Use when a set of older pages is losing traffic and you need to decide which ones are worth the work.
organic-content-decay-triage.md
You are triaging pages whose organic traffic is falling, to decide which to refresh, merge or retire. You are not writing the replacement copy.
Declining pages (URL, topic, clicks by month, current position, last updated): {{DECAY_ROWS}}
Commercial value of each section: {{SECTION_VALUE}}
Refresh capacity this quarter: {{REFRESH_CAPACITY}}
What has changed in the SERPs for these topics: {{SERP_CHANGES}}
Output a ranked table, biggest drop against highest value first: Rank | URL | Decay shape (gradual, cliff, seasonal, intent shift) | Likely cause | Verdict (Refresh / Merge into another URL / Retire and redirect / Leave alone) | Effort (S / M / L) | The first change to make.
Then two sections:
- "Fits this quarter" - only the rows that fit {{REFRESH_CAPACITY}}, in order.
- "Leave alone" - pages where the fall is demand or seasonality rather than a page problem, with the evidence from {{DECAY_ROWS}}.
Constraints:
- Rank on value from {{SECTION_VALUE}} as well as size of drop. A large fall on a page nobody buys from is not a priority.
- Where {{SERP_CHANGES}} shows intent has moved to a format we cannot produce, recommend retire rather than refresh.
- Never recommend adding word count as a fix on its own.
- If a last updated date is missing, write "unknown" rather than inferring one.
Replace each placeholder with your own detail. The more specific you are, the less the model invents.
When do I use this rather than just refreshing whatever dropped most?
When the list of declining pages is longer than the refresh capacity you actually have. Ranking by size of drop alone spends the quarter on pages nobody buys from, which is why this one ranks on section value as well as the fall, then cuts the list down to what fits your stated capacity.
What do I need in front of me?
Monthly clicks per URL rather than a single percentage change, because a cliff and a slow slide have different causes and different fixes. Add the last updated date, the commercial value of each section, an honest capacity figure, and impressions if you have them, since flat impressions with falling clicks is a SERP problem.
What comes back, and which part matters?
A ranked table giving each URL a decay shape, a likely cause, a verdict of refresh, merge, retire or leave alone, and the first change to make. The two sections underneath earn their place: what fits this quarter in order, and the leave alone list where the fall is demand rather than a page problem.
What is the mistake that costs me here?
Acting on a merge verdict before checking link data. Redirecting a page that carries referring domains into a stronger URL is fine; retiring it because traffic fell is not, and you cannot undo it cheaply. The second mistake is answering a decay verdict with more word count, which the prompt refuses to recommend for good reason.