Two of your articles target the same phrase, and neither ranks as well as you'd expect — both flickering around positions 7 and 9, trading places week to week, never reaching the top three. The instinct is to optimize harder: more words, more headings, more of the keyword. That usually makes it worse, because the problem isn't that either page is weak — it's that you have two pages doing one page's job.
The takeaway up front: keyword cannibalization is when multiple pages on your own site compete for the same query, splitting the signals that would otherwise concentrate on one page. The fix is deciding which page should own the query and making the others stop — through content consolidation, a redirect, or re-aiming one page at a different search intent. But first confirm it's actually cannibalization, because the term is wildly overdiagnosed.
What cannibalization actually is — and what it isn't
The name suggests pages "eating" each other's rankings, but the mechanism is mundane: when two URLs are both plausibly relevant to a query, the search engine has to choose which to show, and your link equity, internal links, and topical relevance spread across both instead of concentrating on one. The result is two mediocre rankings where you could have had one strong one.
Here's what trips people up: targeting the same keyword on two pages is not automatically a problem. Search engines routinely rank one page per query per site and hold the other in reserve. Cannibalization only matters when the split is demonstrably costing you — the pages swap positions, the "wrong" one ranks, or neither breaks through.
Three things that look like cannibalization but usually aren't:
- Keyword overlap. Multiple pages mentioning the same term is normal — a guide and a glossary can both say "rank tracking" without competing for the query.
- A pillar and its supporting pages targeting different intents around a theme. That's good topical structure: the pillar owns the broad term, the supporting pages own sub-questions.
- One page outranking another you'd prefer to rank. Annoying, but if only one page ranks and the other never appears, the signals aren't split — that's a preference problem, not cannibalization.
Real cannibalization is specifically same query, same intent, two pages both visible and both underperforming because of each other — and that rules out most false alarms.
How to confirm it (don't act on a hunch)
Before consolidating or deleting anything, prove the conflict exists — guessing leads to removing a page that was helping you.
- Find queries where two URLs both appear. In Google Search Console, open the Performance report, filter to a query, and check the Pages tab. If two of your URLs both pull impressions for that search, you have overlap to investigate.
- Check the search intent of both pages. Open them side by side: do they answer the same question the same way? Two how-to guides for one task genuinely compete; a how-to and a comparison may serve different searchers and not conflict.
- Look for the symptom, not just the overlap. Real cannibalization shows up as position instability (URLs swapping over weeks), the wrong page ranking while the better one sits below it, or both stuck on page two when one merits more. Stable, single-URL rankings mean nothing to fix.
- Confirm it's persistent, not a blip. Rankings wobble naturally, so a one-day swap proves nothing — confirm the conflict holds across more than one check. If you can't show two URLs competing for one intent and a real cost, stop: "fixing" it risks removing a page for no reason.
Pick the page to keep
Once you've confirmed a genuine conflict, decide which page owns the query deliberately — not by defaulting to "the newer one" or "the longer one." Weigh these in priority order:
- Authority and links first. The page with more quality backlinks and stronger internal links is usually the keeper — that equity is the hardest thing to rebuild, and redirecting into it preserves it.
- Then existing performance. A page already pulling more impressions and clicks has proven traction worth consolidating onto.
- Then intent fit. The page that most cleanly matches what searchers want is the right long-term home, even if currently weaker.
- Then completeness. All else equal, keep the more thorough, up-to-date page and fold the other's unique value into it.
Link equity leads because it's the costliest signal to recreate, while content can be improved on whichever URL you keep. Decide, write down why, and commit to one page rather than hedging.
Fix it: the three real moves
There are only three durable fixes, and which applies depends on what the pages are.
1. Consolidate, then redirect
When two pages cover the same intent, merge them. Move any unique, valuable content from the loser into the keeper, then 301-redirect the loser to the keeper. The redirect passes most of the old page's signals to the surviving URL and removes the competitor in one move. This content consolidation is the cleanest fix when the pages are genuinely redundant — and unlike a plain delete, it hands the loser's equity to the page you kept rather than discarding it.
2. Re-aim one page at a different intent
Sometimes both pages deserve to exist; they just drifted onto the same query by accident. Here you don't merge — you differentiate. Re-target one page at a distinct search intent and rewrite its title, headings, and body to own that angle. A broad guide and a narrow how-to coexist once each clearly serves its own searcher, so reach for this when consolidating would cost a legitimate audience.
3. Fix the internal signals
Often the pages are fine but your signals are confused:
- Anchor text and internal links. If both pages are linked internally with the same anchor for the target query, you're telling search engines both are about that term. Point the keyword-rich links at the page you want to rank, and link the other with anchors for its topic.
- Canonical tags for true near-duplicates (printer-friendly or parameter variants) — a canonical names the master URL so the engine consolidates without a redirect.
Internal linking is the most underused lever here, and it costs nothing; it's covered in depth in the on-page SEO guide. Often, re-pointing your anchors to the chosen page resolves a mild conflict without touching the content.
FAQ
Is targeting the same keyword on two pages always cannibalization?
No. Targeting the same term on two pages is normal. It only becomes cannibalization when both compete for the same query and intent and demonstrably hold each other back — swapping positions, the wrong page ranking, or neither breaking through. If one ranks cleanly and the other never surfaces, there's no conflict to fix.
Should I delete the page that's cannibalizing the other?
Usually not — deleting throws away the links and equity that page earned. Consolidate its useful content into the page you're keeping and 301-redirect it there, which passes most of its signals to the survivor. Delete only a worthless page with nothing to preserve.
How do I find keyword cannibalization on my site?
In Google Search Console's Performance report, filter to a query and check the Pages tab for two of your URLs pulling impressions for the same search. Then confirm they share the same intent and that the overlap is costing you — overlap without a cost isn't worth acting on.
What's the difference between cannibalization and duplicate content?
Duplicate content is the same or near-identical text on multiple URLs, resolved by a canonical. Cannibalization is distinct pages competing for the same query and intent. The fixes differ: duplicates usually need a canonical tag, while cannibalization needs consolidation, a redirect, or re-aiming one page at a different intent.
Next step
Don't go hunting for cannibalization everywhere — most keyword overlap is harmless. Take one query you care about, open Search Console, and check whether two of your URLs are genuinely competing for the same intent and underperforming. If they are, make one of the three moves above, then give it weeks to recrawl before re-checking — success looks like one URL ranking, higher than either reached while splitting the signal. To track where your pages stand before and after, visit sbranker.com.
