Key takeaways
- Positions 11–30 with rising impressions are your priority list — nothing else comes close on cost per click gained.
- Diagnose before you edit: intent mismatch, thin coverage, cannibalisation and stale data need different fixes.
- Merging two mediocre pages usually beats improving both.
- Change the publish date only when the change is substantial, and always update the modified date.
- Measure at 30 and 60 days; if nothing moved, the diagnosis was wrong, not the effort.
Build the priority list in twenty minutes
Open Search Console, set the range to the last three months, and export queries with their pages. Then filter for:
- Average position between 11 and 30
- At least 300 impressions in the period
- CTR below the benchmark for that position
Sort by impressions. The top twenty rows of that sheet are your refresh backlog, ordered by opportunity. You now know more about where your next quarter of traffic is coming from than most teams learn in a month of planning.
Add two columns before you start work: current conversions from that page, and whether the page has backlinks. Both change what you are allowed to do with it.
Diagnose before you touch the copy
Refreshing without diagnosis produces longer pages that rank exactly where they did before. Five causes account for nearly everything we see:
| Symptom | Likely cause | Fix |
|---|---|---|
| Ranks for the term but the SERP is a different format | Intent mismatch | Rewrite to the format that ranks, or retarget the page |
| Covers fewer sub-topics than every page above it | Thin coverage | Expand with the missing sections and questions |
| Several of your URLs appear for the same query | Cannibalisation | Merge into one page, redirect the rest |
| Data, screenshots or examples are two years old | Staleness | Update facts, visuals and examples; refresh the modified date |
| Good position, poor CTR | Title and description problem | Rewrite the title tag and meta description; add schema |
Write the diagnosis down for each page before editing. It takes two minutes and it is what makes the 60-day review meaningful.
Decide: keep, refresh, merge or remove
Sort every candidate into one of four buckets:
- Keep — performing adequately, no action beyond a factual check.
- Refresh — close to the first page, sound intent, fixable gaps. The majority of your work.
- Merge — two or more pages competing for one intent. Combine the strongest material into the page with the best links and history, then 301 the others.
- Remove — no traffic, no links, no conversions, no strategic role. Redirect to the nearest relevant page, or let it 410 if there is genuinely nothing relevant.
Merging is the most under-used option. Two pages in positions 18 and 24 combined into one strong page routinely land in the top eight, because you have concentrated the links, the coverage and the engagement signals rather than splitting them.
How to actually do the refresh
A refresh is not an edit pass. Work in this order:
- Re-read the SERP. What do the top five cover that you do not? What format do they use? What has changed since you published?
- Restructure first. Fix the heading hierarchy and section order before writing new copy — often the content exists but is buried.
- Add the missing sections, especially People Also Ask questions your page ignores.
- Cut ruthlessly. Dated context, throat-clearing intros, padding written to hit a word count. Most refreshed pages should lose 15–25% of their original words.
- Update every fact and visual, and re-link statistics to their primary sources.
- Rewrite the title tag and meta description with today's SERP in front of you.
- Fix internal links — in and out. Add links from your three most recent relevant posts.
- Re-submit in Search Console and re-promote.
A worked example
A B2B SaaS client had a 2023 post targeting a term with about 2,400 monthly searches. It sat at position 17 with 4,100 impressions a quarter and 38 clicks.
The diagnosis was mixed: thin coverage (six sections against the leaders' eleven), two stale statistics, and a title tag that read like a chapter heading. It also had four backlinks, which meant deletion was never an option and merging was unnecessary.
The refresh took a writer six hours: five new sections answering PAA questions, a comparison table, two updated statistics, a rewritten title, 400 words of padding removed, and six new internal links from recent posts.
| Metric | Before | Day 30 | Day 60 |
|---|---|---|---|
| Average position | 17.4 | 9.1 | 5.8 |
| Impressions (28 days) | 1,380 | 2,940 | 4,620 |
| Clicks (28 days) | 12 | 96 | 287 |
| Demo requests | 0 | 2 | 7 |
Six hours of work. Compare that with commissioning a new 2,000-word article and waiting four months to find out whether it ranks.
Pruning: when less content means more traffic
Large archives accumulate pages that do nothing: event recaps from 2019, three-paragraph news posts, duplicate category descriptions, tag pages with one entry.
They rarely rank, but they dilute your internal linking, consume crawl budget on big sites, and make your archive expensive to maintain. Pruning them — carefully — tends to help.
The safe process: list every page with zero organic sessions in twelve months; remove from that list anything with backlinks, conversions, offline value or seasonal traffic; redirect the remainder to the closest relevant page; and let pages with no relevant destination return 410. Then wait sixty days before pruning the next batch, so the effect is measurable.
Make the refresh cycle permanent
The teams that compound are the ones that treat refreshing as a standing commitment rather than an annual clean-up.
What that looks like in practice: a quarterly export of the 11–30 query list; two refreshes shipped every month alongside new pieces; a rule that no page goes eighteen months without a factual review; and a simple dashboard showing refreshed pages and their before/after positions, so the work stays visible to whoever approves the budget.
If you would rather hand it over, our content audit and optimisation service does exactly this — inventory, diagnosis, decisions and the rewrites themselves.
Frequently asked questions
Run the audit query quarterly and refresh continuously. In practice most teams should spend roughly a third of their content budget on refreshes and two thirds on new pieces, once the archive reaches about fifty pages.
Update the modified date always. Change the publish date only when the piece has been substantially rewritten — more than half new. Re-dating a lightly edited post is the kind of trick that erodes reader trust without gaining much.
Deleting pages with no traffic, no backlinks, no conversions and no strategic role usually helps. Never delete a page with backlinks — redirect it to the closest relevant page instead, so the equity survives.
Pages in positions 11–20 often move within two to four weeks of re-indexing. Deeper pages take longer and may need a rewrite rather than an edit.
Yes, and it is the step most teams skip. Re-share it, email it to your list, add fresh internal links from recent posts, and request re-indexing in Search Console.
Daniel Okafor
Senior Editor & SEO Lead
Daniel runs SERP analysis, content audits and the 30/60/90-day performance reviews. If a page is stuck on page two, he is the one who works out why.
