Most content retainers are priced by output: so many posts a month. That pricing model quietly shapes the strategy. New posts are billable and visible. Fixing the post that ranked well two years ago and has been slipping ever since is neither, so it rarely gets done.
Yet a page that already has links, history and some rankings is usually the cheapest traffic you will ever win back. This is a method for finding those pages, diagnosing what went wrong, and deciding which ones deserve a refresh before a single new brief is written.
Step one: find the pages that are sliding
Open the Search Console performance report, switch to the pages view, and compare the last three months against the same three months a year earlier. Year-on-year comparison removes most seasonality. If your site is too young for that, compare against the previous three months and keep seasonality in mind.
Sort by the change in clicks, largest loss first. Ignore pages that never had meaningful traffic. You are looking for pages that used to work and now work less. Export the top of that list, usually a few dozen URLs is plenty to start.
Step two: work out which kind of decay it is
Not all decay is fixable, and the fix depends on the cause. For each page, look at clicks, impressions, CTR and average position together. Four patterns cover most cases:
- Position fell, impressions fell with it. Something outranked you. This is the classic refresh candidate. Look at what now ranks above you and ask what it answers that your page does not.
- Position held, CTR fell. You still rank, but fewer people click. The results page changed around you: an AI overview, a new feature, or competitors with sharper titles. Rewrite the title and description before touching the body.
- Impressions fell, position held. Fewer people are searching. The topic is cooling or the terminology moved. A refresh will not create demand that is not there. Update the language if buyers now use different words, otherwise leave it.
- Several of your own pages trade places for the same query. That is cannibalisation. Two half-strong pages lose to one strong one. Merge them.
Also open the page and read it as a buyer would today. Outdated screenshots, old pricing, a product name you retired, advice that stopped being true: these hurt conversion even where rankings hold.
Step three: decide keep, refresh, merge or retire
Give every page on the list one verdict:
- Refresh when the topic still matters to buyers and the page lost ground to better answers.
- Retitle only when the ranking held and only click-through dropped.
- Merge when two or more pages compete for the same intent. Combine the best parts into one URL and 301 redirect the others to it.
- Retire when the topic no longer relates to what you sell. Redirect to the closest useful page, or remove it if nothing fits.
Prioritise by commercial value, not just lost clicks. A post that feeds demo requests is worth more than a post that once went viral with the wrong audience.
Step four: refresh properly, not cosmetically
Changing the date and adding two paragraphs is not a refresh, and readers can tell. A real refresh usually means:
- Re-matching the search intent. If the results now favour a step-by-step guide and yours is an opinion piece, restructure it.
- Replacing outdated facts, examples, screenshots and references. Check every claim you keep.
- Putting the direct answer near the top.
- Adding internal links from newer, related pages to the refreshed one, and from it to your money pages.
- Rewriting the title and meta description for the query the page actually ranks for.
- Updating the published or modified date only when the content genuinely changed.
Keep the URL. The page's history and links live there.
Step five: measure against the page's own before-state
Record the baseline for each refreshed page on the day you ship: clicks, impressions, CTR, position and conversions. Check again once the page has been re-crawled and has had time to settle. Compare the page to itself, not to a sitewide average that other changes are also moving. A running log of refreshes and their outcomes becomes the best evidence you have for where content budget should go next.
When new content is the right call
None of this means stop writing. It means earn the right to. Once the decaying pages worth saving are fixed, new content should fill real gaps: questions buyers ask that no page of yours answers. A good rule is that every new brief should state why an existing page could not be refreshed to do the job instead.
How the crew runs this
Three crews share this job in Spryxa. The Search Crew scans keyword movement every week and flags pages losing ground in its search opportunity brief, using Search Console data. The Measurement Crew runs a daily pulse and flags anomalies, such as a page whose clicks drop sharply, with a note on who should follow up. It reports; it does not change anything itself.
The Content Crew then drafts the refresh in your brand voice, with a rationale and staged edits to the existing page where the connection allows. A person reviews the draft, every claim in it and the publishing destination before anything goes live. The Content Crew is measured on drafts accepted, edited and scheduled, and on the traffic or lead movement later attributed to them, so a refresh that does not move the page shows up as exactly that.
Connect Search Console first; everything above depends on it. The how-to hub has the step-by-step versions of these jobs.
Start this week
Run the year-on-year pages comparison in Search Console, take the five biggest losers that still relate to what you sell, and give each one a verdict. That hour of work will usually produce a better quarter than the next five new posts on the calendar.
If you want to see which of your pages are sliding and why, start with the free audit.