Content Pruning: What to Delete vs Merge vs Update

Content pruning sounds scary because the word “delete” is sitting right there. Like you are about to walk into your own site and start ripping pages out.

But when people say “content pruning” in SEO, they usually mean something more boring and more useful. It’s just deciding what deserves to stay, what needs a facelift, what should be combined, and what should quietly go away.

And honestly. Most sites don’t have a content problem.

They have a too much content with unclear purpose problem.

If you want the short version, it’s this:

• Delete when a page has no organic value, no strategic value, and no realistic path to improvement.
• Merge when multiple pages fight each other, overlap, or split backlinks and relevance.
• Update when a page already has a footprint and just needs help.

That’s the whole game. The rest is just how you make those calls without wrecking your rankings.

Why pruning works (and why it’s not just “cleanup”)

Google doesn’t rank “sites”. It ranks pages. But it evaluates a site as a whole in a bunch of indirect ways.

When you have hundreds of thin posts, outdated pages, duplicate angles, and keyword-cannibalizing articles… you basically create a mess of signals:

• unclear topical focus
• internal links spread too thin
• crawling and indexing wasted on junk
• users landing on the wrong page and bouncing
• multiple pages competing for the same query

Pruning fixes that.

Not because Google rewards you for deleting content. It doesn’t.

But because pruning forces you to concentrate your authority, consolidate intent, and make the remaining pages stronger.

If you want a deeper walkthrough of the pruning process itself, we have a full guide here: SEO content pruning (delete vs update vs merge).

Before you touch anything, pull the 3 numbers that matter

You can get fancy with spreadsheets and dashboards, but you really only need three buckets of data per URL:

1. Performance: clicks, impressions, average position (GSC).
2. Engagement: time on page, bounce, conversions, assisted conversions (GA4 or whatever you use).
3. Signals: backlinks, internal links, index status, canonical, crawl errors (SEO crawler + GSC + backlink tool).

If you have no system for collecting this, it turns into vibes-based pruning. Which is how people accidentally delete pages that were quietly making them money.

A quick way to do this is to run a proper audit first. Here’s a solid starting point: content audit workflow and tools for quick wins. And if you want something more hands-on, this is the tool inside SEO.software: Content Audit.

Now, once you’ve got a list of URLs and basic metrics, the real question becomes:

“What action should this page get?”

Let’s break that down.

The “Delete” bucket: what to remove (and why it’s usually safe)

Deleting is correct when the page is basically dead weight. Not “low traffic”. Dead weight.

Delete it if…

1. It has no meaningful impressions in GSC (for a long time).
Not “it dipped last month”. I’m talking months of nothing.

2. It targets a keyword you don’t care about anymore.
Maybe you changed positioning, products, audience. It happens.

3. It has no backlinks and almost no internal links.
If nothing points to it, removing it rarely causes ripple effects.

4. It’s thin and unfixable.
Some posts are so shallow or off-topic that updating them is basically rewriting from scratch. And if the topic itself is weak, you are rewriting for no reason.

5. It’s a duplicate page created by accident.
Old tag pages, internal search pages, weird URL parameters, print pages, “/amp/” relics. Delete or noindex, depending.

What people get wrong about deleting

They delete and forget the redirect.

If a URL has any history, or if it has internal links, do one of these:

• 301 redirect to the closest relevant page (best option when there’s an obvious match).
• 410 (gone) when there is no relevant replacement and you want it removed faster.
• 404 is fine too, but it’s slower and messier.

Also, check internal links. If your navigation or old posts still link to a deleted URL, that’s just bleeding crawl budget and user trust.

Quick “delete” examples that are usually correct

• a 2019 “Top SEO tools” list that you never updated, has no backlinks, and is outranked into oblivion
• a 200 word announcement post about a feature you removed
• three near-identical location pages that were auto-generated with swapped city names (and don’t rank)

The “Merge” bucket: when two pages should become one stronger page

Merging is the most underrated move in pruning. Because it’s not destructive. It’s consolidating.

This is what you do when your site has multiple pages with the same intent, or mostly the same intent. In other words, you accidentally created a little internal competition.

Merge if…

1. You have keyword cannibalization.
Two or more pages alternate ranking for the same queries. Or both sit on page 2 and never break through.

2. The topics overlap heavily.
If you can outline both pages and 60 to 80 percent of the headings match, that is a merge candidate.

3. One page has links, the other has better content.
This is a classic. The “good” page has no authority. The “bad” page has the links. Combine them and you stop splitting value.

4. The user intent is basically identical.
For example: “SEO content optimization checklist” vs “on-page SEO checklist for blog posts”. Sometimes those should be one monster page instead of two okay pages.

How to merge without causing chaos

Pick a primary URL. Usually it’s the one with:

• better backlinks
• stronger historical rankings
• cleaner slug
• more internal links

Then:

1. move the best sections from the secondary page into the primary page
2. add anything missing (often examples, visuals, tools, FAQs)
3. 301 redirect the secondary URL into the primary
4. update internal links to point to the primary URL (don’t rely only on the redirect)
5. resubmit in GSC if needed

If you want a simple tool for the messy part, where you’re literally combining drafts and removing repetition, there’s a free one here: merge texts.

Merge example (realistic)

You have:

• “How to do a content audit”
• “SEO content audit checklist”
• “content audit template spreadsheet”

If those posts are all decent but none are dominant, merging them into one complete “Content audit” hub page can be a win. You build one page that actually deserves to rank.

The “Update” bucket: the easiest wins (because Google already knows the page exists)

Updating is usually the highest ROI action. Because you’re not starting from zero.

You’re taking a page that already has impressions, maybe even some links, and making it more relevant, more complete, and more aligned with what searchers want now.

Update if…

1. The page has impressions but low clicks.
This usually screams “title/meta mismatch” or “not satisfying intent”.

2. The page ranks 8 to 20.
This is the sweet spot. It’s close.

3. The page is outdated.
Old screenshots, old steps, old stats, tools that don’t exist anymore.

4. The content is okay but lacks depth, structure, or proof.
Add examples. Add steps. Add visuals. Add expert input. Add citations.

A really practical guide for this is here: content refresh checklist to optimize old posts.

What a good “update” actually looks like

Not “change a few sentences”.

A real update usually includes a few of these:

• rewrite the intro so it matches the search intent immediately
• add missing subtopics (use competitors + PAA questions)
• improve internal linking (both in and out)
• add concrete examples, screenshots, templates
• refresh stats and references
• improve UX: table of contents, skimmable formatting, clearer headings
• tighten fluff and repetitive sections

If you want a quick quality control pass after updating, this is handy: SEO friendly content checklist.

The decision framework I use (so you’re not guessing)

Here’s the thing. The delete vs merge vs update decision feels hard because you’re trying to decide using one signal.

Don’t.

Use a simple matrix. Traffic, intent overlap, links, and business value.

We have a full post on this style of decision-making here: comparison matrices for SEO (decide write vs update vs kill).

If you want the quick version, ask these questions per URL:

1) Does it serve a purpose for the business today?

• yes, it supports a product, feature, lead gen, or funnel step
• no, it’s random, historical, or off-topic

If “no” and it also has no SEO value, it’s probably delete.

2) Does it have unique intent?

If it overlaps with another page, lean merge.

3) Does it have any authority signals?

Backlinks, internal links, rankings history. If yes, lean update or merge, not delete.

4) Can it realistically rank with effort?

If the topic is too broad, too competitive, or you can’t make it better than what’s ranking… sometimes it’s not worth updating.

The most common pruning scenarios (and what to do)

Scenario A: 30 posts targeting the same keyword family

This is almost always merge + update.

Make one primary “pillar” page. Then either delete and redirect the rest, or repurpose them into supporting content targeting narrower long tails.

Also, fix internal links so they support the pillar properly. If internal linking feels like a guessing game, this will help: internal linking simple system for content sites.

Scenario B: Old posts that used to rank but now flatlined

Update first. Delete only if the topic is no longer relevant.

Often the issue is that competitors improved, Google changed the intent, or your page got stale.

Scenario C: Thin programmatic pages

If they are indexable and low quality, they can drag things down.

• if they are useful and can be improved, update templates and content rules
• if they are not useful, noindex or delete at scale

Scenario D: “We used AI to publish 200 articles and now nothing ranks”

You probably have duplication, weak differentiation, and watered-down topical focus.

It’s fixable. But you’ll need pruning and rewriting, not more publishing.

If you’re wrestling with the AI part specifically, read: make AI content original (SEO framework) and also Google detecting AI content signals.

What about E-E-A-T? Pruning helps more than people admit

A lot of sites try to “add E-E-A-T” by sprinkling author boxes everywhere. That’s fine, but it’s not the only lever.

Pruning helps because it reduces the amount of low-trust, low-proof content on your site. The more fluff you have, the more your good content gets diluted.

If you’re updating pages, use an E-E-A-T checklist while you do it. Here’s one: E-E-A-T content checklist.

Small additions can matter a lot:

• first-hand experience notes
• screenshots of the process
• specific examples and outcomes
• citations to credible sources
• clear “last updated” info (when appropriate)

The safe workflow: prune without tanking your site

If you’re doing this on a real business site, don’t do random deletions on a Friday night.

Use a process.

Step 1: Audit and tag every URL

Tag each URL as: Keep, Update, Merge, Delete, Noindex (optional bucket).

Step 2: Handle merges first

Merging reduces cannibalization and strengthens the core pages. It also gives you a cleaner map for internal links.

Step 3: Update your “almost winning” pages

These are pages ranking 8 to 20, or pages with impressions but poor engagement.

Step 4: Delete only after redirects and internal links are mapped

Deleting is last, not first. Because it is the easiest to get wrong.

Step 5: Monitor in GSC for 2 to 6 weeks

Watch:

• index coverage changes
• query distribution changes
• pages that gain impressions after consolidation
• any unexpected drops (usually internal link issues)

A note on automation (because you can absolutely speed this up)

If your site has, say, 300 pages, you can do pruning manually. It will just take time.

If your site has 3,000 pages, manual turns into a part-time job. And the decisions get inconsistent.

This is where using a platform that combines audits, briefs, updates, and on-page checks saves you. That’s basically the entire point of SEO.software. You connect your domain, run a content audit, and then you can systematically refresh and optimize content instead of guessing.

If you want to see the broader tool angle, this is useful: AI SEO tools for content optimization.

Also. Don’t ignore ops. Pruning becomes much easier when you have a content system and someone owns the workflow. Even if that someone is a VA with a checklist. This post gets into that side of it: content creation VA tasks and SOPs.

Practical “rules of thumb” (when you need a fast call)

Not perfect rules. But they work surprisingly well.

Delete when:

• 0 impressions for months
• no backlinks
• not part of your current strategy
• content is thin and not worth rebuilding
• no one would miss it

Merge when:

• two pages serve the same intent
• rankings alternate between pages
• content overlaps heavily
• one has links and one has better writing

Update when:

• the page ranks somewhere already
• it has impressions but low CTR
• the topic is still valuable
• it’s outdated, incomplete, or just kind of “meh”

The little technical stuff people forget (and regret later)

• Update your sitemap after major pruning.
• Fix internal links (don’t just redirect and hope).
• Don’t chain redirects (A redirects to B redirects to C).
• Check canonicals after merges.
• Preserve URL purpose. Redirecting a “how to” to a product page randomly is a bad match and usually a bad idea.

If you only do one thing this week, do this

Open Google Search Console. Sort pages by impressions. Find pages with:

• high impressions
• low clicks
• average position between 8 and 20

Update those first. That’s usually where the easiest growth is hiding.

Then start merging cannibalized pages. Then delete the truly useless stuff.

And if you want to make pruning a repeatable system instead of a once-a-year panic, take a look at SEO.software. It’s built for this exact loop: audit, decide, optimize, publish, track. Without needing five tools and a giant spreadsheet.

That’s content pruning. Not deleting for fun. Just making your site sharper, cleaner, and way easier for Google and humans to understand.