Content pruning

Google has said there is no site-wide penalty for having many pages. It has also said low-quality pages can affect how a site is assessed. Your aim is simple: only pages that earn their keep stay indexed.

You feel the benefit fastest on a small site. Fewer thin or duplicate URLs mean clearer internal links, cleaner sitemaps, and crawl spent on the pages that sell or help. For example, if /blog/how-we-fixed-bug-2019 and /blog/how-we-fixed-bug-2020 both get zero impressions, they do not need to sit in the index.

1. 1

   Start in Search Console, Performance, Pages

   Set the date to several months. Sort by clicks, then impressions. Mark pages with zero clicks and near-zero impressions across the full period.

2. 2

   Open each low performer

   Check queries and positions. If a page shows for no relevant queries, or sits beyond striking distance for months, it is a pruning candidate.

3. 3

   Check duplication and purpose

   Look for overlap. If /guides/getting-started and /help/getting-started cover the same task, choose one home. Keep the clearer URL.

4. 4

   Check links and internal role

   Open the Links report. If a weak page has good backlinks, consider merging its content into a stronger page and 301 redirect. If it supports users but not search, noindex it.

5. 5

   Check index coverage

   Use Indexing reports and the URL inspection tool for edge cases such as Crawled, currently not indexed or Excluded by ‘noindex’ tag. Do not fight the index with dozens of near-duplicates.

A page to keep usually has a purpose, some impressions, a path to clicks, or links. A page to prune has none of those. Write your keep or prune call next to each URL.

• Improve it: expand thin parts, match search intent, add examples, and fix title and H1. Use internal links from /pricing and /features to show importance.
• Merge it and redirect: move the best paragraphs into the stronger page, then 301 the weaker URL. Example: merge /blog/what-is-lcp into /guides/largest-contentful-paint and redirect.
• Noindex it: keep the page for users but add a meta robots noindex. Good for changelogs, archived release notes or legal pages. It leaves the index but stays live.
• Delete with a 410: remove pages with no purpose, no links and no replacements. Use a 410 to signal gone. Keep a clean 404 page for any stragglers.

Pick one choice per URL. Do not both noindex and redirect. If you merge, update any internal links to point to the new home.

1. 1

   Inventory

   Export all pages from your CMS and your XML sitemap. Add Search Console clicks and impressions for several months. Add a notes column for purpose and overlaps.

2. 2

   Decide

   Apply the four choices. Flag any page with backlinks for merge, not delete. For seasonal URLs, wait if the season has not cycled yet.

3. 3

   Execute

   Ship redirects in your server or CDN. Add meta robots noindex where needed. Remove deleted URLs from sitemaps, and keep sitemaps current.

4. 4

   Tidy links

   Fix internal links to avoid chains. Update navigation and related posts so they do not point to redirected or deleted URLs.

5. 5

   Monitor

   Watch the Performance report weekly at first. Expect impressions on the kept pages to stabilise over weeks, not days. Use URL inspection to confirm status changes.

A fixed site feels smaller and clearer. For example, /guides/getting-started ranks, while /help/getting-started redirects there and passes any equity.

• Deleting a URL with backlinks. Merge and redirect instead so you keep value.
• Pruning pages that serve users. Docs, status and legal pages often should be noindex, not gone.
• Cutting during a seasonal dip. Check the same months last year before you judge.
• Creating near-duplicates after pruning. When you improve one page, retire the twin and redirect to avoid cannibalisation.
• Expecting instant gains. As of 2026, changes roll into crawl and ranking over weeks. Keep tracking over months.