Verified against ChatGPT · 2026-08-12
Diagnose why a once-ranking page has decayed and produce a refresh plan that fixes the actual cause, not a generic rewrite
Reads a page's traffic-decay pattern alongside its current content and separates the real causes of decline — staleness, SERP feature displacement, a stronger competitor, or genuine irrelevance — into a targeted refresh plan instead of a blanket rewrite.
The prompt
Ready to copy — highlighted parts are example details you can swap.
This page used to rank well and has lost traffic. Diagnose why before recommending a refresh, since a refresh built on the wrong diagnosis wastes the work. PAGE CONTENT (current) Guide to 'best video conferencing software 2023', last updated 14 months ago, references pricing and features that have since changed for two of the five tools listed. TRAFFIC / RANKING PATTERN Ranked position 3 for 8 months, dropped to position 9 over the last 6 weeks, impressions flat, clicks down 60%. TARGET KEYWORD best video conferencing software WHAT'S CHANGED IN THE SERP, IF YOU KNOW An AI-generated overview box now appears above organic results for this query, summarizing top options directly. DIAGNOSTIC PASS Consider each of these possible causes against the evidence given, and state which one(s) actually fit rather than assuming it's always "the content needs updating": (a) content staleness — dates, statistics, or product details are visibly outdated relative to what the query now expects; (b) SERP feature displacement — a featured snippet, AI overview, or new SERP element now sits above where organic results used to get clicks, meaning the ranking itself may be stable but the click-through opportunity shrank; (c) a specific competitor page objectively out-covers this one now in a way it didn't before; (d) genuine topic irrelevance — the query's actual meaning or dominant intent has shifted since this page was written, and the page answers a version of the question that's no longer what's being asked. Say explicitly which of these the evidence supports, and resist defaulting to "just refresh the content" if the real cause is (b) or (d), since those require a different fix than an on-page rewrite. REFRESH PLAN Based on the actual diagnosis, not a generic list, propose the specific fix: if it's staleness, name exactly what needs updating and why the current version reads as dated. If it's SERP displacement, propose how to compete for the new SERP feature itself rather than just refreshing prose that ranks organically below it. If it's a stronger competitor, name the specific gap using their actual coverage as reference. If it's genuine intent drift, be honest that this may need a substantially different page rather than a refresh, and say so plainly rather than recommending a light edit to a page whose fundamental premise no longer matches the query. OUTPUT FORMAT Diagnosis section naming the supported cause(s) with evidence cited. Refresh plan matched specifically to that diagnosis. If intent has genuinely drifted, a clear recommendation that this is a rebuild, not a refresh.
Customize
Optional — swap in your own details for the highlighted parts above.
Why this works
The default failure mode for a content-refresh request is treating every case of traffic decline as a staleness problem solvable by updating dates and prose, because that's the simplest and most common refresh pattern in the training data — this prompt structure forces the model through an explicit differential diagnosis across four distinct causes before it's allowed to recommend a fix, which matters because staleness, SERP feature displacement, competitive out-coverage, and genuine intent drift each require a materially different response, and applying a content refresh to a page whose real problem is an AI overview or featured snippet sitting above it will not recover the lost clicks no matter how well the prose is updated. Explicitly including SERP feature displacement as a candidate cause matters specifically because this is the category most likely to be missed by a naive "refresh the content" instinct — the organic ranking itself can be perfectly stable while a new on-SERP element eats the click-through opportunity above it, and the fix for that (structuring content to compete for the feature itself, or accepting a lower click ceiling) is completely different from a content update. Requiring the model to name intent drift as a possible diagnosis, and to recommend a rebuild rather than a refresh when the evidence supports it, guards against the sunk-cost pull toward always saving the existing page — sometimes a page's fundamental premise no longer matches what the query means today, and a light edit to an outdated premise produces a page that reads as freshly updated but still doesn't answer the question being asked, which is a worse outcome than an honest rebuild recommendation.
What you get back
Diagnosis: evidence supports both staleness (pricing/feature details for two tools are outdated) and SERP feature displacement (an AI overview now sits above organic results, summarizing the same comparison this page provides) — the position-9 ranking may partly reflect Google deprioritizing a page it can detect as dated, but even a full content update likely won't recover the clicks the AI overview is now absorbing. Refresh plan: update all five tool listings with current pricing and features, and additionally restructure the top of the page as a scannable comparison table, since that format has a better chance of being pulled into the AI overview itself as a cited source.
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
