Two Of Your Own Pages Are Fighting For The Same Search Result

Two of your own pages are fighting for the same search, and neither is winning

You publish a new blog post targeting a keyword. Six months later you publish another one, on a slightly different angle of the same topic, because the first one never quite ranked the way you hoped. Now you've got two pages competing for the same search, splitting whatever ranking signal either one could have earned alone, and both sitting somewhere on page two instead of one of them owning page one.

That's keyword cannibalization, and it's one of the quieter ways sites sabotage their own SEO. No one did anything wrong exactly, the content calendar just grew organically over a year or two, and the overlap only shows up once you actually go looking for it.

The Find where your own pages are competing against each other in search prompt is built for exactly that audit, spotting the real overlaps and telling you which page should win, rather than leaving both to keep splitting the same signal indefinitely.

Why this isn't the same as just having related content

Plenty of sites have multiple pages that touch a similar topic, and that's fine, even good, when each page serves a genuinely different search intent. A page about "email marketing tips" and a page about "email marketing software comparison" both live under the email marketing umbrella but answer completely different questions. That's not cannibalization, that's just a healthy content cluster.

Real cannibalization is narrower: two pages targeting the same query, the same intent, the same searcher need. When that happens, search engines don't reward you for having two attempts at the answer, they get confused about which page to rank, and often split authority between both instead of consolidating it behind either one. The result is two mediocre rankings instead of one strong one, which is worse for traffic than if the weaker page had never been published at all.

How the prompt works

You feed it your page list and target keywords or topics, whatever hunch you have about where an overlap might exist, and any ranking data showing pages performing inconsistently or multiple pages showing up for similar queries. From there it does the actual analysis work: identifying which pages genuinely target the same intent rather than just sharing a broad topic, then for each real overlap, deciding which page is the stronger candidate to consolidate behind based on content depth, backlinks, or current performance.

That second part matters more than people expect. It's tempting to assume the newer, more polished page should win by default, but if the older page has more backlinks and a longer ranking history, redirecting traffic and authority into the newer one might actually be the wrong move. The prompt pushes you to look at the actual evidence rather than assume newer is better.

It also tells you what to do about the losing page in each pair, and the fix isn't always the same. Sometimes it's a straight redirect. Sometimes the weaker page covers enough of a genuinely distinct angle that it's worth rewriting to target something different instead of just killing it. And sometimes the right call is a full merge, combining both into one page that's stronger than either was alone.

A worked example

Say you run a project management SaaS blog with two posts: "How to prioritize your backlog" published eighteen months ago, and "Backlog prioritization frameworks" published four months ago because the first one never ranked well and you assumed a fresh angle would do better. Both are still targeting essentially the same searcher, someone trying to figure out how to decide what to work on next.

You feed both URLs and their target topics into the prompt, along with a note that neither has broken into the top 10 despite decent word counts. The output flags this as a genuine overlap: same search intent, no real distinction between the two beyond publish date. It then looks at the signals, the older post has more backlinks (three from other sites linking to it over its longer life) despite being less polished, while the newer post has better structure and a more current framework list.

Rather than defaulting to "keep the newer one," the recommendation is to merge them, folding the newer post's better structure and updated framework list into the older page's URL to preserve its existing backlinks and ranking history, then redirecting the newer URL into it. Six months later, one consolidated page has a real shot at page one, instead of two mediocre pages permanently splitting the same demand.

Where the real value is

The output isn't just a list of overlapping URLs, plenty of free tools can technically flag that two pages share keywords. The value is in the judgment call about which page should win and why, backed by actual signals rather than a coin flip, and the specific next action for the page that loses, since "redirect it" isn't always right and neither is "leave it alone."

It's also honest about the overlaps that don't need fixing. Not every pair of pages sharing a keyword is actually cannibalizing each other, and forcing a fix onto a healthy content cluster wastes effort and can even hurt a page that was serving a real, distinct purpose.

How to use it

1. Pull a list of your pages and what each one is meant to rank for, along with any Search Console data showing inconsistent rankings or multiple pages appearing for similar queries.
2. Run it through the prompt and let it identify the genuine overlaps versus the pages that just share a broad topic.
3. For each real overlap, follow the specific fix recommended, redirect, rewrite, or merge, rather than defaulting to the same action for every case.

If keyword strategy in general is where you're stuck, this week's batch also covers sorting a keyword list by actual search intent, worth a look before your next content push.