Content Pruning: Remove Outdated Content for SEO | Ighenatt

When Bankrate deleted thousands of low-performing pages in 2022, their organic traffic didn’t drop. It went up. This pattern — where deleting content improves rankings — contradicts the accumulation instinct most marketing teams operate on: publishing more always seems better than publishing less.

The data tells a different story. CNET deleted thousands of articles in June 2023 and recorded a 29% increase in estimated organic traffic in just two months — from approximately 19 million to 24.5 million monthly visits, according to SEMrush analysis published by Fery Kaszoni. An AI SaaS platform working with OneLittleWeb saw monthly traffic grow from 10.5M to 18.5M in a year — a 76% year-over-year increase — after eliminating pages with minimal views and no inbound links.

Content pruning is not deleting randomly. It’s understanding that Google evaluates site quality by looking at the whole, not just individual pages. Keeping content that doesn’t perform can contaminate the quality perception of the pages that do.

What Google Means by “Site Quality” and Why Every Page Counts

Google doesn’t evaluate pages in isolation. John Mueller, Search Advocate at Google, explained it directly: “The problem with trying to keep a frequency up is that it’s easy to end up with mediocre, fluffy content, which search engine quality algorithms might pick up on, and then assume the whole site is like that.”

This has concrete implications. If your site has 400 articles but 200 generate zero traffic, have fewer than 200 words, or cover topics unrelated to your topical authority, those 200 pages affect how Google perceives the other 200. It’s the logic of a neighborhood: if half the shops on a street are empty, the whole street looks like it’s in decline.

Mueller also clarified the debate about whether deleting or improving is the right strategy. His position: ideally improve, but when that isn’t practical, deleting is preferable to keeping low-quality content indexed. What isn’t acceptable, according to his statements in Google Search Central Q&A sessions, is keeping thin pages indefinitely hoping they’ll “maybe rank someday.”

The impact is more severe after Google’s core updates. Lily Ray, Senior Director of SEO & Research at Amsive Digital, analyzed patterns across sites most affected by the Helpful Content Update and documented that many had applied “spray and pray” strategies — publishing about everything without depth, hoping to capture long-tail traffic without providing real value. Recovery, according to Ray, required not just improving existing content but removing content that didn’t belong on the site in the first place.

This distinction between “content that can be improved” and “content that doesn’t belong” is the key to well-executed content pruning.

How to Identify Content That’s Harming Your SEO

The starting point is a Google Search Console export. Download all your site’s URLs with their last 12 months of data: clicks, impressions, CTR, and average position. This gives you the complete picture of real organic performance.

Pages requiring urgent action fall into four categories.

Pages with zero clicks and zero impressions over 12 months. If Google doesn’t show a page in any search over a full year, the page doesn’t exist for organic search. It may be indexed, but it ranks for nothing. Before deleting, check in GA4 whether it receives direct or referral traffic with business value. If it doesn’t, it’s a deletion candidate.

Pages with high impressions but CTR below 1%. This indicates the page appears in searches but nobody clicks it. The problem may be the title and meta description — not compelling enough — or the page isn’t competitive for the queries it appears in. These are update candidates, not deletion candidates.

Duplicate or cannibalizing content. If you have multiple pages targeting the same keyword with similar intent, Google doesn’t know which to rank and may split authority between them instead of concentrating it. The solution is consolidation: one URL wins and the others redirect or are deleted. (This overlaps with keyword cannibalization but surfaces in every content audit.)

Outdated articles with incorrect data. A 2019 post discussing “best SEO practices for the coming year” with five-year-old statistics not only doesn’t rank: it actively damages the site’s authority perception. These need complete updates or deletion if the topic is no longer relevant to your business.

The most useful tool for this analysis is Screaming Frog combined with a GSC export. Screaming Frog crawls all your URLs and exports them with technical metadata; GSC adds real organic performance data. Cross-referencing both sources in a spreadsheet gives you the complete inventory with the information needed to make decisions.

The Delete-Update-Consolidate Framework

The most common mistake in content pruning is applying binary logic: delete everything that doesn’t perform. Before touching any URL, you need to understand what that page has beyond organic traffic.

The right analysis runs through four questions in sequence.

Does it have quality backlinks? Use Semrush, Ahrefs, or Google Search Console (links report) to verify whether external domains link to that page. If they do, deleting without redirecting destroys real authority. The solution is always a 301 redirect to the most relevant page on the site, not direct deletion.

Does it have traffic outside organic search? A page can have zero search clicks but receive significant direct traffic because it’s the destination of a link in an email or social media post. GA4 shows this breakdown by channel. If it has referral or direct value, the deletion decision should involve the teams using it.

Is the content recoverable? Some posts have a good angle but are outdated or too short. An update with new data, more depth, and better structure can convert a 400-word page with zero traffic into a ranking article. The question is whether the update cost justifies the return potential.

Can it be consolidated with another page? If you have two articles on the same topic with mediocre performance in both, consolidation usually outperforms keeping both. One URL with all the authority concentrated performs better than two URLs splitting it. Consolidation means: choose the winning URL, migrate the best content from both, and redirect the losing URL with a 301.

If a page answers “no” to all four questions — no backlinks, no traffic outside organic, no recovery potential, no consolidation possible — then deletion is the right decision.

Crawl Budget: Why Content Pruning Matters More for Large Sites

For sites with fewer than 1,000 pages, crawl budget is rarely an urgent problem. Googlebot can crawl that volume without difficulty. For sites with 10,000, 50,000, or more indexed pages, crawl budget is the most important technical reason for content pruning.

Google’s official documentation for large sites is explicit: “If your site has many URLs that are duplicates, have little content, or are not valuable, this wastes a lot of Google’s crawling time on your site.” The time Googlebot spends on valueless pages is time it doesn’t spend re-crawling and re-indexing the pages that matter.

The practical impact: after publishing a new article or updating existing content, the time until Googlebot detects and re-indexes it depends directly on crawl efficiency. On sites bloated with low-quality pages, that process can take weeks. On pruned sites, it can happen in days.

The Crawl Stats report in Google Search Console shows crawl budget in real time: URLs crawled per day, distribution by response type (200, 301, 404, 5xx), and trend over the last 90 days. If a high proportion of the crawl is consumed by redirects or pages returning 404, there’s cleanup work to do.

The analogy that best describes it: crawl budget is like the time an inspector has to visit your building. If half the apartments are empty, in disrepair, or exact copies of the one next door, the inspector wastes time on them. That’s time not spent reviewing the good apartments — where the people who matter actually live.

The CNET Case and What to Learn From It

The most documented and debated case of mass content pruning is CNET’s. In June 2023, CNET deleted thousands of articles — some AI-generated, some old editorial content with no traffic. The SEMrush analysis published by Fery Kaszoni showed a 29% increase in estimated organic traffic between June and August 2023 — from approximately 19 million to 24.5 million monthly visits.

The analysis has nuances. Scott Rogers-Jones pointed out that part of that increase coincided with the Threads launch and Amazon Prime Day — events that generate news searches that any major media outlet captures. Attributing 100% of the increase to pruning would be inaccurate.

What is attributable to content pruning is the recovery of the long-term trend. CNET had been losing organic traffic steadily for years. The pruning didn’t just produce a one-time spike: it changed the site’s trajectory. That’s the real result of well-executed content pruning — not a guaranteed +29% in two months, but a reversal of decline and a stronger foundation for growth.

The pattern repeats. An AI SaaS platform that worked its content strategy with OneLittleWeb eliminated pages with minimal views and no backlinks. Monthly traffic went from 10.5M to 18.5M in 12 months — a 76% year-over-year increase. E-commerce sites that remove between 5% and 20% of their low-performing product and category pages after a full audit typically see consistent improvements in both traffic and revenue.

What these cases share: none of them simply deleted URLs at random. All started with a data audit, identified clear categories of valueless pages, applied redirects where there was external authority, and executed the pruning in phases to avoid abrupt negative impacts.

How to Run a Content Audit in 7 Steps

The process isn’t complex, but it requires methodological discipline and patience with data.

Step 1: Define the scope. Are you auditing the entire site or starting with one content type (blog, category pages, product landing pages)? For large sites, starting with the blog is more manageable and produces faster results.

Step 2: Export all URLs. Screaming Frog crawls the site and exports every URL with its status code, title, description, and crawl depth. For sites up to 500 URLs, the free version is sufficient. Also export the Coverage report from GSC to have the list of indexed URLs.

Step 3: Cross-reference with GSC data. Download the Performance report from GSC filtered for the last 12 months for all pages. Clicks, impressions, CTR, and average position by URL. This is the real organic performance data.

Step 4: Add GA4 data. Export sessions, time on page, bounce rate, and conversions by URL for the last 12 months. A page can have low organic traffic but a high conversion rate — that changes the decision.

Step 5: Analyze backlinks. With Ahrefs or Semrush, export the link profile for each URL in the inventory. Pages with quality backlinks get flagged as “do not delete without redirect.”

Step 6: Classify each URL. With the data from the previous steps, assign each URL one of four labels: Keep (performs well or has clear potential), Update (good topic but weak content), Consolidate (overlaps with another URL), Delete (no value in any channel).

Step 7: Execute in phases. Don’t implement all changes on the same day. Start with the clearest deletions (pages with zero activity across all channels), wait 4-6 weeks, monitor the impact in GSC, and continue with the next batch. Semrush recommends this phased approach for large sites so you can attribute causality and correct mistakes before they propagate.

What Content Pruning Doesn’t Solve

A caveat that rarely appears in content pruning articles: pruning only works if the content that remains after pruning is genuinely good.

You can delete 40% of your pages and still not improve if the remaining ones are thin content with better presentation. Google evaluates site quality based on what it keeps indexed, not on what you’ve removed. Pruning removes the ballast, but the ship still has to be seaworthy.

The most common failure pattern: sites that do aggressive pruning without a parallel plan to improve existing content. They delete 200 pages, the 300 that remain are still mediocre, and the result is a smaller site with the same quality problems.

Content pruning is a necessary condition, not a sufficient one. It works when paired with a clear strategy for what quality content will replace or strengthen what’s being removed.

If you have a site with several years of posts and suspect some content is doing more harm than good, the first step is a GSC analysis crossed with GA4 metrics. At Ighenatt we run that audit as the starting point for any content SEO project — and the results often surprise: there are more dispensable pages than expected, and removing them unlocks potential the site already had but couldn’t leverage.