Deleting old posts feels obvious because traffic is down and the archive is full of pages nobody reads. Cutting them looks like cleaning house. Sometimes it helps. More often the traffic drop has a different cause, and deleting pages just throws away work you already paid for.
We get this question from founders regularly. A site has years of blog posts, organic traffic has slipped, and someone read that Google punishes thin content. The plan is usually the same: delete half the blog and hope the rest ranks better. That plan can work. It can also make things worse. The difference comes down to why each page is failing, and most teams never check that before they start deleting.
Content pruning is the practice of reviewing every page on your site and deciding whether to keep it, rewrite it, merge it into another page, or remove it. The goal is a smaller library where every page earns its place. Pruning is a decision process, not a delete button.
The gardening comparison holds up. You are not clearing the whole bed. You are cutting what draws energy away from healthy growth. A page that ranks for nothing, earns no links, and answers no real question is drawing energy. A page that quietly brings in a few qualified leads a year is not.
Pruning has four outcomes, and only one of them is deletion. You can keep a page as it is. You can update it. You can merge it into a stronger page and redirect the old URL. Or you can remove it. Most teams jump straight to the fourth option because it is the fastest, and speed is the wrong goal here.
Google's own guidance says no, at least not as a first move. In its documentation on core updates, Google states that "deleting content is a last resort, and only to be considered if you think the content can't be salvaged." That is unusually direct language from Google, and it is worth taking at face value.
The same Google Search Central page sets expectations on timing. It says that after you make changes, "some changes can take effect in a few days, but it could take several months for our systems to learn and confirm that the site as a whole is now producing helpful, reliable, people-first content in the long term." So even a correct pruning decision will not show results next week.
Google also adds a caveat that most SEO advice leaves out. It says there is "no guarantee that changes you make to your website will result in noticeable impact in search results." Pruning is a bet, not a lever, and we tell clients the honest odds up front. Deleting a page removes a signal, and sometimes that signal was quietly earning you work.
Probably more than you think. Ahrefs studied roughly 14 billion pages for its 2023 search traffic research and found that "96.55% of all pages in our index get zero traffic from Google, and 1.94% get between one and ten monthly visits." Zero traffic is the normal state of a web page, not a sign of failure.
That number reframes the whole exercise. If almost every page on the web gets no Google traffic, then a blog post with no Google traffic is not evidence of a quality problem. It might be a targeting problem, a competition problem, or simply a page that was never meant to rank.
This pattern shows up in most audits we run. A client points at a batch of posts with no sessions in Google Analytics 4 and calls them dead weight. Then we open Google Search Console and find that some of them are picking up impressions on queries the client would love to own. They are ranking on page three. Those are not pages to delete. Those are pages to rewrite.
Before you prune anything, pull twelve months of Search Console data and sort by impressions, not clicks. Impressions tell you Google has an opinion about the page. Clicks only tell you whether that opinion was good enough.
Delete when the page cannot be fixed by editing it. That means the topic is no longer relevant to your business, the information is wrong and not worth researching again, the page targets a query you have no business ranking for, or you have three other pages saying the same thing better.
The last case is the most common one we run into. A site publishes four posts over three years that all try to rank for the same query. None of them wins, because they split the signal between them. This is keyword cannibalization, and the fix is usually to merge the four into one strong page and redirect the other three URLs to it. That is a deletion in the sense that three URLs go away, but nothing of value is lost.
Relevance decay is the other honest reason. If you pivoted your business two years ago and still have twenty posts about a product you no longer sell, those pages are working against you. They tell Google and every AI answer engine that your site is about something you have stopped doing. Cutting them sharpens what your site is for.
Everything else is usually an update job rather than a delete job. We wrote more about that tradeoff in our piece on refreshing old content versus publishing new posts, and the short version is that refreshing an existing URL is almost always cheaper than starting over.
Never delete a page with inbound links without redirecting it. Links are the hardest asset on a website to replace, and a deleted URL turns every one of those links into a dead end. Check the page in Ahrefs or Semrush first, then either keep the page or redirect it to the closest relevant replacement.
The rule we use is simple. If a page has even one external link from a site we would be happy to be mentioned by, that page gets a 301 redirect rather than a 410 response. The redirect passes most of the value to the new destination. A deleted page passes nothing.
Be careful where you point the redirect. Sending unrelated posts to your homepage is a pattern Google treats as a soft 404, so the redirect gets ignored and you lose the value anyway. Point each redirect at the page that genuinely answers the same question. If no such page exists, keep the original.
Use a permanent method, not the emergency one. Google Search Central lists three permanent options: remove or update the content, password-protect the page, or add a noindex tag. The Removals tool in Search Console is temporary. Google's documentation states that "requests made in the Removals tool last for about 6 months."
That six month limit trips people up. A team submits a removal request, sees the page vanish from search, and assumes the job is done. Half a year later the page is back, because the underlying URL still returns a normal page. The Removals tool buys time. It does not solve anything.
For a page you want gone for good, the cleanest approach is to return a 410 Gone status code, which tells Googlebot the page was deliberately removed. A 404 works too and Google treats the two almost identically, but 410 states your intent clearly. If you want the page to stay accessible to humans while dropping out of search, use a noindex tag instead. We covered the details of that in our guide to the noindex tag.
One warning. Never combine noindex with a robots.txt block on the same URL. If you block the crawler, it cannot read the noindex tag, and the page can stay indexed indefinitely. This is one of the most common mistakes we find on inherited sites.
For most sites, no. Crawl budget is a real constraint, but it applies to very large sites. Google's documentation says you do not need to worry about it "if your site doesn't have a large number of pages that change rapidly, or if your pages seem to be crawled the same day that they are published."
Google gives specific thresholds. It flags crawl budget as a concern for large sites with over one million unique pages that change moderately often, and for medium or larger sites with more than ten thousand unique pages and very rapidly changing content. A four hundred post marketing blog is nowhere near either line.
So when someone tells you to delete two hundred posts to save crawl budget, ask how many URLs the site has. If the answer is under ten thousand, crawl budget is not the problem. Something else is.
Google does say consolidating duplicate content helps, in its words to "eliminate duplicate content to focus crawling on unique content rather than unique URLs." That is an argument for merging near-duplicate pages, which is different from mass deletion.
It raises the cost of a wrong deletion. ChatGPT, Perplexity, Claude, and Google AI Overviews pull from indexed pages, and they often cite pages that get very little direct click traffic. A post with two hundred sessions a year can still be the source an AI model reaches for when someone asks a narrow question in your field.
This is the part most pruning advice has not caught up with. Traffic used to be the only scoreboard, so a low-traffic page looked worthless. Now a page can be cited in an AI answer, influence a buying decision, and never register a session in your analytics. Deleting on traffic alone throws away pages that are quietly doing work.
Our practical adjustment is to weight specificity higher than volume. A narrow, precise post that answers one question well is the kind of page an answer engine reaches for. A broad, generic post that restates what everyone else says adds nothing whether a human or a model reads it. When we prune, generic goes first.
We start with data, not opinion. We pull twelve months of Google Search Console impressions and clicks, crawl the site with Screaming Frog, check inbound links, and then sort every URL into keep, update, merge, or remove. The sorting takes longer than the work that follows, and it is the part that protects you.
We also move slowly on purpose. Rather than deleting two hundred pages in one afternoon, we work in batches and watch what happens between them. If organic performance drops after the first batch, we still have the rest of the archive and a clear signal that the criteria were wrong. Mass deletion gives you no way to learn from a mistake.
The honest truth is that pruning is rarely the highest-value work available. In most audits we run, the bigger wins are fixing a slow site, rewriting weak pages that already rank on page two, and building internal links between pages that should support each other. Pruning is tidying. Those three are growth.
Only after you have checked Search Console. If your low-traffic posts are getting impressions, they are candidates for rewriting, not deletion. If they get no impressions, no links, and no relevance to what you sell today, then yes, merge or remove them. Start with twenty pages, not two hundred.
The mindset that helps most is treating your archive like a product rather than a pile. Every page either supports what you want to be known for or dilutes it. That question is easier to answer than "is this post good", and it leads to better decisions.
If you are staring at a few hundred posts and not sure where to start, we are happy to walk through it. We do this kind of audit as the first step on most site rebuilds, and we can usually tell within a day whether pruning is your real problem or a distraction from a bigger one. Reach out through phoenix.studio and tell us what you are working with.
Tell us where you want to go. We'll tell you how we'd get you there.