UI & UX Design

Verified against ChatGPT · 2026-08-13

Review a landing page's UX against the one conversion decision it's actually asking a visitor to make

Reviews a landing page's layout, copy hierarchy, and CTA placement against a single named conversion goal, flagging every element competing with that one decision instead of a generic visual critique.

ChatGPT (GPT-5.1)4 fillable variables

The prompt

Ready to copy — highlighted parts are example details you can swap.

LANDING PAGE CONTENT AND LAYOUT (describe section by section, top to bottom)
Hero with headline + CTA -> logo bar of customer names -> three-feature grid -> customer testimonial -> pricing table -> FAQ accordion -> footer CTA repeat.

SINGLE CONVERSION GOAL
Get the visitor to start a free 14-day trial by entering their email.

VISITOR CONTEXT (how they arrived, what they already know)
80% arrive from a paid search ad for a generic category term, meaning most have never heard of this specific product before landing.

CURRENT CONVERSION RATE OR KNOWN ISSUE
Conversion rate is 1.2%, well below the 3-4% this category typically sees for a similar offer.

Name the one decision CONVERSION_GOAL is actually asking a visitor to make, then go through PAGE_SECTIONS top to bottom and, for each section, judge whether it moves the visitor toward that specific decision, is necessary supporting context the decision genuinely depends on, or is competing with it — a second CTA, a distracting link that leads away from the page, or content that would be valuable on a different page but only delays this one's actual ask. Be specific about competition: a navigation menu that lets a first-time visitor wander to five other pages before ever reaching the CTA is a form of competition even though nothing about it looks like a rival call-to-action.

Check whether the page's information order matches VISITOR_CONTEXT — a visitor arriving cold from a broad ad needs context this page may be skipping straight past into a hard ask, while a visitor arriving from a specific, already-informed source (an email to existing customers, a search for the exact product name) is being made to wade through context they don't need before reaching the CTA, which reads as friction rather than persuasion for that specific audience. If CURRENT_PERFORMANCE names a known issue or number, prioritize whichever section-level finding most plausibly explains it over items that are technically true but less consequential.

Evaluate the CTA itself: is its label a specific, low-ambiguity commitment ("Start your 14-day trial") or a vague action ("Learn more", "Get started") that doesn't tell the visitor what happens on click; is there exactly one primary CTA repeated at logical points down the page, or multiple different asks competing for the same click.

WHAT NOT TO DO
Do not produce a generic visual-design critique (color contrast, font pairing, spacing) unless a specific instance of it is actually interfering with the one named decision — this is a conversion-focused review, not a general design audit. Do not recommend adding more content or reassurance sections as a default fix; more sections between the visitor and the decision is itself often the problem, not the solution.

OUTPUT FORMAT
1. The one decision this page is asking for, stated in one sentence.
2. Section-by-section table: section | verdict (advances / necessary context / competes) | reasoning.
3. Visitor-context fit: is the information order right for how VISITOR_CONTEXT actually arrives.
4. CTA evaluation: label clarity, count/placement, one specific recommended change.
5. If CURRENT_PERFORMANCE names an issue, the single finding above that most plausibly explains it.

Customize

Optional — swap in your own details for the highlighted parts above.

Why this works

Naming the single decision the page is asking for before reviewing anything else prevents the model's default posture toward landing-page reviews, which is a general aesthetic and best-practices critique (contrast, whitespace, font choices) — that's the most heavily represented pattern in generic "review this landing page" material it has seen, and it produces a report full of true but low-leverage observations unless the review is explicitly anchored to what the page is actually trying to get someone to do. Framing internal navigation and unrelated content as forms of competition with the CTA, not just a second visible call-to-action, catches a real and common miss — a model asked generically to check for "competing CTAs" will look for a second button, but a nav bar that lets a first-time cold visitor click away to four other pages is functionally the same failure even though it doesn't look like a rival ask, and it takes an explicit instruction to get the model to name that as competition rather than treating navigation as structurally exempt from the analysis. Checking information order against how the visitor actually arrives (cold ad traffic versus a warm, already-informed visitor) targets a mismatch that's invisible if you only look at the page in isolation — the same page can be correctly sequenced for one traffic source and badly sequenced for another, and a model reviewing the page without being told who's actually looking at it will judge the sequencing against an assumed generic visitor rather than the real one. The explicit ban on defaulting to "add more reassurance content" as a fix matters because that's the single most common generic landing-page recommendation, and it's frequently wrong for a conversion problem, since more sections between the visitor and the ask is itself often the mechanism suppressing conversion, not a fix for it.

What you get back

The one decision: whether to hand over an email address to start a 14-day trial. FAQ accordion section: verdict — competes; for cold ad traffic with no prior product context, an FAQ this early reads as a reason to hesitate rather than reassurance, and it sits above the pricing table a warm, cold visitor hasn't earned context for yet. CTA evaluation: 'Get started' used in hero and footer is ambiguous about what happens next; recommend 'Start your 14-day trial' to make the actual commitment explicit and reduce the anxiety of an unstated next step for a visitor who has never heard of the product before this page.

Verified against

ChatGPT GPT-5.1 · 2026-08-13

Changelog

  • 2026-08-13 Initial publish, verified against ChatGPT GPT-5.1.

Need this built into your business?

If a prompt isn't enough — branding & design, built and maintained for you — that's Scult's day job.

EXPLORE BRANDING & DESIGN
All UI & UX Design prompts

Check your AI visibility

One URL in, a 0–100 score and the exact fixes out.

RUN THE CHECK

Browse all the tools

15 tools across six categories
13 of them never send your data anywhere

Free · No signup · No trial clock

SEE THE DIRECTORY