Most sites over three years old are carrying dead weight. Event announcements from 2019. Three articles covering the same question because three different people wrote them. Category descriptions nobody edited. A hundred posts that have never received a visitor from search.
The instinct is to leave them alone, because deleting content feels destructive & the pages cost nothing to host. But they are not free. They compete with your good pages, they dilute what your site appears to be about, & they make every other content decision harder. Pruning is one of the few SEO activities where doing less produces more.
Audit before you judge
Build one spreadsheet containing every indexable URL, with columns pulled from four sources.
- A full crawl for URL, title, word count, status code, canonical & internal links pointing in.
- Search Console for clicks, impressions, & distinct queries over the last twelve months.
- Analytics for sessions, engagement & any conversions attributed to the page.
- Backlink data for referring domains pointing at each URL.
Twelve months is the right window. Shorter periods punish seasonal content unfairly, & longer periods keep pages alive on the strength of traffic they earned years ago.
Score every page against four questions
Resist the temptation to judge by feel. Apply the same four questions to every row.
- Does it earn search traffic? Clicks & impressions over twelve months.
- Does it earn links? Any referring domains at all.
- Does it serve the business? Conversions, assisted conversions, or a clear role in the sales process even without traffic.
- Is it accurate & current? Would you be comfortable if a prospective client read it today.
A page failing all four questions has no defenders. A page passing any one of them needs a decision rather than a deletion.
Five possible outcomes
Every page gets exactly one of these, recorded in the sheet.
Keep as is. Performing, accurate, no action needed. Most sites have fewer of these than they expect.
Update. The topic is valuable & the page has some traction, but the content is dated, thin or poorly structured. This is usually the highest return action in the whole exercise, because you are improving a page search engines already trust rather than starting a new one from zero.
Consolidate. Several pages cover the same ground. Choose the strongest as the destination, merge the genuinely useful material from the others into it, then redirect the others to it. Consolidation frequently produces a page that outranks all of its predecessors combined.
Redirect. The page has no ongoing value but has links or historical traffic. Redirect it to the most closely related surviving page. Never to the homepage, which is treated as a soft 404 & wastes the equity you were trying to preserve.
Remove. No traffic, no links, no business purpose, no relevance. Return a 410 status so the removal is unambiguous, or a 404 if your platform makes 410 difficult.
The pages people wrongly delete
Three categories look like candidates & are not.
Pages with links but no traffic. They are carrying authority into your site. Update them or redirect them; do not delete them.
Pages with a business role & no search traffic. Detailed technical specifications, policy pages, or the article your sales team sends to every prospect. Traffic is not the only measure of value.
Recently published pages. Content needs several months before its performance means anything. Exclude anything published in the last six months from the audit entirely.
Execute in batches & watch what happens
Do not action 400 URLs in one deployment. If something goes wrong you will have no way to identify the cause.
Work in batches of 30 to 50, starting with the clearest cases. Record the date each batch shipped. Leave three to four weeks between batches & check overall organic sessions, impressions & conversions before continuing.
Expect indexed page count to fall & total impressions to dip. Both are intended. Watch clicks & conversions instead. If clicks fall meaningfully, review the most recent batch for a page that was more valuable than the data suggested.
Consolidation done properly
Merging is the part most often done badly. Copying three articles into one long page produces a repetitive document that serves nobody.
Instead, work from the reader's question. Decide what the merged page needs to answer, then take the best explanation of each point from wherever it exists, discard the rest, & rewrite the connective tissue so it reads as one piece. The output should be shorter than the sum of its inputs & better than any of them.
Then update internal links to point at the surviving URL directly rather than relying on redirects, & check that the redirected URLs are not still listed in your sitemap.
Prevent the archive rebuilding itself
Pruning is a symptom treatment. The cause is publishing without a review cycle.
- Give every substantial page an owner & a review date at the point of publication.
- Run a light audit annually rather than a heavy one every four years.
- Before commissioning anything new, search the existing archive. Expanding a page usually beats adding one.
- Set a rule for time-sensitive content: either it gets updated on a schedule or it gets an expiry.
What to expect
Results are not instant. Search engines need to recrawl every affected URL, & consolidation benefits accumulate over weeks. Judge the programme after a full quarter.
The secondary benefit is often the more valuable one: a smaller, better maintained archive is dramatically easier to work with. Internal linking becomes coherent, updates become manageable, & the next content decision starts from a clear picture rather than a pile.
If you want a pruning audit run properly, see our content marketing & technical SEO services, read more on the blog, or send us your page count.