How to identify SEO cannibalisation in Search Console

Inspect sources and measurements associated with an intervention. The supported automatic review path currently uses WordPress and Search Console.

SEO cannibalisation is not simply two pages receiving impressions for similar keywords. It becomes a problem when multiple URLs compete to fulfil the same intent and that competition weakens the preferred result, divides signals or sends users to the wrong destination.

Google can legitimately show different pages for broad and specific variations. The task is to prove harmful overlap before merging or redirecting anything.

Define the intent and preferred page

Choose one query family and write the reader’s task. Then identify which URL should perform that job and why. Without a preferred destination, every alternation can look like a problem.

Examples of distinct intents:

These can coexist. Two introductory pages giving the same MCP definition probably cannot justify separate roles.

Inspect queries and pages together

In Search Console, filter the query family and compare page rows across complete periods. Record clicks, impressions, CTR and average position for each URL. Then inspect the queries associated with each page.

Look for:

• two or more URLs with meaningful, comparable impressions;
• frequent switching of the top URL;
• a less suitable page replacing the strategic one;
• divided links and internal anchors;
• weaker combined outcomes after the overlap appears.

Search Console’s visible query rows can be privacy-filtered, so absence from the table is not conclusive. Preserve filters and dates. The guide to reading Search Console metrics explains the limits.

Distinguish normal overlap from a problem

Observation	Interpretation
One page clearly dominates; another gets rare impressions	Usually not urgent
Different pages win for different intents	Healthy specialisation
Locale URLs appear in their intended markets	Expected separation
Two pages alternate for the same high-value task	Cannibalisation candidate
Wrong page gets clicks and converts poorly	Stronger problem signal
Canonical or redirect conflict	Technical issue to resolve first

Do not rely on a single day. SERPs and query mixes fluctuate.

Read the actual content

Titles alone can hide duplication. Retrieve or open both pages and compare:

• primary promise and audience;
• headings and questions answered;
• products, entities and examples;
• internal links and anchor text;
• conversion role;
• canonical and indexability.

Use the Firecrawl page audit when you need a content-backed comparison. A technical canonical conflict should not be “fixed” by rewriting copy alone.

Choose the right remedy

Consolidate

Choose a primary URL when the pages truly do the same job. Merge unique useful information, preserve valuable incoming links and implement redirects or canonical changes with technical review.

Differentiate

Keep both when each can serve a defensible intent. Clarify title, H1, opening, examples, internal anchors and CTA so their jobs no longer compete.

Strengthen the preferred page

If one URL is already correct but weakly supported, update its answer and add relevant internal links. Remove misleading links that point the same anchor to another page.

Do nothing and monitor

If overlap is minor, intentional or unsupported by enough data, preserve the baseline and watch it. Avoid destructive action for a theoretical issue.

The guide on updating, consolidating or creating content helps select the editorial path.

Verify the result

Before changing anything, save page-query metrics, dates and preferred URL. After implementation and recrawling, compare equivalent periods. Check whether:

• the intended URL receives the relevant query family;
• the secondary page retains its distinct intent or redirects correctly;
• clicks and CTR do not deteriorate;
• internal links consistently support the new structure;
• the pages still convert as intended.

Do not expect immediate stability. Low-volume query families need longer observation.

Keep multilingual equivalents in one translation group rather than labelling EN, IT and ES as competing URLs. Analyse each intended market separately and verify hreflang and canonical behaviour technically when needed.

Frequently asked questions

Is ranking with two URLs always bad?

No. It can reflect different intents or normal testing and alternation.

Should I delete the weaker page?

Not automatically. Review backlinks, conversions, unique content, locale, redirects and canonical state.

Can changing internal links solve cannibalisation?

Sometimes it strengthens the preferred destination, but true duplication may require consolidation or differentiation.

How many impressions prove a problem?

There is no universal threshold. Use enough data for the site and query to distinguish a persistent pattern from noise.

Prove the conflict before resolving it

Cannibalisation is an intent and outcome problem, not a keyword-count problem. Compare page-query evidence, read both pages and choose the smallest remedy that gives each URL one clear job.

What BeKnow keeps

Keep interventions, hypotheses, assets, observations and decisions in a searchable workspace. Record what is known and what still needs checking.

A change in performance after an intervention does not prove that the intervention caused it. Cross-platform attribution and a complete analytics dashboard are not available today.

Project memory and data imports do not require an AI model key. A compatible external AI client may have its own costs. BYOK applies only to available functions that actually call an external provider.

Next step

Start with one project, one documented change and the evidence needed to review it. Source connections.

Record your first intervention. How it works.