How to fix keyword cannibalization starts with one question. Are two of your pages really competing, or do they just overlap? Many sites treat normal overlap as a crisis and break pages that worked.
This guide shows how to confirm the problem in Google Search Console. Then it walks through four fixes and when each one fits. It ends with a worked example, common mistakes, and what to check afterward.
Reporting note: We checked this guide against Google's public documentation on November 15, 2026.
Key takeaways
- Some overlap between pages is normal and harmless
- Search Console's Pages tab shows which URLs compete for a query
- Merge and redirect only when two pages do the same job
- Wait out this month's core update before big changes
- Update internal links after any redirect
What keyword cannibalization actually means
Keyword cannibalization happens when two of your pages chase the same search need. Google then has to guess which one to show. The pick can flip from week to week, so neither page settles.
The term isn't Google's, and its documentation talks about duplicate pages and canonical URLs instead. The idea is the same. One clear page per job helps readers and search engines.
Overlap alone isn't a problem. A service page and a blog post can both show for one phrase. They can still serve different needs.
Trouble starts when both pages try to do the same job for the same reader.
Think of it from the searcher's side. They want one good answer. If your site offers two half-answers, Google may show neither one well.
How do you know two pages are competing?
Look for a pattern, not a single odd week. These are the signs worth noting:
- Two URLs trading places for the same query over several weeks
- Neither page holding a steady position for that query
- The wrong page showing, like a blog post instead of the service page
- Near-identical titles and main headings on two pages
- City or service pages that repeat the same text with a new place name
- Customers landing on a page that doesn't answer their question
One sign alone proves little. Look for two or three together before you change anything. Also note which page you'd want to win, since that guides every fix below.
How to check for cannibalization in Search Console
Google Search Console is free and shows the searches that already reach your site. Its Performance report is where you confirm the problem. Follow these steps for each phrase you care about:
- Open the Performance report and set the date range to the last three months
- Click the Queries tab and find a phrase that matters to your business
- Click that query to filter the report to it
- Open the Pages tab and note every URL with impressions
- Add a page filter for each URL in turn and watch its position line
- Write down which page shows more often and which one earns clicks
Search Console counts results differently in the chart and the table.
The chart counts your site once per search. The Pages table counts each URL on its own. Two URLs there for one query is your clue.
You can also filter queries with regular expressions. That helps you group spelling variants and plurals of one phrase. It saves you from checking each version by hand.
Keep your notes in a simple sheet. List the query, each URL, and what each page is meant to do. That sheet becomes your plan for the next steps.
What changed this month: a core update
Google started rolling out the November 2026 core update on November 10, 2026. That's according to its Search Status Dashboard. Rankings can shift for many pages while a rollout runs.
That churn can look like two of your pages trading places. Google's guidance says to confirm the rollout has finished first. Then wait at least a full week before you analyze changes.
Google also suggests comparing the week after an update with the week before it started. So hold off on merges or redirects based only on this month's swings. Take notes now and decide once the data steadies.
How to fix keyword cannibalization: four options
Most cases fit one of four fixes. The table sums them up, and the sections below explain each one.
Scroll sideways to read the full table.
| Fix | Best when | What happens to the weaker page |
|---|---|---|
| Merge and redirect | Both pages cover the same thing | Its best parts move over, then it redirects |
| Retarget one page | Each page could serve a different need | It gets a new focus, title, and heading |
| Canonical tag | Near-duplicates must both stay live | It stays live but points to the main version |
| Leave it alone | Pages serve different needs and both perform | Nothing changes |
Merge and redirect
Pick the page with the stronger record. Look at clicks, links from other sites, and how well it matches the searcher's need. Move any unique, useful content from the weaker page into it.
Then add a permanent 301 redirect from the old URL to the winner. Google says permanent redirects are a strong signal that the target should be canonical. It also says server-side redirects are the most reliable kind.
Keep the redirect in place for a long time. Google's site move guide suggests keeping redirects for at least a year. From a visitor's point of view, it suggests keeping them for good.
Retarget one page
Sometimes both pages deserve to live. They just need clearer jobs. Rewrite the weaker page around a different question that your customers ask.
Change its title, main heading, and intro to match the new focus. Then link from it to the main page with clear anchor text. Our guide to internal linking covers how to word those links.
Use a canonical tag
A rel="canonical" tag tells Google which version you prefer. It suits pages that must stay live, like filtered product lists or printable versions. Google treats it as a strong signal, but it still makes the final call.
Don't use noindex or robots.txt to choose between two pages. Google advises against both for that job. Noindex removes a page from Search entirely, which is a bigger step than you need.
Leave it alone
Two pages showing for one query can be fine. Maybe one answers a quick question and the other sells the service. If both earn clicks and hold steady, let them be.
Which fix should you choose?
Work through these questions in order. Stop at the first one that gives you a clear answer.
- Do both pages serve the same reader need? If not, retarget one or leave both
- Does the weaker page have unique, useful content? If so, merge it into the stronger page
- Must the weaker page stay live for visitors? If so, use a canonical tag
- Otherwise, redirect the weaker page to the stronger one
When in doubt, start with the smallest change. A new title is easy to undo. A redirect is much harder to walk back once Google has processed it.
Back up both pages before a merge or redirect. Save the text, the title, and the URL. If the change goes badly, you can rebuild what you had.
A worked example
Here's a hypothetical example. Say a remodeling company has a service page called "Bathroom Remodeling." It also has a blog post called "Bathroom Remodeling Guide."
In Search Console, the query "bathroom remodel" shows both URLs. Over three months, the blog post gets more impressions but brings few calls. The service page shows up some weeks and disappears in others.
The blog post answers early questions about layouts, timelines, and materials. The service page sells the work. They serve different needs, so a merge would throw away useful content.
The fix is to retarget the blog post. Its new title becomes "Bathroom Remodel Timeline: What to Expect Week by Week." Its intro now answers that question first.
The post also links to the service page with the words "bathroom remodeling services." The service page keeps the main phrase in its title and heading. The team doesn't delete or redirect anything.
Then the team waits several weeks and checks the same query again. The aim is for the service page to hold the main phrase. The blog post should start showing for timeline searches instead.
Common mistakes to avoid
Most damage comes from fixing too much, too fast. Watch out for these:
- Merging pages in the middle of a core update rollout
- Redirecting every page that shares a word or two
- Redirecting to a loosely related page, like the homepage
- Deleting a page with links pointing to it and no redirect
- Using noindex to choose between two pages
- Leaving old internal links pointing at the redirected URL
- Making big changes without a backup
Internal links matter most after a merge. Google's site move guide says to change internal links to the new URLs. Redirects still work, but direct links are faster for visitors.
How to prevent cannibalization going forward
Keep a simple keyword map. List each page and the one main phrase it owns. Check the map before you write anything new.
- One main phrase for each page
- A new page only for a new search need
- An update to an old page before a new page on the same topic
- A quick review of the map every quarter
Be careful with near-identical city pages. Google's spam policies list city pages that funnel users to one page as doorway abuse. Give each location page real local substance or skip it.
The same goes for blog posts. Before you write a new post, search your own site for the topic. Often the better move is to improve the post you already have.
What to check after you fix it
A fix isn't done when you hit publish. Run through this list once the change is live:
- The old URL returns a 301, not a 302
- Internal links point straight to the winning page
- The winning page's title and heading match the main phrase
- The old URL is gone from your sitemap
- Search Console shows one main URL for the query over time
Use the URL Inspection tool to ask Google to recrawl the winning page. Then give it time. Compare equal periods, like four weeks before and four weeks after the change.
