Content Pruning: When to Delete, Merge or Rewrite a Page

Content pruning has a reputation problem.

Some people treat it as a magic trick: delete half the blog, watch rankings climb.

Others will not touch an old post in case it breaks something.

Both are wrong. Pruning is an editorial call made one URL at a time, and most calls are not deletions.

In this guide you'll learn:

• What content pruning is, and when it actually helps
• How to build a content inventory that makes the decisions obvious
• The rules I use to decide between keep, rewrite, merge and remove
• How to remove or merge pages safely

What is content pruning?

Content pruning is the process of auditing every indexable page on a site and deciding whether each one should be kept, rewritten, merged into another page or removed. The goal is fewer, stronger pages that each answer one search need well.

The word suggests cutting. In practice, it looks more like editing.

On the sites I audit, deletions are usually the smallest group.

Most of the work is rewriting pages that are close to ranking and merging pages that split one topic across several URLs.

Content pruning vs a content audit

A content audit is the review. Pruning is what you do with the findings.

You can audit content without pruning anything.

But you should not prune without an audit, because every decision depends on data you only get from one.

Does content pruning help SEO?

Sometimes. It depends on what problem you are fixing.

Google is blunt about the ritual version. Its guidance on helpful, people-first content lists, as a warning sign, removing a lot of older content "primarily because you believe it will help your search rankings overall by somehow making your site seem fresh". The answer it gives: "No, it won't."

So deleting pages for the sake of a smaller site is not a strategy.

Pruning helps when it fixes a specific problem:

• Cannibalisation. Several pages compete for the same query, and none of them wins. Merging them gives Google one clear answer. My guide to avoiding keyword cannibalisation covers how to spot this.
• Split authority. Backlinks and internal links are spread across near-duplicate pages. Consolidating them puts those signals on one URL.
• Outdated advice. Pages with old prices, rules or product details mislead readers and lose trust.
• Crawl waste on large sites. Thousands of thin URLs can slow how quickly Google finds the pages that matter. On small sites this is rarely the issue, as I explain in what crawl budget is.

If none of those apply, you may not need to prune at all.

Your time might be better spent on new content that fills real gaps.

When should you prune content?

Pruning earns its place when the site has grown faster than anyone has maintained it.

The signs I look for:

• Two or more posts on almost the same topic, written years apart
• Old posts that rank on page two or three for queries that matter
• Pages full of advice that no longer applies
• Tag, category or archive pages indexed with nothing unique on them
• A blog that grew by posting target rather than by plan
• Location or service pages cloned from one template with a swapped name

Another trigger is a strategy change.

If you have just mapped your topics into clusters, the old content needs sorting into that structure. My guide on how to build topic clusters covers that mapping step.

How to build a content inventory

The inventory is a spreadsheet with one row per URL and the data that decides its fate.

Build it once and the decisions mostly make themselves.

Step 1: List every indexable URL

Start with a full crawl of the site, then add every URL from the XML sitemaps.

Remove anything that already returns a redirect or a 404.

Keep a note of noindexed pages, but separate them out.

Step 2: Add search performance data

From Search Console, export clicks, impressions and average position per page for the last 12 to 16 months.

Use a long window.

A page about a seasonal topic can look dead in spring and earn most of its clicks in December.

Add the top query for each page too.

It tells you what Google thinks the page is about, which is not always what you intended.

Step 3: Add business data

From GA4, add conversions or key events by landing page.

A page with modest traffic that produces enquiries is not a pruning candidate.

Then add anything your analytics will not show.

Does sales send this page to prospects? Does support link to it? Ask before you cut.

Step 4: Add backlinks

Pull referring domains per URL from a backlink tool.

A page with links from real, relevant sites needs a redirect plan even if nobody reads it.

My notes on checking backlink quality help separate links worth keeping from noise.

Step 5: Map each URL to a topic and a query

Finally, write the topic or cluster each page belongs to, then sort by topic.

This is where the overlaps jump out. Five posts about the same question, sitting next to each other in a spreadsheet, make the merge decision for you.

Pro tip: Put a rough monetary value on the traffic you are about to touch. The organic traffic value calculator gives you a number to weigh against the effort of fixing a page.

When I planned content for a clinic, I listed every URL before planning anything new. You can see how that shaped the plan in the medical aesthetics case study.

How to decide: keep, rewrite, merge or remove

Every row in the inventory gets one of four decisions. Here are the rules I use.

Keep

A page you keep still gets a quick check.

Is the information current? Are its internal links pointing to live pages?

Keeping a page is not the same as ignoring it.

Rewrite

Rewrite when the topic is right and the page is not.

This is usually the biggest win in a pruning project.

Pages sitting on page two for a query with real demand are close. A rewrite that matches the search intent better, adds what the ranking pages cover and removes what they do not often moves them. If you are unsure what the query wants, my explainer on search intent shows how to read the results.

Write a proper brief first, using my checklist of what goes in an SEO content brief.

A rewrite without one tends to become a longer version of the same weak page.

Merge

Merge when two or more pages answer the same question.

Choose the strongest URL as the survivor. Usually that is the one with the most links and clicks, or the cleanest address.

Move the best sections from the other pages into it, then 301 redirect the old URLs to the survivor.

Do not just redirect without moving content.

If the useful parts of the old page disappear, the redirect is passing readers to a page that no longer answers their question.

Remove

Remove when a page has no search value, no links, no business use and covers a topic you do not need.

Think of announcements for events long past, thin tag pages, or posts about a service you stopped offering.

Return a 404 or 410 if there is no close replacement.

Google's guidance for large sites recommends exactly that for permanently removed pages. If a relevant page exists, redirect to it instead.

When noindex is the right call

Noindex suits pages people need that searchers do not: thank-you pages, account pages, internal search results.

It is a poor fit for weak content. The page stays live, keeps getting crawled and still needs maintaining.

If a page is not good enough to be indexed, ask whether it is good enough to exist.

How to remove or merge pages without losing traffic

The execution is where pruning projects lose traffic.

Treat it like a small site migration.

1. Build a redirect map. Old URL, new URL, status code, for every merged or removed page. My guide on how to create a redirect map has the format.
2. Move the content before you redirect. The survivor page should be finished and live first.
3. Update internal links. Point every internal link at the final URL, not at the redirect. Redirects are a safety net, not a linking plan.
4. Update the XML sitemap. Remove old URLs and add any new ones.
5. Avoid chains. If an old URL already redirects somewhere, update that rule to point at the final page.
6. Annotate the date. Note when each batch went live so you can compare before and after.

Work in batches, not all at once.

If something goes wrong, a batch of twenty pages is easy to trace. A batch of four hundred is not.

If your site has hundreds of URLs to process, this crosses into technical work. That is where my technical SEO checks come in: status codes, redirect rules and re-crawls after each batch.

Content pruning on different types of site

The rules stay the same, but the usual candidates change.

Blogs and resource centres

Overlapping posts are the main issue. A blog that has published on the same handful of topics for years usually has several near-identical posts per topic. Merging is the main job.

Service businesses with location pages

Cloned town pages with a swapped place name are the usual problem. Keep pages for places where you have real presence or real work to show. Merge or remove the rest.

Online stores

Discontinued products and thin filter pages are the candidates. A discontinued product with links or steady searches can redirect to its replacement or closest category. Products coming back into stock should stay live with a clear status rather than be deleted. My e-commerce content strategy page covers how category and product content fit together.

Legal, health and finance sites

Outdated advice is a real risk here, not just a ranking issue. Every page needs a review date and a named reviewer. Pages nobody can stand behind should be updated or removed promptly.

How to measure the results of pruning

Measure the topic, not the individual URL.

When you merge five pages into one, the survivor should end up with more clicks than the five had combined.

That is the test.

What I track for each batch:

• Clicks and impressions for the affected topic, compared over matching date ranges
• Rankings for each survivor page's main queries
• Indexing status of redirected URLs in Search Console
• Conversions from the affected landing pages
• Crawl errors and any redirect chains that appear

Give each batch a full quarter before you judge it. Rewrites in particular take time to settle.

If your clicks for a topic drop and stay down after a batch, check the redirects and the survivor page first. Usually a merge lost a section that readers relied on, and putting it back fixes the problem.

Then repeat the review once a year.

Pruning is maintenance, not a one-off project. Once you know what is missing as well as what is surplus, a content gap analysis tells you where new pages should go.

Common content pruning mistakes

Most pruning damage I have had to repair came from one of these:

• Judging pages on three months of data. Seasonal pages look worthless out of season. Use at least a year.
• Deleting pages with backlinks. Links pointing at a 404 stop helping you. Redirect them to the closest match.
• Redirecting everything to the homepage. Visitors land somewhere unrelated, and the old page's relevance is lost.
• Merging without editing. Pasting three posts together creates one long, repetitive page. The survivor needs a proper edit.
• Forgetting internal links. Menus, footers and old posts keep pointing at redirected URLs for years unless someone updates them. My internal linking strategy guide covers how to keep those tidy.
• Pruning instead of publishing. If your site is missing pages for questions buyers ask, cutting old posts will not fill that gap.

The common thread is speed. Pruning rewards patience: decide carefully, ship in batches and measure each one.

Prune with a spreadsheet, not a shovel

Start with the inventory.

Sort your pages by topic, mark the overlaps and decide on merges before you even look at deletions.

You will probably find that the pages you most wanted to delete are worth rewriting, and the real wins sit in consolidation.

If you would like a second opinion on your own site, every content strategy engagement I run starts with this kind of keep, improve, merge or remove call for each URL. For a review that also covers technical, link and local issues, an SEO audit is the wider starting point. Either way, the content planning side picks up once the pruning calls are made.