On every site I have run, the same pattern appears: a handful of old posts drive most of the traffic, and they decay quietly — a dated screenshot here, an outdated plugin reference there, rankings sliding a position or two each quarter. Publishing something new feels productive; updating a winner is productive. But updates need rules, because refreshing the wrong posts wastes the same hours.
The four signals that a post needs a refresh
- Ranking slide on a previously stable query. A post holding position 4-6 that drifts to 8-9 over two quarters is the classic refresh candidate — Google still considers it relevant; it is just less helpful than it was.
- Position 8-20 with decent impressions. These “almost” pages need expansion, not rewriting — add the missing section searchers want. This is the cheapest traffic on any site, straight from Search Console.
- Dated mechanics. Version numbers, UI screenshots, prices, deprecated functions. Even great posts rot mechanically, and stale screenshots quietly tell readers the whole page is stale.
- Query drift. Search Console shows the post ranking for questions it does not answer. That gap is a new section asking to be written.
Refresh, rewrite, merge, or ignore — the decision
- Refresh when the post’s structure is right and the information is mostly right: update numbers, re-screenshot, add a section, tighten the intro. Ninety percent of updates are refreshes.
- Rewrite when the core approach is wrong for the intent — the post targets the query badly. Rare, and expensive, so be honest that this is usually better spent on a new post in the same cluster.
- Merge when two posts split one topic’s potential. Combine into the stronger URL, 301 the weaker one. I do this ruthlessly: three thin posts on adjacent sub-topics consistently underperform one complete post, per the coverage logic in my post-count math guide.
- Ignore everything else. Posts with no rankings, no impressions, and no strategic role are not “opportunities” — they are clutter, and refreshing them is procrastination with a productivity costume.
How to refresh without tanking the post
Updates can backfire when done carelessly. My rules: change what is stale, keep what earns. If a post ranks because of a specific section, that section stays — expand around it, do not restructure it. Preserve the title and slug unless the query itself changed; changing slugs means redirects and risk for zero benefit. And keep the publish date, updating the “last updated” only when the changes are substantial — cosmetic date-bumping is a trust killer.
One more: log what you changed and when. Without a record, you cannot tell whether an update helped, which makes every future update decision worse. This tracking is part of the measurement setup in my content ROI guide.
Scheduling updates into the calendar
Updates compete with new posts for the same hours, so they need their own queue. Mine: one update slot per production cycle, chosen from the four signals above, ranked by traffic value. During heavy months the update queue is what pauses first — per the busy-month protocol in my calendar guide — because new-posts-from-buffer keeps publishing alive and updates resume when calm returns. And the reason this discipline matters at all: maintained posts are the moat described in my 2026 strategy piece. Anyone can publish; fewer people maintain.
How often should I update old WordPress posts?
Review monthly, update what signals demand. A useful rhythm: revisit your top 20 posts once a quarter and fix mechanical rot (screenshots, versions, prices) wherever it appears.
Does updating a post improve its Google ranking?
Substantial updates often do, especially for posts sliding from stale content. Cosmetic date-bumps do nothing and erode trust. The update has to add real value to move rankings.
Should I change the publish date when I update a post?
No. Keep the original publish date and show a last-updated date instead. Faking freshness via date manipulation is detectable and damages credibility with readers and search systems alike.
When should I merge two old posts instead of updating both?
When they compete for the same intent and split weak rankings. Merge into the stronger URL, redirect the other with a 301, and the combined page usually outperforms both originals.
Which old posts should I never bother updating?
Posts with zero rankings, zero impressions, and no role in a topic cluster. Refreshing dead content is busywork; spend the hours on positions 8-20 pages with real impressions instead.
