Keyword Cannibalization: A Free Checker Method and How to Fix It
Keyword cannibalization happens when two or more pages on your site target the same search intent, so Google keeps switching between them and neither performs as well as one strong page would. You can find it for free with Google Search Console: filter by a query and see how many of your pages get impressions for it. This guide shows that method step by step, how to tell real cannibalization from harmless overlap, and the four fixes that work.
What keyword cannibalization is, and when it is not a problem
The term is SEO jargon, not a Google penalty. Google's documentation on consolidating duplicate URLs describes the closest official concept: when several pages are very similar, Google decides which one to treat as canonical and shows that one. When two of your pages compete for the same intent, a few things can go wrong:
- Google alternates between them, so neither builds stable visibility.
- Links and mentions you earn are split across two pages instead of one.
- The page Google picks may not be the one you want readers to land on.
But overlap alone is not cannibalization. Two pages showing up for the same query is fine, and sometimes great, when:
- They serve different intents. A "best free tools" list and a "paid tools under $50" list can both appear for a broad tool query.
- Both rank well. Two results from your site on page one is more real estate, not a problem.
- The query is broad and minor. Many queries with a handful of impressions touch several pages. Ignore those.
The real signal is the same intent, split performance, and neither page doing well.
Free checker method 1: Search Console, query first
This is the core method and needs nothing but a verified Search Console Performance report:
- Open Performance > Search results and set the date range to the last 3 months.
- In the Queries tab, sort by impressions and note the queries that matter to you, especially those where your average position is between about 5 and 30. Those are the ones where a fix can move something.
- Click one query. This adds a query filter.
- Switch to the Pages tab. You now see every URL on your site that got impressions for that query, with clicks, impressions, and position for each.
- Write down any query where two or more pages have meaningful impressions.
Then check the pattern over time. With the query filter still on, filter by one page, then the other, and compare the impressions chart. If one page's impressions rise when the other's fall, Google is swapping them. That swap is the clearest cannibalization signal you can get for free.
One detail matters when you read the numbers. Google's help page explains that the chart is aggregated by property: "if two results from the same site appear for one query, they count as a single impression in the chart total," while the Pages table counts per page. So the page rows can add up to more than the chart total. That is expected, not a bug.
Free checker method 2: Search Console, page first
The reverse view finds overlap you did not think to look for:
- In the Performance report, filter by one page (Page > URLs containing, or exact URL).
- Open the Queries tab and export the list.
- Repeat for the other pages in the same topic.
- In a spreadsheet, put all lists together and look for queries that appear for more than one page.
On a blog with under 100 posts, doing this for each topic cluster takes an evening.
Free checker method 3: Looker Studio for the whole site
If you want query and page in one table without exporting page by page, Looker Studio's free Search Console connector can do it. Choose the URL Impression table when you connect (it reports per URL rather than per property), then build a table with the query and landing page fields, impressions, clicks, and average position. Sort by query, and every query with two or more rows is a candidate. Export to Sheets and filter out queries with very few impressions.
Free checker method 4: a site: search
For a quick sanity check on a specific topic, search Google for site:yourdomain.com topic. If you see several posts with nearly identical titles answering the same question, you have likely found the cause even before Search Console confirms it. It is not a ranking report, only a way to spot overlapping content.
How to decide what to do
For each pair of competing pages, answer three questions: do they serve the same intent, which one performs better (clicks and position), and which one has external links or mentions pointing to it? Then pick a fix:
| Situation | Fix | Notes |
|---|---|---|
| Same intent, one page clearly stronger | Merge the weaker into the stronger, then 301 | Most common fix on small blogs |
| Same intent, both weak | Rewrite into one better page, 301 the other | Keep the URL with more links |
| Near-duplicate pages you must keep (print versions, parameters) | rel="canonical" to the preferred URL | Both stay accessible |
| Different intents, blurred content | Differentiate: rewrite titles, headings, and focus | Then link between them |
| Overlap on minor queries, both doing fine | Leave it | Not a problem worth fixing |
Fix 1: Merge and redirect
Move the useful, unique parts of the weaker page into the stronger one: missing steps, examples, data, an FAQ answer. Update the stronger page so it fully covers the intent. Then set a permanent redirect from the old URL. Google's redirect documentation says 301 and 308 mean a page has permanently moved, and that the indexing pipeline uses a redirect as a signal that the target should be canonical. Use permanent redirects only when you are sure you will not reverse the change.
After merging, update your internal links so they point straight to the surviving URL instead of through the redirect, and remove the old URL from your sitemap.
Fix 2: Canonical, when both pages must stay
Sometimes you need both URLs to exist, for example a filtered version of a page or a syndicated copy. Then add rel="canonical" on the duplicate pointing to the preferred page. Google's consolidation guide ranks the signals: redirects and rel="canonical" are strong signals, sitemap inclusion is a weak one, and you can combine them. Google still makes the final choice.
What not to do: Google advises against using robots.txt or noindex to pick a canonical within a site. Robots.txt can leave a URL indexed without its content, and noindex removes the page from Search entirely rather than consolidating it.
Fix 3: Separate the intents
If both pages deserve to exist, make each one clearly about something different:
- Rewrite the title and H1 of each so they target different questions.
- Cut sections from one page that duplicate the other, and link to the other page instead.
- Make the broader page the hub and the narrower one a spoke, with links both ways.
An example from this blog
Our post on free Ahrefs alternatives started picking up Search Console impressions for a mix of intents: free tools, cheaper paid tools, Semrush-style suites, and free authority checkers. One page was trying to answer four different questions.
Instead of stuffing all of it into one page, we split by intent while planning. The free-tools post stays the hub for "free." Cheaper paid options went to a separate post on cheaper Ahrefs alternatives, and the other intents were assigned to their own planned posts. The hub links to each spoke, and each spoke links back. What Ahrefs itself offers for free stays inside the hub, so no new page competes with it. We are not claiming a ranking result from this; the point is that the split was decided on paper, before writing, which is the cheapest moment to prevent cannibalization.
Prevention: a 5-minute check before every new post
- Search your own site for the target query (
site:yourdomain.com query) and check your content plan. - In Search Console, filter by the target query and see if an existing page already gets impressions for it.
- If one does, update that page instead of writing a new one, or define the new post's different intent in one sentence.
- Assign the new post to a topic, with one hub, before you write.
- Keep a simple sheet: URL, primary query, intent, hub. One primary query per URL.
Our guide to SEO for niche blogs covers building topic clusters that make this sheet easy to keep. And if a page gets impressions but almost no clicks, cannibalization is one possible cause among several; see impressions but no clicks.
What to do next
Today: open Search Console, take your ten most important queries, and run method 1 on each. This week: decide merge, canonical, separate, or leave for every real case you found, and fix the first one. This month: add the 5-minute check to your publishing routine.
Fixing overlap makes your best pages clearer. To see how your site looks from the outside before you plan the next posts, check your site's Quality Score, built mostly from open data, with the evidence behind each number in about 30 seconds.
FAQ
What is keyword cannibalization?
It is when two or more pages on the same site target the same search intent, so Google alternates between them or splits visibility, and neither performs as well as one consolidated page would. It is an SEO term, not a Google penalty.
Is there a free keyword cannibalization checker?
Google Search Console works as one. In the Performance report, click a query and open the Pages tab to see every URL that got impressions for it. For a site-wide view, Looker Studio's free Search Console connector with the URL Impression table can list queries and pages together.
Is keyword cannibalization always bad?
No. Two pages appearing for the same query is fine if they serve different intents or both rank well. It is a problem when the pages target the same intent, Google swaps between them, and neither performs well.
Should I use a 301 redirect or a canonical tag to fix cannibalization?
Use a 301 when you merge pages and no longer need the weaker URL. Use rel=canonical when both URLs must stay accessible, such as filtered or syndicated versions. Google treats both as strong canonical signals but makes the final decision.
Can I use noindex to fix keyword cannibalization?
Google does not recommend noindex for choosing a canonical within a site, because it removes the page from Search entirely instead of consolidating signals. Merge and redirect, use rel=canonical, or differentiate the pages instead.
What is a keyword cannibalization example?
Illustrative case: a blog with two posts titled 'How to Feed a Sourdough Starter' and 'Sourdough Starter Feeding Guide' that answer the same question. In Search Console, both get impressions for the same queries and swap positions over time. Merging them into one post usually fixes it.
Read next
Check your site's Quality Score
Free, no sign-up: a 0–100 score from mostly open data. Then add your site to Satelo and get partner articles from sites of your level.
Enter a domain like example.com
Add my site