Melbourne-based

Rank in the map pack

City and suburb pages built to win local search across Australia.

Free local audit
Content Strategy · Guide

Keyword Cannibalisation: Find It and Fix It

Andy Merritt, founder of SEO Soar
Andy MerrittFounder, SEO Soar · 6 min read
Two web pages tugging the same keyword orb in opposite directions, a picture of keyword cannibalisation

More content is supposed to mean more rankings. Then one day you notice two of your own pages trading places for the same search, neither ever cracking the top five, and the more you publish the worse it gets. That's keyword cannibalisation: competing pages from the same site splitting the vote for one query. It's one of the most common problems we find in site audits, it's entirely self-inflicted, and it's very fixable. This guide shows you how to spot keyword cannibalisation in your own data and the four ways to fix it, from consolidation to a simple rewrite.

What keyword cannibalisation is (and what it isn't)

Keyword cannibalisation happens when two or more pages on one site target the same query and the same intent, forcing Google to choose between them, and forcing your clicks, links and relevance signals to split between competing pages instead of stacking on one. You'll also see it spelled keyword cannibalization in US sources; same problem, same fix. What it isn't: several pages ranking for related but different queries. A service page ranking for "hot water repair" and a guide ranking for "why is my hot water not working" aren't cannibalising, they're covering two intents properly. The test is intent, not overlapping words. If both pages would satisfy the same searcher equally, they're competing. If each serves a different need, they're a team.

Keyword cannibalisation usually creeps in rather than arriving. A blog accumulates three posts on the same subject written years apart by different people. A services menu grows a "blocked drains" page and a "drain cleaning" page because both sounded right at the time. An online store's category, brand and filter pages all chase the same product query, so common a pattern that large catalogues treat overlap checks as routine maintenance. None of these were decisions to compete with yourself; they're what happens when publishing runs ahead of planning.

The damage competing pages do

Google generally shows one or two results per site for a query, so competing pages fight for one slot. The damage compounds quietly: backlinks that would have stacked onto one URL scatter across several, internal links point in different directions, and click history splits, so no single page ever builds the momentum to climb. The classic symptom is a keyword that "flickers", ranking with one URL this week and another the next, each reset burning position. Meanwhile a competitor with one focused page collects everything on one URL and sits above you with half your content. It isn't a penalty. It's friendly fire.

Two runners from the same team tripping each other mid-race, like competing pages costing each other the ranking

How to find keyword cannibalisation

Start in Google Search Console. Open the performance report, filter to a single query, then switch to the Pages tab: more than one URL earning impressions for that query is your evidence, and a ranking URL that keeps swapping over time is confirmation. For a fast manual check, search site:yourdomain.com "your keyword" and read the list of pages Google associates with the phrase. Anywhere two results could answer the same searcher, dig in. It's also worth doing this the way we do in a competitor analysis: pull your top queries, map each to its intended page, and flag every query where the data disagrees with the map. Sites that publish frequently without a keyword map are where keyword cannibalisation breeds, because nobody decided which page owns which query.

A rank tracker adds the time dimension. If you track keywords weekly, export the ranking URL alongside the position and scan for keywords where that URL changes month to month. The flicker is keyword cannibalisation caught in the act, often visible in the graph as a sawtooth: position climbing on one URL, then resetting when Google swaps to the other.

Quick self-test: pick your most important keyword and ask, "which single URL on my site should rank for this?" If you can't answer instantly, Google can't either.

Four ways to fix it

1. Consolidate. The strongest fix for true duplicates: merge the weaker page's best material into the stronger page, then 301 redirect the old URL so its links and history transfer. Our guide to 301 redirects covers the mechanics. Consolidation turns two half-strength pages into one full-strength one, which is why it so often produces a ranking jump on its own. 2. Canonicalise. When both pages must exist for users, near-identical product variants, a printable version, point the duplicate at the preferred URL with a canonical tag so signals flow to one page. 3. De-optimise and differentiate. When both pages deserve to live, give each its own job: retarget the secondary page to a different query and intent, rewrite its title and headings, and adjust internal links and anchor text so every "blocked drains" link points to the one page that owns that phrase. 4. Delete. For thin, outdated pages with no links, traffic or purpose, removal is honest housekeeping. Redirect if there's any equity, let it 404 or 410 if there's none.

A keyword orb split and dimmed between two competing pages, showing ranking signals divided by cannibalisation

Choosing which page survives matters as much as the method. Keep the URL with the stronger history: more backlinks, more impressions in Search Console, the better fit for the query's intent. Where the two are close, prefer the URL whose path will still make sense as the site grows, and if you're genuinely unsure, run the numbers side by side for a fortnight before merging; the data usually makes the call for you. Make one person the owner of the decision, because keyword cannibalisation fixes stall when nobody feels entitled to retire a colleague's page.

Keeping it from coming back

Keyword cannibalisation is a planning failure before it's a technical one, so the lasting fix is upstream. Keep a keyword map, one primary query per URL, as part of a documented content strategy, and check every new brief against it before writing begins. When a new piece overlaps an old one, update the old page instead of publishing a rival. And tune titles and headings so each page's on-page signals declare a distinct job. Do that, and every new page you publish makes the site stronger instead of noisier, which is the whole point of publishing at all.

Finally, put a light content audit on the calendar. Once or twice a year, pull the site's queries, re-run the overlap check, and deal with anything new while it's small. Keyword cannibalisation is cheap to fix when it's two pages and expensive when it's twenty, and a site that publishes weekly can drift from one to the other inside a single year.

Suspect your pages are fighting each other? Run the free audit to see how your site presents in search, then talk to us about a full cannibalisation check and fix plan.

Andy Merritt, founder of SEO Soar
Andy MerrittFounder of SEO Soar. Senior SEO specialist, Melbourne. I work on every campaign directly, no juniors, no runaround.
Questions

Quick answers

How do I know if my site has keyword cannibalisation? +

Check Google Search Console's performance report: filter by a query and look at the Pages tab. If two or more URLs earn impressions for the same query, and especially if the ranking URL keeps swapping, you have cannibalisation. A quick manual check is searching site:yourdomain.com plus the keyword to see every page competing for it.

Is keyword cannibalisation a Google penalty? +

No. It's not a penalty and Google doesn't punish you for it. It's self-inflicted inefficiency: your pages split clicks, links and relevance signals between them, so neither ranks as well as one consolidated page would. That's why fixing it often lifts rankings without any new content or links.

Should I just delete the competing pages? +

Rarely delete, usually merge. Move the best content from the weaker page into the stronger one, then 301 redirect the weaker URL so its links and history transfer. Straight deletion throws away the equity the old page earned. Only delete outright when a page is thin, outdated and has no links or traffic worth keeping.

Free, no obligation

Stop your pages competing with each other

Run the free audit to see how your site shows up in search, where your own pages are splitting the vote, and the consolidations worth making first.

Get my free SEO audit📞
SEO ServicesIndustriesLocationsResultsResourcesAboutContact Free SEO audit