A diagnostic guide to the real causes, in the order worth
checking them. Then how to fix the ones that are fixable.
Diagnose first
Site-wide or page-level? The answer changes everything you do next
Decay
The slow, common one
Drift
The invisible one
Technical
The fixable one
Before chasing causes, find the shape of the drop. It splits the whole diagnosis in two
A sudden, site-wide drop — most or all pages falling together on a specific date — points to something systemic: a Google algorithm update, a technical catastrophe (accidental noindex, robots.txt block, botched migration, manual penalty), or a tracking error making a drop look worse than it is. Line your traffic chart up against known update dates and your deploy history first; if the cliff matches a date, that's your lead.
A gradual, page-by-page decline — different articles slipping at different times, adding up over months — is the far more common pattern for a mature blog, and a different animal entirely. This isn't one event; it's content decay and drift accumulating across the library, one article at a time, until the totals finally show it. Nearly 60% of posts lose their rankings within 12–24 months of publication [Draft.dev, 2025] — for most established sites, this slow erosion is the real story, not a single bad day.
THE USUAL SUSPECTS, IN ORDER
Roughly in the order worth checking. Some you fix in an afternoon; some are a standing discipline; some aren't a content problem at all — and we'll say so.
Technical & indexing (check first)
Accidental noindex, blocked resources, broken canonicals, a migration that dropped redirects, Core Web Vitals regressions. Fast to rule out, catastrophic if missed. A site crawler is the right tool here.
Algorithm updates (check the dates)
A confirmed update near your cliff means the bar moved — usually toward quality, depth, and freshness. The response is content quality work, not a quick fix. No tool 'recovers' you; better content does.
Content decay & ICP drift (the slow bleed)
Articles staling as intent shifts, competitors update, and your own positioning moves past them. The most common cause for mature libraries — and the one that hides, because pages can drift while still ranking.
Sometimes traffic isn't falling — but it stopped being worth anything
There's a decline that never shows up as a decline: the article still ranks, still pulls traffic, but the visitors are the wrong ones. Your ICP moved upmarket; the page keeps ranking for the SMB queries you no longer serve. The traffic line looks fine while the qualified traffic quietly collapses — and no rank tracker or analytics dashboard flags it, because by their metrics nothing broke.
This matters even when your totals look healthy, because the library is your pipeline engine: 92% of HubSpot's monthly blog leads came from older posts [HubSpot]. If those older posts drifted toward the wrong audience, your traffic can hold while your pipeline contribution falls — the most expensive decline there is, and the hardest to see. Diagnosing it requires checking every article against your current ICP, not just its rankings.
Draftcamp addresses the content-side causes. We won't pretend it touches the others
The questions people search mid-panic.
Book a 30-minute demo — connect GSC and the audit shows you, page by page, what's decayed and what's drifted, with the reasoning per article.
✓ 30 minutes ✓ Page-level diagnosis ✓ We'll tell you if it's not a content problem