Content Pruning vs Content Refresh: How to Choose

Content pruning removes a page from your site. Content refresh rewrites one you are keeping. They are not two intensities of the same job, they are two different calls, and only one of them can be taken back.

Most guides on this query spend their length on the first decision, which posts deserve your afternoon. That decision has a grid, and it is not on this page. This page starts where the grid ends, on the posts it has already excluded from a refresh, and it answers the question the grid leaves open: merge it, or cut it.

The question most guides answer, and the one they skip

Sorting a backlog into keep, update, consolidate and remove is a job with a known method, and it belongs to the main guide: refresh old WordPress posts. Four buckets, one Search Console signal each.

Two of those buckets end in a refresh. The other two do not, and that is where guides stop being useful. A post with impressions and almost no clicks, and a post with neither over a year, get the same two-word answer in most of what is published: prune it. Those two posts need opposite treatments, and telling them apart takes about a minute.

When a post has no traffic but is still relevant

Four things save a post that brings you nothing today. Any one of them is enough.

• An external link points at it. Someone vouched for that URL. Deleting it throws the vouch away, and nothing you publish later recreates it.
• Five or more of your own posts link to it. Your site already treats it as a reference. Removing it breaks those links and the reader hits the break, not you.
• It converted. A post can bring forty visits a year and two quote requests. A traffic threshold kills it; your accountant would not.
• It still answers a question people ask, even if a stronger post of yours now answers it better.

That last case has a name. When two of your posts split one query, that is cannibalization, and the fix is not to rewrite either of them. Fold the weaker into the stronger, redirect it, and you stop dividing the signal instead of strengthening the wrong half of it. Whether the drop you noticed is really that, or one of four things that draw the same falling line, is its own question: content decay.

Merging is the answer here far more often than cutting is. A merge keeps the link, keeps the URL working, and keeps the paragraph that was worth keeping.

When a post has no traffic and no value

None of the four applies. No link, no internal references, no conversion, nothing left that another page of yours does not say better. That post is a candidate for removal, and the word candidate is doing real work in that sentence.

Here is what the field will tell you instead. We read the six guides ranking on this query on September 17, 2026. Five of them hand you a numeric threshold: under fifty clicks in twelve months, under fifty impressions, fewer than five referring domains. None of the five sources its numbers. Two say so themselves, in smaller type, calling their own grid illustrative rather than a benchmark.

Google's John Mueller addressed that exact use of those exact numbers. Listing age, traffic, bounce rate and time on site, he wrote that they are worth a list but that "it's a starting point, but don't use it as a deciding factor alone" (Mastodon, reported by Search Engine Land on August 9, 2023). Danny Sullivan, in the same article, went further: Google has never advised deleting content just because it is old.

That is the same position this site already publishes about age, arrived at from the other direction. Age measures elapsed time, not decline. It tells you where to look first, and the looking is the part that decides. The sorting itself, and the threshold behind it, are find outdated content on your site.

Noindex, delete, 410 or redirect: what each one costs

You have decided to remove a post. You now have four ways to do it, they are not interchangeable, and not one of the six guides we read explains the difference between the two that look identical.

The 404 and the 410 are the pair nobody separates. Both make the page disappear for a reader. Only the 410 tells a crawler the disappearance is deliberate and permanent, which is why it settles faster and why it is the harder one to regret.

The 301 has its own failure, and two of the six guides do name it: redirecting a retired post to your home page, or to any page that does not answer what the old one answered. Google reads that as a soft 404 and treats it as a removal anyway, so you paid the cost of the redirect and got none of its benefit. A redirect toward nothing relevant is not a redirect, it is a deletion with extra steps.

Does pruning actually lift the rest of the site?

This is the claim the whole category is built on, and the six guides we read on September 17, 2026 contradict each other about it. Two refute it. One explains it by crawl budget. One asserts it from its own case. One raises it and then walks it back. One asserts it across five thousand words without citing a source.

Exactly one of the six cites Google. Google answers, and Google says both halves.

"Deleting content is a last resort, and only to be considered if you think the content can't be salvaged."

"If that's the case for your site, then deleting the unhelpful content can help the good content on your site perform better."

Both sentences sit in the same section of Google Search's Core Updates documentation, updated December 10, 2025 and read on September 17, 2026. John Mueller pulled harder in one direction two years earlier: "keep in mind that removing content doesn't make the rest rank higher" (X, September 22, 2023, reported by Search Engine Roundtable two days later).

Hold the two together, because together they say something neither says alone. Deleting is not a lever you pull for growth. It is what is left when a page cannot be saved. Which is the whole reason the previous two sections are longer than this one: the posts worth merging outnumber the posts worth cutting, on every backlog we have measured.

Google documentation quoted under CC BY 4.0.

What you cannot undo, and what merging quietly erases

A rewrite that lands worse is a bad afternoon. You read the diff, you roll it back, the post is what it was. A deletion that lands worse is a different category of problem, because there is no version to go back to. Of the six guides we read, one mentions this. None measures it.

Merging is the gentler move, and it has a cost of its own that we can put a number on. When the stronger post absorbs the weaker one's vocabulary, the rare terms the two used to share stop being rare. Their weight is halved in both articles, and the pair that a detector used to see stops scoring. On a real pair, on September 10, 2026, the score fell from 0.675 before the merge to 0.100 after, under the 0.40 threshold that reports anything at all.

Read that carefully, because it is not an argument against merging. Both articles were still published, still on the same query, and the map had gone blank. A tool that finds duplicate coverage by measuring vocabulary will forget the duplicate it just helped you fix, unless it is built to remember. Unstale remembers the merge for exactly that reason.

What Unstale does here, and what it does not

Unstale never deletes a post. It does not trash one either. Nothing in the plugin removes your content, by design, because the last section is the reason.

What it does do: it sorts your backlog so the candidates surface, it merges two posts when they split a query, it serves the redirect for a post its Pro add-on retired after a merge, and it lets you undo that redirect from the admin screen.

What it does not do, and you should know before you follow this page's advice: there is no screen in Unstale for creating a redirect by hand. If you retire a post yourself, the redirect is yours to set up, with a plugin made for it or in your server config. We would rather tell you that here than let you find out after you hit delete.

What the other branch costs, if you decide the post is worth saving after all, is what an AI refresh costs.

Questions people ask about pruning and refreshing

Will I lose the backlinks if I delete the post?

Yes, unless you redirect it to a page that answers the same question. A 301 toward relevant content carries the link's value. A 301 toward your home page carries almost nothing, because Google reads an irrelevant redirect as a removal. Check for external links before you decide, not after.

How many posts should I prune at once?

In batches, with a gap between them. The gap matters more than the size: remove two hundred posts in one afternoon and your traffic moves, you cannot tell which removal did it. Smaller waves, far enough apart to read the result.

How long before I know whether pruning worked?

A quarter at least, and you read it on the rest of the site, not on the pages you removed. Those are gone, so of course their traffic is zero. The question is whether anything else moved.

Does unpublishing a post count as pruning it?

No. A WordPress draft serves a 404 without meaning to: no redirect, no signal that the removal is permanent, no note about why the post left. Choose one of the four dispositions above deliberately instead. If you are unsure, noindex it and leave it published, because that is the one you can take back.

What happens to the comments on a post I merge?

They stay attached to the post being absorbed and disappear with it. Nothing moves them to the survivor. If the thread under an old post is worth something, and on some posts it is the best part, move what matters across before you merge.

Can I prune a page that ranks but converts nothing?

You can, and it is usually the wrong worry. The documented trap runs the other way: the page that ranks for nothing and quietly brings two enquiries a month. A page that ranks is answering somebody. Find out who before you remove it.

Does Unstale delete posts for me?

No. Unstale sorts, rewrites and merges. It never deletes or trashes a post, and no setting changes that. Removal stays a decision you make in WordPress, one post at a time.