Content Decay
- Content decay is the gradual erosion of a page's visibility as it ages and rivals refresh.
- In the AI layer decay is invisible to rank tracking — it shows as falling citations, not falling position.
- Recency is a maintained signal: date, verify, and re-publish decision-critical pages on a schedule.
- Not all decay is refresh-reversible — stronger competitors and shifted intent need more.
Consensus definition
Content decay is a well-established SEO phenomenon: pages that once ranked and drew traffic lose ground as their information dates, fresher competitors appear, and search intent drifts.1 The classic remedy is the refresh — update the facts, re-publish, recover the ranking. What changes in an AI-mediated regime is the signal: decay no longer shows up cleanly as a ranking drop, because retrieval and AI answers favour current, well-maintained sources when they assemble a response. A page can hold its blue-link position and still decay out of the AI answers that now sit above it — invisible decay that classic rank tracking misses entirely.
rhinegold operator refinement
Rhinegold treats content decay as a two-layer problem now. The classic layer (ranking and click erosion) is still real and still fixed by disciplined refreshing. The new layer is decay in the AI surface: as models and retrieval corpora update, stale pages are passed over in favour of current ones, and the loss appears as falling Citation Rate or Mention Rate rather than falling position. This makes recency a signal in its own right, and it changes the maintenance calculus: the pages most worth refreshing are not only the high-traffic ones, but the ones that anchor the brand in decision-relevant answers. The trap is measuring decay only through GSC clicks and concluding a page is healthy when it has already gone quiet in the AI layer.
Operational use
Audit on two axes: classic decay (declining clicks/impressions and slipping rank over a trailing window) and AI-layer decay (declining citation or mention presence on the page's topics). Prioritise refreshes by decision-relevance, not just traffic, and treat update cadence as a maintained signal — date, verify, and re-publish decision-critical pages on a schedule rather than once.
Measurement boundary
Classic decay is cleanly measurable in GSC and rank tools; AI-layer decay is not — it requires tracking citation and mention presence over time, which is noisier and provider-specific. Decay is also not always reversible by refreshing: where loss is driven by a genuinely stronger competitor or a shifted intent, an update alone will not recover it.
Distinct from
Against a one-off ranking drop from an algorithm update — decay is gradual erosion, not a step change. Against Semantic Anchoring, its temporal counterpart: anchoring is how durably a brand sits in model knowledge, decay is how that standing erodes without maintenance. Against pruning a page deliberately — decay is unwanted loss, pruning is a chosen removal.
Observed pattern in practice
Common mistakes
- Measuring decay only by GSC clicks — a page can look healthy while going quiet in the AI answer layer.
- Refreshing by traffic rank alone — decision-relevant pages that anchor AI answers may matter more than high-traffic informational ones.
- Assuming every refresh recovers the loss — decay driven by a stronger competitor or shifted intent needs more than a date change.
Where consensus is missing
Classic content decay is well documented; AI-layer decay is an emerging observation with no standard metric, no agreed window, and no benchmark for how fast AI presence erodes relative to classic rankings. Whether recency is a direct ranking/retrieval factor or a correlate of other quality signals is unsettled.
Sources & deeper reading
- rhinegold Insights, Episode 05 — Content Architecture for LLM Authority
- rhinegold two-layer decay audit — classic GSC erosion plus AI-presence erosion tracking, shared with engaged clients and partners.