Most content programmes only add. The result is a library where a handful of posts earn everything and the rest quietly decay. Refreshing what already ranks is usually the highest-return work available.
Updating existing posts is generally more effective than publishing new ones once you have a reasonable library, because you are improving pages Google already trusts rather than asking it to evaluate more. Prioritise posts ranking in positions 5 to 20, and update the content genuinely before changing the date.
Why refreshing wins
- The page is already indexed and has accumulated whatever authority it has.
- It already has internal links and possibly external ones.
- You know from Search Console exactly which queries it appears for.
- Improving a page at position 8 to position 4 usually produces more traffic than a new post starting at zero.
- It costs less than writing something new.
New content is for topics you genuinely do not cover. Once the obvious topics are covered, the marginal new post is worth less than an hour spent on a post already earning impressions.
Choosing what to update
- Open Search Console, Performance, last three months.
- Filter to average position between 5 and 20. These are the near misses.
- Sort by impressions. High impressions with low clicks means the opportunity is real.
- Cross-check against the page: is it genuinely worse than what ranks above it?
- Also flag anything with declining clicks over six months, which usually means decay.
| Signal | What it usually means |
|---|---|
| Position 5 to 20, high impressions | Best refresh candidate |
| High impressions, very low clicks | Title and description problem, not content |
| Steady decline over months | Content has aged or competitors improved |
| Position 1 to 3 | Leave it alone unless facts are wrong |
| No impressions at all | Refreshing will not help; the topic or targeting is wrong |
What to actually change
- Check every fact. Prices, tool names, platform behaviour, statistics. Outdated specifics are the fastest way to lose a reader's trust.
- Answer questions the page misses. Compare against what currently outranks you and against the People Also Ask box.
- Improve the opening. Give a direct answer in the first paragraph rather than a preamble.
- Add structure. Question-form headings, a table where there is a comparison, a summary near the top.
- Fix internal links. Link out to newer relevant posts, and link in from them.
- Rewrite the title and description if click-through is low relative to impressions.
- Remove what has aged badly rather than leaving it and hoping nobody notices.
Dates: the honest way
Update dateModified when you genuinely change the content, and show it as an updated date. That is legitimate, supported by Google, and it tells readers the page is maintained.
Changing the date without changing the content is a false claim in your structured data. Google has demoted sites for date manipulation, and readers who return to an unchanged page notice.
When to consolidate instead
If you have three thin posts on nearly the same topic, they are competing with each other. Merging them into one strong page and redirecting the others usually outperforms updating all three.
- Pick the strongest URL as the survivor, usually the one with the most links or best position.
- Merge the useful content into it.
- 301 redirect the others to it.
- Update internal links to point at the survivor.
- Keep the redirects permanently.
A workable cadence
For most businesses, one refresh a week beats four new posts a week once the library covers the topics that matter. Pick the highest-impression near miss, spend two hours making it genuinely better, update the date, and move on.
Frequently asked questions
Is updating old content better than writing new?
Once you have a reasonable library, usually yes. You are improving pages Google already indexes and trusts rather than asking it to evaluate more. New content is for genuine gaps, not for volume.
Which posts should I update first?
Those ranking between positions 5 and 20 with high impressions. They are close enough that improvement moves them onto page one, and the impressions prove there is demand. Pages ranking first already need only fact-checking.
Should I change the publish date when I update a post?
Update dateModified, not datePublished, and display it as an updated date. Changing the original publish date without changing the content is a false claim in structured data and Google has acted against date manipulation.
How much do I need to change for it to count?
Enough that a reader would notice. Correcting facts, adding sections that answer missing questions, restructuring for clarity. Swapping a few words and updating the date is the practice Google's guidance is aimed at.
What if several posts cover the same topic?
Consolidate. Merge them into the strongest URL, redirect the others to it permanently, and update internal links. Three thin posts competing for one query is worse than one thorough page.