How to find and fix keyword cannibalization in Search Console

Keyword cannibalization happens when two pages on your site compete for the same query with the same intent, and Google keeps changing its mind about which one to show. You find it by filtering the Search Console performance report to a single query and adding the page dimension. If two URLs split the impressions and neither holds a stable position, you have something to fix.

Before you go hunting, calibrate. Several pages ranking for one query is often fine. Asked about it in September 2025, Google's John Mueller said that three pages showing in the same search result doesn't seem problematic just because it's more than one. Two of your URLs on page one is a good day. The cost shows up when the two pages are near-copies, when they trade places week to week, and when the links and clicks that should sit on one page get spread across two.

What keyword cannibalization actually costs you

Three concrete losses. Your internal links point at two targets, so neither page accumulates much authority. Your backlinks do the same. And the URL Google picks is often the weaker one: a four-year-old post with a thin answer that happens to have more history, instead of the guide you wrote last month.

The expensive version is a commercial page losing to a blog post. Someone searches with money in hand, Google serves your 900-word explainer, and the reader never reaches the page built to convert.

Find it in Search Console in ten minutes

1. Open Performance, then Search results. Set the date range to the last 3 months.
2. Sort by impressions and pick a query that matters commercially rather than the one with the biggest number.
3. Filter to that exact query, then open the Pages tab. You now see every URL that earned an impression for it.
4. Flag the query if a second URL takes more than roughly 20 percent of the impressions.
5. Compare the last 3 months against the previous 3 and watch whether the split is moving. One URL handing impressions to another is the clearest signal you have.

Run that on ten queries and you'll usually surface two or three real cases. Put them in a sheet with the query, the competing URLs, and the impression split. That sheet is your work list, and it stops the audit from becoming an opinion contest.

Then pick one of three fixes

Merge when the pages answer the same question

Take the strongest sections of both, put them on the URL with the better link profile and the longer history, and 301 the other one into it. Leave the redirect in place permanently, and update the internal links that pointed at the retired URL so the redirect works as a safety net rather than the main path. Merging is the right call more often than teams expect: one page with 1,400 useful words beats two with 700 each. We covered that trade-off in write one page that ranks instead of ten that don't.

Sharpen the intent when the pages answer different questions

Sometimes both pages deserve to exist and the titles blur them. A pricing page and a "how much does X cost" post serve two different readers at two different moments. Rewrite the title, the H1, and the opening paragraph of each so the difference lands in the first line, then link them to each other with anchor text that names the distinction. Reading the live results page before you decide is the whole skill, and our post on matching your page to search intent walks through it.

Leave it alone when both pages earn clicks

If both URLs pull clicks at a reasonable CTR and their positions have been steady for three months, you're occupying two slots on page one. Do nothing. Write the decision in the sheet so the next person auditing the site doesn't merge them by reflex six months from now.

What to watch after you ship the change

Give it two to four weeks. Google has to recrawl both URLs and process the redirect before anything settles, and the first week after a merge usually looks worse than the fourth.

Then check three numbers on the surviving page: impressions for the target query, average position, and clicks. The pattern you want is impressions holding roughly flat while position and clicks climb, because the same demand now lands on one page instead of splitting.

If clicks drop and stay down past four weeks, the merge probably folded two intents into one page. Split it back out, keep the redirect on the old URL, and publish the second intent on a fresh URL with a title that says plainly who it's for.

Keep the sheet and re-run the check every quarter. Cannibalization creeps back in every time someone publishes a post without asking which existing page already owns the query, so the ten-minute audit is cheaper than the cleanup.

If you want help turning this into a plan for your site, that's the kind of work we do.