The Traffic You Lost Last Year Did Not Leave in a Single Day
Sudden drops get attention. Somebody notices, somebody investigates, somebody fixes it. Slow drops get rationalized. Two percent this month, three percent the next, and by the time the annual review arrives you are down thirty percent with no single event to point at.
That is content decay, and it is the most common untreated condition on established sites. Think of it as the slow version of what happens when you stop doing SEO entirely. It does not feel like a problem because no individual month looks bad. It only becomes visible when you compare a year to the year before it, at which point you have lost a great deal of ground that was expensive to acquire.
The good news is that decay is the cheapest problem in SEO to fix, because the pages already have authority. You are restoring something, not building it. At Sympler we consistently find refresh work outperforms new publishing on established sites, often by a wide margin.
What Causes Content to Decay
Decay is not one phenomenon. Diagnosing which kind you have determines the fix, and applying the wrong fix wastes the effort entirely.
Factual Decay
The content was accurate and is no longer. Interfaces changed, prices moved, a tool was discontinued, a regulation was updated. This is the most damaging kind, because a reader who spots one wrong detail discounts everything else on the page.
Competitive Decay
Your content did not change. Everyone else’s got better. You published the definitive guide in 2023 and three competitors published deeper ones since. Your page is exactly as good as it was and is now the fourth best resource available.
Intent Decay
The query still exists but what searchers want from it has shifted. A term that once wanted a definition now wants a comparison, and re-running your keyword research on it usually makes the shift obvious. Your page answers the old question well and the new question not at all. This one is easy to miss because your content looks fine in isolation.
Structural Decay
The page is fine, but its position in your site has weakened. Internal links pointing to it were removed during a redesign, or it fell out of navigation, or newer pages on similar topics are now competing with it for the same terms.
Why Decay Is So Hard to See
Decay hides in plain sight, and the reason is resolution. Month over month reporting is dominated by seasonality and noise, so a page shedding two percent a month never registers as an event. The decline only becomes legible when a full year is compared against the year before it, page by page, and almost nobody looks at their content that way on any regular basis.
Volume reporting hides it further. The pages losing the most ground in percentage terms are rarely the pages at the top of a traffic report, so the ones most worth fixing sit well below where anyone is looking.
The deeper difficulty is that very different problems produce an identical looking decline in a standard analytics view. A page that is losing the click and a page that is losing its ranking outright look the same from the top, and the fixes have nothing in common. Reading a decline correctly is what separates a refresh program that pays for itself from one that burns a quarter rewriting pages which did not need rewriting. It is also the part we are most often brought in to do, because the same data will happily support several contradictory conclusions.
One step no report will ever do for you is opening the page and reading it as a prospect would. Most causes are obvious within ninety seconds: a stale date, a dead screenshot, advice that no longer applies, a competitor’s version you would rather read yourself.
Not Every Declining Page Deserves Attention
The instinct on seeing a decline report is to start fixing from the top. That is how refresh programs consume a quarter and return very little.
A declining page has several possible futures. Some are worth rewriting where they stand. Some need to absorb a neighbouring page rather than keep competing with it. Some should hand their accumulated authority to a stronger page and step aside. A few should simply go, because they earn nothing, are linked from nowhere, and no longer describe anything the business does.
Choosing wrongly is expensive in both directions. Rewriting a page that should have been consolidated leaves you competing with yourself at greater length. Removing a page that still holds links throws away something you paid for and cannot easily buy again.
One principle holds across nearly all of it. Publishing a fresh URL and abandoning the old one discards the accumulated links and history, then asks the new page to earn all of it again from zero. That is a strange trade to make voluntarily, and it is among the more common unforced errors we find on established sites.
What a Real Refresh Involves
Changing the date and adding a paragraph is not a refresh, and search engines are not fooled by it. A refresh worth doing changes the substance of the page, applying the same standards as writing effective content for search in the first place.
In practice that means the page gets re-reported rather than re-worded. Claims are verified again from source. Gaps are identified against what readers now expect to find. The structure is reconsidered for how answers actually get extracted today rather than how they were written three years ago. The page’s relationships with the rest of the site are rebuilt in both directions. It is closer to producing a second edition than to touching up a draft, which is precisely why refresh work is so consistently underestimated when it gets scoped internally.
One rule deserves stating on its own: only update the modified date when the content genuinely changed. Sites that programmatically bump dates without changing anything are manufacturing a signal that means nothing, and they lose the ability to use dates meaningfully at all. It is a short term trick with a permanent cost.
Running It as a Program Rather Than a Project
One big cleanup feels productive and solves nothing permanently. Decay resumes the moment the project ends, and a year later the same report produces the same list with different page names on it.
What works instead is a standing cadence: a regular rhythm for surfacing candidates, a fixed and realistic amount of refresh work per cycle, and a periodic check that the pages you fixed actually recovered. The specifics have to be tuned to a team’s real capacity rather than its ambitions, and that is the part that usually goes wrong. An ambitious cadence gets skipped in the first busy month and never restarts.
The harder question is how to divide effort between refreshing what exists and publishing something new. On a mature site with substantial existing content, refresh work is usually where the return is, and it is chronically under-resourced for a straightforward organizational reason: new content is visible and easy to champion internally, while a recovered page looks like nothing happened. Both halves depend on the same content writing standards, and both benefit from the structural principles in our on-page SEO checklist.
The AI Search Angle
Decay has taken on a second dimension worth understanding.
Outdated content is not merely less likely to rank. It is less likely to be used as a source in a generated answer, and if it is used, it may propagate an outdated claim attributed to you. That is worse than being ignored, because a wrong answer with your name on it damages trust with someone who never visits your site to see the context.
Pages making time sensitive claims deserve tighter review cycles for exactly this reason. If your content states a price, a specification, or a current best practice, treat the review date as a commitment rather than a suggestion. How this interacts with your broader AI SEO program is one of the more underappreciated reasons to take refresh work seriously.
Frequently Asked Questions
How do I tell decay apart from an algorithm update?
Timing and breadth. Algorithm impacts hit many pages at once on a specific date. Decay affects individual pages on their own timelines. Plot your declining pages against a date axis. Clustering suggests an update, scattering suggests decay.
Should I keep the same URL when refreshing?
Almost always yes. The URL carries the links, history, and accumulated signals. Change it only if the existing URL is genuinely misleading about the content, and if you must, redirect properly with a single hop.
How much do I need to change for it to count?
Enough that a returning reader would notice. There is no percentage threshold. If someone who read the original would find meaningful new value, it is a refresh. If they would not, it is a date change.
Can I refresh a page that never performed well?
You can, but check the diagnosis first. A page that never ranked usually has a targeting or intent problem, not a freshness problem. Refreshing content that was aimed at the wrong query just produces a more current version of the wrong page.
How long until a refresh shows results?
Typically a few weeks for recrawling and reassessment, longer for competitive terms. If nothing moves after roughly two months, the diagnosis was probably wrong rather than the execution. Go back to the four decay types and reclassify.
Systemic or Isolated?
The most useful thing to establish early is whether your decay is confined to a handful of neglected pages or is running quietly across the whole library. Those are very different problems carrying very different budgets, and most teams assume the first while having the second.
If you would like help building the reporting and the standing cadence rather than running another one off cleanup, our team designs these programs as part of a broader content SEO engagement. Get in touch with Sympler and we will start with your decline report and show you what it is telling you.
