Verified against ChatGPT · 2026-08-12
Audit one page's on-page SEO against its actual target keyword, flagging only what would genuinely move the needle
Runs a focused on-page audit against a single stated target keyword and business goal, separating real fixes from cosmetic nitpicks so a small team doesn't burn a sprint on changes that were never going to affect rankings.
The prompt
Ready to copy — highlighted parts are example details you can swap.
Audit this page's on-page SEO against its stated target keyword, and be honest about which findings would actually move rankings versus which are cosmetic. PAGE CONTENT (paste full text, or headings + key paragraphs) H1: Corporate Travel Booking Software. Intro paragraph doesn't mention 'corporate travel' until sentence 4. Three H2s, all flat, no H3 nesting despite covering 5 distinct subtopics. Body ~450 words total. TARGET KEYWORD corporate travel booking software CURRENT RANKING POSITION (if known) Position 14, hasn't moved in 3 months TEAM CAPACITY FOR FIXES One developer available for 4 hours this sprint, one writer available for a half-day. Check each of the following against the actual page content given, not in the abstract: title tag keyword presence and framing, H1 presence and match to intent, heading hierarchy logic (does H2/H3 nesting reflect actual subtopic structure or is it flat), keyword presence in the first 100 words, content depth relative to what the target keyword's intent requires, internal links present in the content, image alt text if described, and any obvious duplicate or thin-content signal (very short body text competing for a keyword that clearly needs depth). For every finding, classify it as high-impact (directly affects how the page is evaluated for this specific keyword and intent), medium (helps but isn't likely to move rankings alone), or cosmetic (technically correct practice with negligible ranking effect on its own, like exact keyword density counting). Do not present a cosmetic finding with the same urgency as a high-impact one — a common failure of generic SEO checklists is treating a missing alt tag as equally important as thin content that fundamentally can't satisfy the query, when in reality one of those is a rounding error and the other is the whole game. Given the stated team capacity, recommend the two or three fixes to do first, and explicitly deprioritize the rest rather than handing back an undifferentiated list of twelve things to fix, which in practice means nothing gets fixed because nothing was prioritized. If the current ranking position is given and it's already reasonably strong for this keyword, say so and recommend caution about making large structural changes purely for audit-checklist completeness, since a page that already performs well carries real risk from an unnecessary rewrite. OUTPUT FORMAT A findings table: Check | Finding | Impact (high/medium/cosmetic) | Fix. Followed by a "do these first" shortlist matched to stated capacity, and a one-line caution note if the current ranking is already strong.
Customize
Optional — swap in your own details for the highlighted parts above.
Why this works
A generic on-page SEO checklist treats every item — title tag, alt text, keyword density, heading structure, content depth — as equally worth fixing, which is a real and common failure mode because each item is individually a legitimate best practice; the actual skill in an audit is knowing that a page failing on content depth for a competitive keyword has a fundamentally different problem than a page missing alt text on three images, and this prompt forces that distinction by requiring an explicit impact tier per finding rather than a flat pass/fail list. GPT-5.1 without this constraint will happily generate all twelve checklist items with equal-weight phrasing, because listing checks is the easier default than weighing which ones actually correlate with ranking movement for this specific keyword and page — the impact classification forces the harder, more useful reasoning step. Tying the final recommendation to stated team capacity rather than an idealized "fix everything" list matters because an audit handed to a resource-constrained team with no prioritization functionally produces zero action; a ranked shortlist scoped to what two people can actually do in a sprint is the difference between an audit that changes something and one that becomes a document nobody acts on. The caution flag for pages that already rank reasonably well addresses a specific risk in on-page auditing: rewriting a page purely to satisfy a checklist, when that page is already performing adequately, risks a ranking drop from disrupting whatever combination of factors is currently working, and an audit that doesn't weigh current performance against proposed changes can recommend a net-negative rewrite in the name of technical completeness.
What you get back
Check: Keyword in first 100 words | Finding: target keyword doesn't appear until sentence 4 of the intro | Impact: high — this page is competing on a commercial keyword where early keyword match to search intent is a strong relevance signal | Fix: move a keyword-bearing sentence into the first two sentences of the intro.
Verified against
ChatGPT GPT-5.1 · 2026-08-12
Changelog
- 2026-08-12 — Initial publish, verified against ChatGPT GPT-5.1.
Building this for real?
This is a free starting point. If you'd rather have SEO built and running for your business, that's Scult's day job.
EXPLORE SEO
