Should you delete old content? Usually update, sometimes remove
You have accumulated pages that do nothing and you are wondering whether removing them helps.
Short answer
Split dead pages by whether they get impressions. Impressions with no clicks means demand exists and your result is losing, which is a title fix. No impressions at all means nobody searched or nobody found it, and those are candidates for merging or deletion. Deleting pages does not itself improve the rest.
The split that decides everything
Pull every page with fewer than ten clicks and sort by impressions. Pages with impressions have confirmed demand and a losing snippet. Pages with none were never searched for or never found. These two groups need opposite treatments and are routinely handled as one pile.
Group one: rewrite the result
Demand is confirmed and relevance is confirmed; only the click is missing. Read your own snippet on the results page, check whether the title truncates, and whether it answers the query the impressions are coming from. This is the cheapest traffic on any site.
Group two: merge where there is overlap
Several thin pages covering adjacent aspects of one topic often become one page that is genuinely comprehensive, with the others redirected into it. This consolidates whatever links they had and produces something with a reason to rank, which none of them had separately.
Group three: delete honestly
Pages with no impressions, no links and no overlap with anything worth keeping. Return a 404 or 410 rather than redirecting them to the homepage, which engines treat as a soft 404 and which makes your own reporting harder to read.
The claim to be sceptical about
"Deleting thin pages lifts the rest of the site" is widely repeated and hard to demonstrate, because it is almost always done alongside other work. Prune to reduce maintenance and to stop misleading your own reporting, and treat any site-wide lift as a bonus you did not pay for.
Check for links before removing anything
A page with no traffic can still carry external links, which are the most expensive thing on your site to replace. Check before deleting, and redirect those specific pages to the closest real equivalent rather than dropping them.
Do it once, then prevent it
Pruning is a symptom of publishing without a demand check. The permanent fix is the test before publishing: name the query, name what this page does that page one does not. That habit stops the pile accumulating again.
Questions
- Will deleting pages hurt my rankings?
- Deleting pages that rank for something will lose that something. Deleting pages with no impressions and no links generally changes nothing measurable except your maintenance burden.
- Should I noindex instead of delete?
- Noindex is reasonable when the page has a real purpose for users but no search value, such as internal utility pages. If it serves nobody, deleting is cleaner.
- How often should I prune?
- Once a year is plenty for most sites, and only after you have a demand check in place for new pages. Pruning repeatedly without changing the publishing habit is treating the symptom.
Measured, not asserted
On ihatepdf.cv, 93 of 160 pages earned fewer than ten clicks and 22 earned none across 224 days and 100,542 users. Every one of the zero-click pages still collected impressions, which is the distinction that matters: no impressions is an authority or demand problem and takes a year, impressions without clicks is a title problem and takes twenty minutes.
Free tool for this: Programmatic Set Analyzer. No account, nothing uploaded.
Where this goes deeper
Every number on this page comes from one complete dataset: one product taken from zero to 100K+ users on search alone, with nothing spent on advertising. The full argument is Chapter 36 of the book. Five chapters are free to read.