At a glance
A useful content refresh strategy starts with purpose, not age. Prioritise pages that still answer a valuable question, then correct, expand, reposition, merge or retire them according to evidence, accuracy and the reader’s next useful step.
An old publication date does not make an article bad. A new date does not make it useful. A content refresh should improve the answer, the evidence or the reader’s next step, not just make the archive look busy.
Start with the articles that still serve a real purpose. Some need a few corrections. Others need a new argument. A few should stop competing with a stronger page.
Give each article a reason to stay
Record the audience, the question the article answers and the next useful action. If nobody can explain those three things, traffic alone is a weak reason to keep investing in it.
Google’s people-first content guidance asks publishers to consider whether their work adds useful value and demonstrates care. It also warns against changing dates simply to make content seem fresh. That aligns with a sensible editorial rule: update the substance before updating the label.
Use search data to identify opportunity, but read the actual page. A decline can reflect changing demand, stronger competition, a technical problem or an answer that no longer fits. The same remedy will not solve all four.
Sort the archive into decisions
| Decision | When it makes sense | What changes |
|---|---|---|
| Correct | The argument works but facts or links are stale | Sources, examples and specific statements |
| Expand | The page answers part of the reader’s question | Missing steps, trade-offs or practical detail |
| Reposition | The audience or search intent has shifted | Promise, structure and relevant next step |
| Merge | Several pages substantially answer the same question | One stronger destination with appropriate redirects |
| Retire | The subject no longer belongs on the site | Removal or redirection based on a genuine replacement |
Do not redirect every retired article to the homepage. A replacement should help the person who followed the old link. If no relevant replacement exists, forcing an unrelated destination creates a poor experience.
Prioritise with three simple questions
Does the subject still matter commercially? Is there evidence that people want the answer? Can a realistic edit make the page meaningfully better?
A page with modest traffic but strong relevance to a valuable buying decision may deserve attention before a popular, unrelated article. This is particularly important when a site has changed direction. Historical traffic can pull the editorial calendar back towards an audience you no longer serve.
Our SEO performance measurement guide helps separate page-level changes from sitewide averages. Look at relevant queries, not just total impressions. A software download query and a marketing strategy query do not represent the same opportunity.
Edit the answer before the keywords
Read the opening as a new visitor. Does it answer the title’s promise? Remove scene-setting that delays the useful point. Replace unsupported claims, update screenshots where they matter and add a concrete example where the advice remains abstract.
Check tables especially carefully. Product capabilities, prices and comparison criteria age at different rates. Record when material facts were checked and link to the original source.
Then review internal links. Add a relevant route to a deeper guide or commercial next step, such as a martech stack audit for readers deciding whether to replace software. Avoid inserting links simply to reach a quota.
Measure after the page has had time to be seen
Save the old title, content, dates and a performance baseline. Compare equivalent periods after recrawling and allow for seasonality or concurrent changes. A refreshed page can improve usefulness without producing an immediate ranking movement.
Review the outcome against the purpose: better relevant visibility, clearer navigation, more useful enquiries or fewer outdated claims. If performance does not improve, investigate rather than refreshing the date again.
Our preference is a smaller archive of pages with clear jobs over a large archive that nobody is willing to maintain. Publication starts the responsibility; it does not finish it.
