Update pages that rank or convert but have gone stale. Merge several thin pages competing for the same search into one strong page, then redirect the old URLs. Delete only pages with no traffic, no links, and no purpose — and 301 them to a relevant page rather than letting them 404.
- The decision comes down to four signals per page: does it get organic traffic, does it earn backlinks, does it serve a unique search intent, and does it convert. Pages strong on any of these get kept; pages weak on all four are deletion or merge candidates.
- Updating is the default move for pages that already rank or convert but have outdated facts, stale dates, or thin coverage — refreshing them is far cheaper and faster than building ranking authority from scratch.
- Merging is the fix for keyword cannibalization: when several pages target the same query, Google splits signals across them and none ranks well. Consolidate into one page and the combined authority often lifts it above where any single page sat.
- Whenever you remove or merge a URL, 301-redirect it to the closest relevant page. A 301 passes most of the old URL's link equity and ranking signals; a 404 throws them away and frustrates anyone who clicked an old link.
- Pruning genuinely dead pages can help: Google crawls a finite share of any site, so removing hundreds of zero-value URLs concentrates crawl attention and topical focus on the pages that earn money — but the bar for 'dead' is no traffic, no links, and no purpose, not simply 'old.'
The Four-Signal Decision: Update, Merge, or Delete
Don't decide page by page on instinct — pull the data and run every old page through four questions. First, does it get organic traffic? Check Google Search Console for impressions and clicks over the last 6–12 months. Second, does it earn backlinks? Even a couple of links from real sites is equity worth preserving. Third, does it serve a unique search intent, or does it overlap with another page you already have? Fourth, does it convert — calls, forms, sales — even occasionally?
The answers map cleanly to an action. A page with traffic or links or conversions, but stale or thin content, gets updated — you keep the URL and improve what's on it. Several pages that overlap in intent and individually rank for nothing get merged into one, with the weaker URLs redirected to the survivor. A page weak on all four signals — no traffic, no links, no distinct purpose, no conversions — is a delete-and-redirect candidate.
The expensive mistake is treating 'old' as a synonym for 'bad.' Age is not a ranking problem; irrelevance and neglect are. A 2019 service page that still ranks and books jobs needs a refresh, not a funeral. Likewise, a brand-new page that duplicates an existing one is a merge problem despite being recent. Judge by the four signals, not the publish date. One more rule before you touch anything: export your full URL list with its traffic, links, and conversion data first, so every decision is reversible and you can prove the before-and-after later.
When to Update (the Default for Pages That Already Earn)
Update any page that already ranks, attracts links, or converts but has drifted out of date. This is the highest-return work in the whole exercise, because you're improving an asset Google already trusts rather than starting a new one from zero. A page sitting at position 6–12 for a valuable query is often one solid refresh away from page-one traffic — far cheaper than publishing something new and waiting months for it to mature.
What 'update' actually means depends on the page. For a service or location page, that's accurate pricing ranges, current process, fresh photos, new FAQs, and internal links to related pages you've added since. For an informational article, it's correcting outdated facts, removing dead references, expanding thin sections to genuinely answer the query better than the pages above you, and refreshing the title and meta description if click-through is weak. Updating the visible publish or 'last updated' date honestly — when you've made real changes — signals freshness; faking it on an untouched page does nothing.
Updating is also the right call when search intent has shifted under a page. A query that used to want a definition might now reward a comparison or a how-to; check what currently ranks and reshape the page to match. The trap to avoid is the cosmetic refresh — swapping a date and changing two words. Google's freshness signals respond to substantive change, not timestamps. If a page is worth keeping, give it a real update: better answer, better structure, better internal linking. If it isn't worth that effort, the honest question is whether it should be kept at all, which moves you to merge or delete.
When to Merge (the Fix for Cannibalization and Thin Pages)
Merge when several pages compete for the same search intent, or when individually thin pages would be stronger as one. The classic symptom is keyword cannibalization: you have three blog posts and a service page all loosely targeting 'commercial roof repair,' and in Search Console you can see Google rotating between them, none holding a strong position. Splitting your topical authority across four weak pages almost always ranks worse than concentrating it in one comprehensive page.
To merge well, pick the survivor — usually the page with the most traffic, the most backlinks, or the cleanest URL — and fold the genuinely useful content from the others into it, so the combined page is more complete than any original. Then 301-redirect every retired URL to the survivor. Done right, the merged page often inherits the combined link equity and ranking history and climbs above where any single page sat. Update internal links across your site to point at the new URL rather than relying solely on the redirect.
Merging also rescues thin content. A dozen 200-word posts each answering one tiny sub-question typically perform better consolidated into one thorough guide that covers them as sections. The same logic applies to near-duplicate location pages built from a template with only the city name changed — if they offer no genuinely local information, Google may treat them as doorway pages; merging or substantially differentiating them is the fix. Be careful not to over-merge: two pages that target genuinely different intents — 'how much does X cost' versus 'best X near me' — should stay separate, because a buyer and a researcher want different things and one page can't rank well for both.
When to Delete — and Always Redirect
Delete a page only when it fails all four signals: no organic traffic, no backlinks, no unique purpose, and no conversions. Expired promotions, old event pages, discontinued products, duplicate tag and category archives, and auto-generated thin pages are the usual suspects. Removing this dead weight has a real upside — Google crawls only a finite share of any site, so pruning hundreds of zero-value URLs concentrates crawl budget and topical focus on the pages that actually earn. This is content pruning, and on bloated sites it can lift the survivors.
The critical rule: deleting a URL almost never means letting it 404. Redirect every removed page with a 301 to the closest relevant page — a discontinued product to its category or replacement, an old city page to your main service page, a dead article to the updated one that superseded it. A 301 passes most of the page's accumulated link equity and ranking signals to the destination and sends anyone clicking an old link somewhere useful. A 404 throws all of that away and creates a dead end. Only return a true 410/404 when a page genuinely has no relevant successor anywhere on your site.
A few cautions. Don't bulk-delete based on a single low month — use 6–12 months of data so you don't kill a seasonal page in its off-season. Don't redirect everything to your homepage; Google often treats irrelevant mass redirects as soft 404s and ignores them, so match each redirect to the most relevant page. And remove deleted URLs from your sitemap and update internal links pointing at them. Pruning is powerful but unforgiving — measure traffic before and after, and keep your exported backup so any mistaken cut can be restored. If you'd rather have a second set of eyes run the audit and map the redirects safely, that's exactly the kind of cleanup an SEO team handles routinely.
Related questions
It helps when you delete genuinely dead pages — no traffic, no links, no purpose — and redirect them to relevant pages, because it concentrates crawl budget and topical focus on what earns. It hurts when you delete pages that quietly attract traffic, hold backlinks, or convert. Always check 6–12 months of data and 301-redirect, never 404, anything you remove.
Merging preserves the useful content: you combine two or more pages into one stronger page, fold in the best material, then redirect the old URLs to the survivor. Deleting removes a page that has nothing worth keeping. Both should end in a 301 redirect, but merging is about consolidating value while deleting is about removing dead weight.
Only if you've made substantive changes. Updating a 'last updated' date honestly after a real refresh signals freshness to both users and search engines. Changing only the date on an otherwise untouched page does nothing for rankings and erodes trust if readers notice the content hasn't actually changed. Freshness signals respond to real updates, not timestamps.
In Google Search Console, look at the queries report and the pages report side by side. If you see multiple URLs ranking for the same query — especially if their positions rotate and none holds strong — those pages are competing with each other. Consolidating them into one page, with the weaker URLs redirected, usually resolves it and often lifts the combined ranking.
Send a 301 to the closest relevant page: a discontinued product to its category or replacement, an old article to the one that superseded it, a retired location page to your main service page. Avoid redirecting everything to the homepage — Google often treats irrelevant mass redirects as soft 404s and ignores the passed equity.
Want a second opinion on your situation?
Get a free, no-obligation proposal. We’ll look at your site and your market and tell you honestly what we’d do — and what we wouldn’t.
Get Free Proposal →