Content Refresh and Decay Audits
A page losing traffic and a page having a quiet month look identical in a single screenshot. Ivan Sp tracks pages losing search visibility over time and flags them before the decline becomes a traffic problem. A page that decayed slowly over six months is a cheaper fix than the same page after it has fallen out of the results entirely.
Reading the Trend, Not the Snapshot
A single month of data cannot separate genuine decay from normal fluctuation. The audit tracks impressions, clicks, and average position over a rolling window. A page gets flagged only once the trend is consistent, not off a one-month dip.
Pages flagged for volatility at the topical-map stage, covered under topical maps and pillar-cluster structure, get checked on a tighter schedule. Pricing and regulation content decays faster than evergreen material.
A Low-Cost Fix Before a Full Rewrite
Some underperforming pages have fine underlying content and a heading structure underserving it. Google’s indexing process reads headings as an early signal of what a page covers. One generic heading gives it little to work with, regardless of how good the paragraph below actually is.
Adding more specific, unique headings, strictly grounded in the existing body text, is a lower-cost fix than a full rewrite. New claims never get invented to fill a heading. This suits pages where the content itself has not actually gone stale.
When the Content Itself Has Gone Stale
A heading fix will not help a page whose facts are simply out of date. Pricing changes, regulation updates, and statistics carrying an as-of date need the actual figures corrected, not just the framing around them.
The decay audit separates the two causes before recommending a fix. Applying a heading fix to a factually outdated page wastes the effort without addressing the actual problem.
Confirming the Refresh Worked
A refreshed page gets tracked against its own pre-refresh trend line, not against an assumption that the update fixed things. Search Console data over the following weeks confirms whether the trend actually reversed, or only leveled off.
A leveled trend with no genuine reversal usually means the decay had a cause the refresh did not address. That routes back into a fuller diagnosis rather than a second identical refresh.
Where the Standard Refresh Advice Breaks Down
Do you know why your declining pages are declining, or only that they are?
- The usual advice
- Refresh old content regularly, since Google favours recently updated pages.
- What holds up in practice
- Changing a date without changing the content signals nothing and can waste a recrawl. Freshness matters for queries where recency is genuinely part of the answer, and matters very little for a stable how-to page.
Picture a team updating publication dates across 200 archived posts in a single afternoon. Every page now reads as current.
Nothing in the content changed, and the pages that were declining continue declining. The only measurable result is 200 URLs recrawled for no reason.
I identify why each page is declining before touching it. Stale facts, a stronger competitor, and a shifted intent all need different work, and only one of them is a rewrite.
A date change is the cheapest thing to do and the least likely to help. The pages that needed real work stay untouched while the report shows 200 updates.
A page that has quietly lost half its impressions over eighteen months still costs you hosting and attention. It now earns a fraction of what you paid for it.
Four Things Commonly Assumed About Content Decay
Which of these is behind your current refresh schedule?
Ivan Sp meets these four assumptions whenever refreshing has become a calendar task.
| Commonly assumed | What actually holds |
|---|---|
| Updating the date makes a page fresh. | Freshness is a property of the content. A changed date with unchanged content signals nothing and consumes a recrawl. |
| All old content decays. | Stable topics hold position for years. Decay concentrates in pricing, regulation, tooling, and anything year-dated. |
| A traffic drop means the content got worse. | The content usually stayed the same while a competitor published something more complete, or the intent behind the query shifted. |
| Refreshing means rewriting. | Some pages have sound content buried under headings that describe nothing. A heading rewrite is a fraction of the cost. |
| One month of data shows the trend. | A single month cannot separate decay from ordinary fluctuation. Twelve months against the prior twelve is the honest comparison. |
Every month a decaying page goes undiagnosed is a month of compounding loss on an asset you already paid to create. The diagnosis costs an hour.
How Decay Is Measured
- Minimum window
- 12 months, so a seasonal dip is not mistaken for decline.
- Data source
- Search Console, which retains roughly 16 months of performance data.
- The decay signal
- A sustained downward trend in impressions or position across several months.
- Not a decay signal
- A single low month, or a drop matching the same month a year earlier.
- Common causes
- Stale facts, a competitor publishing something more complete, and intent drift.
- Cheapest intervention
- A heading rewrite, where the content is sound and the structure buries it.
- Most expensive intervention
- A full rewrite, warranted where the underlying facts have aged out.
- Verification
- The refreshed page compared against its own pre-refresh trend line.
A Worked Example: Four Declining Pages, Four Causes
Consider four pages all down roughly 40 percent year on year. The diagnosis below is illustrative, written to show why one intervention never covers a set. No client engagement is described here.
| Page | What actually changed | Correct intervention |
|---|---|---|
| Pricing guide | The figures aged out | Full content update |
| How-to article | Nothing. Headings describe nothing | Heading rewrite only |
| Regulation summary | The regulation changed | Rewrite, then a scheduled review |
| Comparison page | A competitor published something deeper | Expand coverage against the new bar |
| Seasonal landing page | Nothing. The comparison window was wrong | No action |
I diagnose before intervening, because the last row needed nothing at all. A blanket refresh would have spent budget on it.
Refreshing every old page is repainting the whole house because one room has damp. The paint is real work, and the damp is still there.
Only two of the five rows need a rewrite. The other three need a heading pass, a schedule, or nothing, which is a very different quote.
How a Decay Audit Runs
I establish the trend before diagnosing the cause, since one month proves nothing either way.
- Export 16 months of page-level data from Search Console, including impressions and average position.
- Compare each page against the same period a year earlier rather than against last month.
- Flag pages showing a sustained decline across at least three consecutive months.
- Check each flagged page for factual content that has aged, such as prices, dates, or regulations.
- Search the target query and read the pages now outranking it, noting what they cover that it does not.
- Read the page’s heading outline alone, since buried structure is a common and cheap cause.
- Check whether the query’s intent has shifted, which shows as a changed ranking URL mix.
- Assign each page the cheapest intervention that addresses its actual cause.
- Record the pre-refresh trend line before making any change.
- Compare the page against its own recorded trend line 60 to 90 days after the refresh.
How This Engagement Is Scoped
I scope decay work by the number of declining pages rather than by archive size.
- What arrives
- A ranked list of declining pages, the diagnosed cause per page, and the assigned intervention.
- Typical duration
- Two to three weeks for diagnosis across a few hundred pages.
- What the client supplies
- Search Console access, and subject input where facts need updating.
- Included
- Trend analysis, cause diagnosis, intervention assignment, and a post-refresh comparison.
- Not included
- Writing the refreshed content, quoted separately once the causes are known.
- What changes the estimate
- Large archives, several service lines, or heavily regulated content.
- How it is quoted
- After a short scoping call, once the archive size is known.
Tools and Primary Sources
The references below supply the trend data and the freshness context. Each opens in a new tab.
- Search Console Performance report supplies the 16-month window a decay comparison is built on.
- Creating helpful, people-first content describes the standard a refreshed page still has to meet.
- Google guidance on removing outdated content covers the options where a page is beyond refreshing.
- Google Trends separates a topic in genuine decline from a page losing share within a stable topic.
Diagnose One Declining Page
I suggest a comparison that takes about ten minutes.
- Open Search Console, then Performance, and select the last three months.
- Turn on the compare option and choose the same three months a year earlier.
- Open the Pages tab and sort by the largest impression decline.
- Open the worst page, then search its main query and read what now outranks it.
If the pages above it cover something yours does not, the cause is competitive rather than technical. If nothing has changed above it, the cause is usually the page’s own structure.
Common Questions
How is decay different from a sudden ranking drop?
A sudden drop usually traces to a specific event, such as an algorithm update or a technical fault. Decay is a gradual decline over months, more often tied to a competitor improving or the content simply aging.
How often should a site get checked for decay?
Monthly for larger sites with meaningful traffic at stake, quarterly for smaller sites where month-to-month fluctuation is harder to distinguish from a genuine trend.
Does every declining page need a full rewrite?
No. Many pages recover with a targeted heading fix or a factual update instead. The audit’s diagnosis step decides which fix a given page actually needs.
How often should content actually be refreshed?
On a schedule set by the topic rather than by the calendar. Pricing, regulatory, and year-dated pages need a review date at publication. Stable how-to content can hold position for years without any intervention.
Discuss This Work
The first call is a free 30-minute consultation with no pitch attached. You receive an honest assessment of where the site stands and what can realistically be built.