What Is Content Pruning in SEO and When Should You Do It?
Removing pages can improve a site. That is the claim and it is what most readers arrive doubting. Almost every established website would perform better with fewer pages than it currently has. The reasoning matters more than the conclusion, because the conclusion is easy to apply badly.
Removing Pages Can Improve A Site
Taking pages away can make the remaining pages perform better. Not the same as before with less overhead. Actually better, which is counterintuitive enough to be worth the rest of this page.
Why it sounds wrong. Publishing feels additive.
Every page is a possible entry point, so more pages ought to mean more ways in. That reasoning is sound at small scale and stops being sound once a site is large enough for its pages to affect each other.
What the resistance actually is. Loss aversion.
Removing something that might rank one day feels like closing a door. It is a real feeling and it is why this exercise stalls at the first page somebody defends, usually one they wrote.
What is being proposed. Not a cull.
Pruning is a review with three possible outcomes, only one of which is removal. Block three sets that out. Businesses that skip to deletion do the damage this page exists to prevent.
Why this belongs in an advanced cluster. It only applies at scale.
A site of forty pages has nothing to prune. This is a problem created by years of publishing, which is why it appears in businesses that have been doing everything right for a long time.
Why It Works
Three reasons. They operate at the level of the site rather than the page. That is the part most explanations leave out. Without it the claim sounds like superstition.
A site is assessed on the whole of what it publishes. Not page by page.
Weak pages do not sit in a separate compartment from strong ones. They form part of what the site collectively demonstrates about itself, which affects how everything on it is regarded.
Attention spent on pages nobody wants is attention not spent elsewhere. Direct arithmetic.
Whatever attention a site receives is divided between its addresses. Fewer addresses means more reaching each, which is the allocation argument set out in what is crawl optimisation at scale.
Weak pages compete with strong ones. The one businesses miss.
Several pages on adjacent subjects mean none is clearly the answer to the question. Removing or combining them lets one page be the obvious response rather than three being partial ones.
Why the effect is not immediate. Reassessment takes time.
Nothing changes the moment a page is removed. The site has to be seen again in its new state, so this is work whose result arrives later than the effort. Businesses need telling that before they start.
Pruning Is Not Deleting
Three options, in order of preference. Improve, consolidate, remove. Reaching for the third first loses value that could have been kept. That is the commonest way this exercise goes wrong.
Improve. Where the subject deserved a page.
If the topic warrants a page and this one simply did not deliver, finish the job. The best outcome and the least common, because most weak pages are weak structurally rather than through lack of effort.
Consolidate. The default, also the most overlooked.
Several thin pages on adjacent subjects become one that covers the ground properly. Nothing is lost, whatever those pages had earned is preserved and the result is genuinely better than any of them.
Remove. Legitimate, though last.
Some pages should simply go. Removal is a valid answer and it discards whatever the page had accumulated, which is why it sits third rather than first.
Why the order gets inverted. Deletion is easiest.
Removing a page takes a moment. Consolidating several takes a writer, a decision about what survives and somebody handling the addresses. The cheapest option is the one that loses the most.
The test before deleting anything. One question.
Could this be part of something better rather than gone. If yes, that is consolidation and it is worth the extra work.
Consolidation Is Usually The Answer
Several thin pages on adjacent subjects frequently make one good page. This is the option most likely to be right and least likely to be chosen, because it is more work than deletion and less exciting than writing something new.
What makes pages consolidation candidates. Adjacency.
Pages that each address a fragment of one subject, where a reader landing on any of them would want most of what is on the others. If somebody has to visit three of your pages to understand one thing, those are one page.
What consolidation preserves. Everything the originals earned.
References pointing at them, the standing they built and their history all transfer when the addresses are handled properly. Deletion discards all of it, which is the difference between the two options.
Why the result beats all the originals. Completeness.
One page covering a subject properly is more useful than three covering thirds of it. It competes as a complete answer rather than as a partial one.
The example you are reading. This section.
This guide section was rebuilt by exactly this exercise. A substantial proportion of its original pages were merged into the pillars you can see in it, because each was a fragment of a subject that reads better whole. The block you are reading is part of the result.
Why we mention it. The same reasoning applies to your archive.
An agency recommending consolidation that had never done it to its own material would be worth less attention.
How To Decide What Goes
Four questions, applied together. A page failing all four is a candidate. A page failing one is not. Treating any single question as decisive is how good pages get destroyed.
Does anything arrive on it. The obvious one, also the weakest.
Visits, references, internal traffic. Necessary to know and dangerous alone, which is the whole of the next block.
Does anything link to it. From inside or outside.
A page nothing points at is disconnected from the site whatever its content. That is either a reason to link to it properly or a sign it never belonged.
Does it serve a purpose no other page serves. The consolidation test.
If another page covers the same ground, you have a duplication question rather than a deletion one. The answer is in the previous block.
Would it be written today. The most revealing question.
Would anybody commission this page now, for a reason they could state. Pages that survive only because they exist are the core of the accumulation problem in block eight.
Why all four together. Each catches what the others miss.
Applied individually they produce confident wrong answers. Applied together they produce a shortlist somebody then has to think about, which is the correct amount of work for a decision that is hard to reverse.
Do Not Prune By Traffic Alone
A page with no traffic may still be doing a job. Removing on a traffic threshold alone destroys things quietly. It is the commonest way a pruning exercise causes harm rather than fixing it.
Why the threshold is so tempting. It removes the thinking.
A number can be applied by anybody in an afternoon. Judgement requires somebody who knows the business and takes considerably longer. The easy method is the damaging one.
It may support other pages. The invisible job.
A page nobody arrives on directly can be what connects a section, passes standing to the pages that do earn, else provides the context that makes a stronger page credible. None of that appears in its own visit figures.
It may serve a rare but valuable query. Common in business selling.
A page receiving a handful of visits a year can be the one that produces your largest enquiries. Volume and value are unrelated. A threshold treats them as the same thing.
It may exist for a reason unrelated to search. Frequently forgotten.
Pages linked from an email signature, referenced in printed material, required by a contract, sent to customers by your own staff. None of that shows in search reporting.
What to do instead. Ask why before asking how much.
For every candidate, establish why the page exists before looking at what it receives. That is slower and it is the difference between pruning and vandalism.
Handle The Addresses Properly
Every page you remove or consolidate leaves an address behind. What happens to those addresses decides whether the exercise preserves value or destroys it. This is where a well judged prune still goes wrong.
Why the address outlives the page. Things still point at it.
Other sites, your own pages, bookmarks, emails and search results all continue to reference an address after whatever was there has gone. Deleting the page does not delete the references.
What consolidation requires. The old pointing at the new.
Where several pages become one, each old address should lead to the page that now covers that ground. That is what carries their accumulated value across. Skipping it turns consolidation into deletion with extra steps.
Where removal is genuinely the answer. A decision rather than a default.
Some removed pages have no successor. Pointing them at something unrelated is worse than letting them go cleanly. That is a judgement per page rather than a blanket rule.
What not to do. Send everything to the homepage.
Redirecting a large number of removed pages to one general destination is the standard shortcut and it wastes most of what the exercise was for.
Where this is covered properly. Its own material.
Mapping addresses to destinations is a migration discipline. Our site migrations material covers doing it at scale.
The Accumulation Problem
A site can gather a large number of pages that each seemed entirely reasonable when it was made. Collectively they read as low effort. This is extremely common and it happens to careful businesses. Recognising it is harder than fixing it.
How it happens. One defensible decision at a time.
A page for a campaign, a page because a competitor had one, a page to maintain a publishing schedule, a page for a service you no longer offer. Not one of those was careless. The total is the problem rather than any part of it.
Why nobody notices. There is no moment.
No page arrives and makes the site worse. The change is gradual and the people closest to it are the least able to see it, because they remember the reason for each one.
Why reviewing page by page never finds it. Wrong unit.
Every page can be individually defensible while the whole reads as volume for its own sake. That is a judgement about the site, so it cannot be reached by examining its parts.
That this applies to us. Said plainly.
An agency publishing at volume across many sectors is exactly the kind of publisher this describes. We have examined our own material on that basis and reworked what did not hold up. We publish no counts, because those came from client specific work and a number without context misleads.
Why it is reversible. Nothing was done wrong.
This is untidiness rather than misconduct. Untidiness can be cleared up.
Do It Once, Properly, Then Stop
Pruning is something you do occasionally and finish. A site being constantly pruned is being constantly destabilised, which produces the opposite of the intended result.
Why continuous pruning is harmful. Nothing settles.
Every removal and redirect is a change. A site in permanent change never reaches a state that can be assessed cleanly. You also lose the ability to tell whether the last round helped.
What a proper exercise looks like. Scoped and ended.
A defined review, decisions made across the whole archive, changes implemented together, then a period of leaving it alone long enough to read the result.
How often is reasonable. Rarely.
Once every year or two for most publishing businesses, else after something that changes the picture, such as a market maturing past the content built for it. That last case is covered in how search intent evolves.
What prevents needing it. Deciding before publishing.
The question that stops accumulation is asked at the start rather than years later: is this page worth existing. A business asking that consistently needs this exercise far less often.
The commitment to make before starting. Finish it.
A half completed prune leaves a site with the disruption and none of the benefit. Agree the scope, complete it, then stop. The full series is on the advanced SEO guide.
Improve.
Consolidate.
Then remove.
In that order, because deletion is the cheapest option and the one that loses most. We review an archive against four criteria rather than a traffic threshold. We handle the addresses so consolidation preserves what those pages earned. We will tell you plainly if your site is not large enough to need this.
What the national tier covers:
The national tier is £1,550 a month. It suits businesses competing beyond one town. We will say so if a lower tier would serve you better.
Every guide.
One practice.
Semantic search, machine learning, topical authority, programmatic production, rendering, log files, crawl allocation, information architecture, content pruning and how strategy changes with scale.
