SEO & GEO/AEO

Verified against ChatGPT · 2026-08-08

Tear down a competitor's top-ranking page into a specific list of what your draft is missing

Compares your existing page against a named competitor URL that outranks it and produces a concrete gap list — missing subtopics, entities, and structural elements — instead of a vague 'improve your content' verdict.

ChatGPT (GPT-5.1)4 fillable variables

The prompt

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

You are a senior SEO strategist doing a competitor teardown for one specific keyword where a competitor outranks us, not a general competitive audit.

OUR PAGE
1,400-word guide on 'switching payroll providers mid-year,' covers timeline and a generic checklist, no cost breakdown or state-specific rules.

COMPETITOR PAGE
2,600-word guide with a state-by-state compliance table, a cost calculator embed, and named case studies from three mid-size companies.

TARGET KEYWORD
switch payroll provider mid-year

WHAT WE ALREADY TRIED
Added an FAQ section last month, no ranking change after three weeks.

RULES
Do not restate generic SEO advice like "add more detail" or "improve E-E-A-T" — every finding must point to something specific and missing that you can name: a subtopic the competitor covers and we don't, an entity or named example they reference that we skip, a structural element (comparison table, FAQ, decision framework) they include that we lack, or a level of specificity (numbers, named tools, concrete steps) they hit that we stay abstract on. If you cannot tell from the summaries whether a gap is real, say so explicitly and ask for the missing detail rather than inventing a plausible-sounding gap. Do not assume the competitor's higher ranking is only about content — note if the gap looks like it could instead be backlinks, domain authority, or page experience, and flag that as a separate finding rather than folding it into the content gap list. Rank the gaps by how directly each one addresses the actual target keyword's intent, not by how easy each gap is to fix.

OUTPUT FORMAT
1. A ranked list of specific content gaps (subtopic, entity, structure, or specificity), each stated as one sentence naming exactly what's missing.
2. A separate short list of gaps that look like they might not be content-related at all.
3. Three subtopics or sections to add, each with a one-line reason tied to the target keyword's intent.
4. One paragraph on what NOT to copy from the competitor even though it ranks — anything that looks like keyword stuffing, thin filler, or an approach that wouldn't fit our brand voice.

Customize

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

Why this works

GPT-5.1 defaults to generic SEO commentary ("add more depth," "improve authority") when given two unstructured page summaries and asked to compare them, because those phrases are statistically the most common completion pattern in SEO-advice training data and require no actual claim about either page's content — the explicit ban on that language forces the model to instead search the two summaries for named, checkable differences, which is the only kind of finding a content team can actually act on. Separating content gaps from non-content gaps (backlinks, domain authority, page experience) matters mechanically because ranking correlation is not causation, and a model asked only "why does this page rank higher" will confidently attribute the entire gap to content even when the summaries given don't support that conclusion — forcing it to flag ambiguity keeps it from overstating certainty it doesn't have. Asking the model to explicitly withhold judgment when the summaries don't contain enough detail to identify a real gap directly counters the tendency of large language models to fill an evidence gap with a plausible-sounding but fabricated specific, since a request for "specific gaps" without that guardrail tends to produce confident-sounding invented details rather than an honest "insufficient information" flag. The final instruction to name what NOT to copy addresses a real failure mode of competitor teardowns: the highest-ranking page sometimes wins despite thin or manipulative tactics rather than because of them, and copying those tactics without acknowledging the trade-off would optimize for the wrong signal.

What you get back

1. Missing: state-by-state compliance breakdown — competitor lists specific rules for CA, NY, TX; ours stays generic. 2. Missing: named case studies — competitor cites three real company examples with numbers; ours has none. 3. Possibly non-content: competitor's domain likely carries more backlinks in this niche — recommend checking referring domains before assuming content alone explains the gap. Do not copy: the embedded calculator appears to duplicate content already covered in prose, which may be padding word count rather than adding real value.

Verified against

ChatGPT GPT-5.1 · 2026-08-08

Changelog

  • 2026-08-08 — 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
All SEO & GEO/AEO 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