The fastest SEO wins usually sit in content you already have. A content refresh strategy uses Search Console to find pages losing clicks and queries stuck just off page one, decides whether each page should be refreshed, merged or removed, and then makes substantive updates. Changing the date and a few sentences is not a refresh.
What content decay looks like in Search Console
Decay rarely looks like a cliff. A sudden drop across many pages usually points to a technical problem, a migration or an algorithm update. Decay is slower: a post peaks, then loses clicks a little every month as competitors publish better pages and the topic moves on.
Open the Performance report, switch to the Pages tab and compare the last three months with the previous three. For seasonal topics, compare against the same months a year earlier instead. Search Console keeps 16 months of data, so export regularly if you want a longer history. Then read the pattern:
| Pattern | What it usually means |
|---|---|
| Impressions steady, clicks falling | A click-through problem: stale titles, stronger competitor snippets or more SERP features above you |
| Position and impressions both falling | Competitors now have better or fresher pages for the same queries |
| Position steady, impressions falling | Demand for the topic is shrinking; content changes won’t fix that |
| Clicks for a query moving to another page on your site | Two of your pages are competing with each other |
| Losing some queries, gaining unrelated ones | The page has drifted, or search intent has shifted |
The fourth pattern needs a different fix than a refresh. Keyword cannibalization covers how to confirm it and which page to keep.
Building the refresh list: traffic trend, position and conversion value
Export page-level clicks, impressions and average position for both periods into a sheet. Then add what Search Console can’t tell you: which pages bring leads, signups or sales. Pull conversions by organic landing page from GA4 or your CRM.
Score each page on three things:
- Lost clicks, in absolute numbers, not percentages. A page that fell from 2,000 to 1,600 monthly clicks lost more than one that halved from 40 to 20.
- Business value, rated 1 to 3. Does the page convert, support a commercial page or attract the buyers you sell to?
- Effort, rated 1 to 3. A title rewrite is a 1; a full rebuild with new research is a 3.
Priority is lost clicks multiplied by business value, divided by effort. Keep the top 10 to 20 pages for the first cycle. A small list you actually finish beats a large one that stalls halfway.
Quick wins: pages ranking in positions 5 to 20
Decay is only half the list. The other half is near-miss queries: searches where you already rank, but not high enough to get real traffic.
In the Performance report’s Queries tab, export the table (or use the table filter on Position), keep queries with an average position between 5 and 20 and sort them by impressions. Average position is blended across every impression, so treat it as a signal, not an exact rank. Click a query, then open the Pages tab to see which URL is earning it.
You will usually find one of four gaps:
- The query’s exact framing never appears in a heading or the opening lines.
- The page answers a neighboring question, not this one.
- A subtopic the ranking pages cover is missing entirely.
- Few internal links point to the page, so it looks unimportant.
Positions 5 to 10 are often a click-through problem as much as a ranking one, so rewrite the title and meta description around the query’s wording. Positions 11 to 20 usually need more depth and stronger internal links before the title matters.
Refresh, merge, redirect or delete: a decision tree
Not every declining page deserves more work. Run each page on the list through these questions in order:
- Does it earn anything? If it has no meaningful traffic, no backlinks, no conversions and targets a topic you no longer care about, remove it.
- Does another page on your site serve the same intent? Merge the weaker page into the stronger one and 301 redirect the old URL.
- Is the topic still relevant to your buyers? If not, but it has backlinks, redirect it to the closest relevant page.
- Is the format still right for the query? If the results page now wants something different, rebuild the page on the same URL.
- Is it on target but outdated or thin? Refresh it.
| Situation | Action | Watch out for |
|---|---|---|
| On-topic page, outdated or thin | Refresh on the same URL | Keep the URL; don’t start over on a new slug |
| Two pages chasing the same query | Merge into the stronger URL, 301 the other | Move the useful sections, not just the redirect |
| Off-topic page with backlinks | 301 to the closest relevant page | Redirecting everything to the homepage wastes the links |
| Off-topic page, no links, no traffic | Delete and return a 404 or 410 | Remove internal links to it and drop it from the sitemap |
| Right topic, wrong format for current results | Rebuild on the same URL | Treat it as a new brief, not an edit |
What a real refresh includes and what is only cosmetic
When I audit content programs, the refreshes that fail usually added words instead of answers. Use this checklist before calling a page done:
- Search the primary query again and note the format, angle and depth of the top results
- Put the direct answer in the first two or three sentences
- Add subtopics the ranking pages cover and your query data shows people want
- Replace outdated facts, product names, interface steps, screenshots and prices
- Cut sections that no longer earn their place, including padded intros
- Add what only you can say: real examples, process detail, decision criteria
- Rewrite the title tag and meta description around the queries the page actually earns
- Fix broken links and add internal links in both directions
- Check structured data, including the modified date
Cosmetic changes don’t count: moving the date forward, swapping synonyms, adding a filler paragraph to raise word count, or reordering sections without changing what they say. Search engines compare your page with the alternatives, and a page that answers the query no better than before has no reason to move up.
Updating for shifts in search intent
Sometimes the page didn’t get worse; the query changed. A search that once returned definitions now returns comparison lists. A how-to query now shows templates and tools. Nothing on your page is wrong, it just isn’t what searchers want anymore.
Check for intent shift in two places:
- The results page. Search the primary query in a private window. Look at the format of the top five results, the angle their titles take, and which features appear, such as featured snippets, video, forum threads or product listings.
- Your query mix. If Search Console shows the page picking up queries it doesn’t address, searchers are telling you what they expected to find.
Then decide. If the new intent still fits the page’s topic, rebuild the page to match it. If it doesn’t, let the page target the queries it can win and create a separate page for the new intent, with a clear internal link between them.
Republishing, dates and internal links
How you republish matters almost as much as what you change:
- Keep the URL. A new slug means a redirect, lost history and a slower recovery.
- Be honest with dates. Keep the original published date and show an updated date when the changes are substantial. Update the modified date in your structured data and sitemap to match. Don’t move dates for cosmetic edits.
- Ask for a recrawl. For priority pages, use URL Inspection in Search Console to request indexing, then confirm the new version was crawled.
- Send links to it. Add links from newer, related posts and relevant service pages, using descriptive anchor text. An internal linking strategy makes this routine instead of an afterthought.
- Clean up after merges. Update internal links that point to redirected URLs so they go straight to the destination.
Measuring refresh results
Keep a change log with the URL, the date, what changed and the target queries. Without it, you won’t know six weeks later which edit caused which movement.
Compare the four weeks after the page was recrawled with the four weeks before the change, and hold a few similar untouched pages as a control. If the control group rose by the same amount, the gain came from seasonality or an algorithm shift, not your edit.
| Metric | Where to find it | What you want to see |
|---|---|---|
| Clicks and impressions for target queries | Search Console, filtered by page | Growth versus the pre-refresh period and the control group |
| Average position for target queries | Search Console | Near-miss queries moving onto page one |
| Click-through rate | Search Console | Higher CTR at a similar position after a title rewrite |
| Organic conversions | GA4 or CRM, by landing page | Leads or sales rising with traffic, not just visits |
Give each page six to eight weeks from the update before judging it. If nothing moved, check in order: whether the new version is indexed, whether the page matches current intent, and whether competitors simply have more authority. The last one needs links, not more editing.
Running this cycle every quarter is a standing part of how I handle technical SEO and content systems: one prioritized list, real updates and results measured against a control.
Get it built
If your blog has plenty of posts and flat organic traffic, I can build the refresh list and ship the updates with you. Start with a Growth Audit, $1,500 fixed and credited if we continue. See pricing or get in touch.