Verified against ChatGPT · 2026-08-11
Read the actual SERP for a target keyword before deciding what kind of page to write
Analyzes what's currently ranking for a keyword — result types, format patterns, and what questions are already answered — to decide what kind of page is worth building before a single word of content gets drafted.
The prompt
Ready to copy — highlighted parts are example details you can swap.
Before I write anything for this keyword, analyze the current SERP as it actually is, not what I assume it should be. TARGET KEYWORD how to calculate freelance hourly rate CURRENT TOP RESULTS (titles, formats, brief description of what each covers) Position 1: interactive rate calculator tool. Position 2: Reddit thread with 200+ comments. Position 3: blog post with a formula and worked example. Position 4-5: more calculator tools. WHAT WE'D WANT TO PUBLISH A 1,500-word blog post explaining the formula for calculating hourly freelance rate. PHASE 1 — READ THE SERP AS EVIDENCE OF INTENT Look at the actual result types occupying the top of this SERP (listicles, single-product pages, forums, comparison tools, video, tool/calculator pages) and use that mix as direct evidence of what Google's ranking systems currently believe searchers want — do not override this evidence with an assumption about what intent "should" be for this keyword. If the results are mixed (some commercial, some informational), say so explicitly rather than forcing a single clean intent label onto an ambiguous SERP. PHASE 2 — GAP CHECK Check whether our intended content format actually matches what's currently ranking. If we planned a blog post but the SERP is dominated by interactive tools or forum threads, name that mismatch directly — this is the single most common reason a well-written page fails to rank, and it needs to be caught before drafting starts, not after. PHASE 3 — DIFFERENTIATION CHECK Among the results that do match our intended format, identify what none of them currently do — a specific angle, level of specificity, or structural element absent from all of the current top results. Do not suggest "be more comprehensive" as the differentiation; name an actual specific thing to include that isn't already there. PHASE 4 — RECOMMENDATION Give a clear go/reconsider/no-go recommendation on the originally intended content format, with the reasoning stated in one paragraph a content lead could relay to a stakeholder without further explanation.
Customize
Optional — swap in your own details for the highlighted parts above.
Why this works
The instruction to treat the actual SERP composition as evidence rather than override it with an assumed intent label directly counters GPT-5.1's default behavior when asked to classify search intent, which is to reach for one of the standard four-way taxonomy labels (informational, navigational, commercial, transactional) even when the real SERP is genuinely mixed — forcing the model to describe the SERP as evidence first prevents it from flattening a legitimately ambiguous ranking pattern into a false certainty. The gap check in phase 2 exists because the most common and most expensive content-strategy mistake is deciding on a format before checking what format Google is actually already rewarding for that specific query — a blog post can be well-researched and well-written and still structurally lose to a SERP dominated by interactive tools, and catching that mismatch before drafting saves the actual writing effort rather than diagnosing the failure after publication when traffic doesn't materialize. Banning "be more comprehensive" as a differentiation answer matters because it's the single most common non-answer a model gives when asked what's missing from a set of ranking pages — it sounds like actionable advice but names nothing a writer could actually go implement, whereas requiring a specific named element forces the model to actually compare the given result descriptions against each other rather than defaulting to generic completeness advice. The final go/reconsider/no-go framing, rather than an openended discussion, respects that this prompt is meant to inform a real resourcing decision before writing time is spent, and a content lead needs a clear recommendation to act on, not just an analysis to interpret.
What you get back
Phase 1: SERP is mixed — two calculator tools, one forum thread, one blog post, suggesting Google is satisfying multiple sub-intents (some want an instant number, some want community discussion, some want the underlying logic). Phase 2: intended blog-post format does match the one blog post already ranking at position 3, so format isn't a blocker. Phase 3: none of the current top results explain how to adjust the rate formula for different tax situations (W-2 vs 1099) — that's a specific, missing angle. Recommendation: go, with the tax-situation adjustment as the differentiating section.
Verified against
ChatGPT GPT-5.1 · 2026-08-11
Changelog
- 2026-08-11 — 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
