every audit →
Are testimonials, authority signals, and pricing real, specific, and verifiable — or smuggled in?
Testimonial has name, photo, role, company, specific outcome. "Jane S., happy customer" is not a testimonial.
Featured customers match ICP. Cialdini's similarity amplifier: peers persuade more than celebrities.
Credentials, "as featured in" logos, certifications link to verifiable sources. FTC-actionable if false.
Price visible or one click away. "Contact sales" without anchor is friction for transactional offers.
Plus 4 more trust checks in every audit →
Does the path from interest to action have any unnecessary friction in the CTA, the form, or the contrast against the page?
One dominant CTA. Secondary CTAs subordinate. Two equal-weight buttons confuse the decision.
Not "Submit" or "Learn more." Specific action + outcome. Peep's TextMagic test: +37.6% from CTA copy alone.
Contrast matters. Color doesn't. Kadavy on CXL: "There's no best button color."
Only fields essential for the next step. Forms with ≥7 fields warrant testing reduction.
Plus 1 more friction check in every audit →
Does the page focus attention on the primary action — or compete with itself through carousels, social icons, and interruptions?
Above-fold has ≤2 CTAs total. More = committee anti-pattern.
Static hero. Motion fights for attention. Birkett on CXL: "Moving images don't do well."
Social icons leak traffic from page. If present, keep below fold or in footer.
Auto-open before user reads hero = anxiety + distraction.
Cialdini calls these "smuggler" patterns — fake invocations of real psychology principles. We flag them. Some are also FTC- or EU-actionable.
A countdown that resets when it hits zero is fake scarcity. Princeton dark patterns research. The EU Digital Fairness Act enforces against it.
A stock counter that doesn't decrement is fake scarcity. FTC-actionable under Rule on Use of Consumer Reviews.
Fake reviews. The FTC issues civil penalties up to $50K per violation under 16 CFR Part 465 (Oct 2024).
"Most people don't [desired behavior]" actually PROMOTES the unwanted behavior. The Petrified Forest field study showed a 3× increase in theft.
Plus 4 more dark patterns checks in every audit →
Four canonical sources don’t always say the same thing. Where they conflict, here’s how we read them.
Ogilvy says long copy sells. CXL data shows short pages often win. Both true: Schwartz's awareness state determines which applies. Most-aware audiences need less; unaware need more.
Most blogs still teach 6. We use 7 — Cialdini added Unity (shared identity) in Pre-Suasion (2016). Different from Liking. Worth knowing.
Cialdini shows 7 levers to amplify response. Schwartz says you can't push — only channel existing desire. Schwartz is precondition; Cialdini is amplifier.
Peep Laja: "Best practices suck." Button-color folklore ("green converts!") doesn't survive A/B testing. We flag contrast, not color. Context > rules.
Every check ships with one of three labels in the audit, so you can tell folklore from research at a glance.