The page nobody planned to win
Ask an owner which of their pages brings the most search traffic and they will name the one they spent longest on. Ask the analytics and the answer is usually a different page - a support note, a comparison buried in a footer, an answer written in ten minutes to a question a customer asked twice.
That gap is not a failure. It is the most useful thing on the site, because a page that already ranks is a page search engines have already decided is relevant. You do not have to earn that decision twice.
Why the accidental page wins
A page ranks when it answers a query better than the pages competing for it. Owners write the page they want to rank: the product, the pitch, the offer. Visitors type the problem. The gap between those two is where the accidental page lives, and it wins because it closed the gap without being asked to.
Three things usually make the accidental page the better one:
- It answers a single question, so it is unambiguously relevant to one query.
- It is written in the reader's words, because it was written for one reader.
- It has no offer to place, so nothing competes with the answer.
Find it before you improve it
Do not guess. Look at the last ninety days of search traffic and sort pages by impressions rather than clicks. The page with high impressions and low clicks is the interesting one: search engines already think it is relevant, and people are choosing not to click. That is a title problem, not a content problem, and it is a far cheaper fix than writing anything new.
Then sort by position, filtering to positions four to fifteen. Those pages are already close. A page at position eleven needs a nudge; a page at position forty needs to exist differently. Most sites have a handful of pages sitting at nine, ten and eleven, and those are the ones worth a morning each.
The uncomfortable conclusion
If the page that ranks is not the page you wrote, the useful move is not to rewrite the one you wrote. It is to ask what the ranking page does that the planned page does not, and then to stop competing with yourself. Two pages on one site aiming at the same query split their own authority. Choose the one that already ranks, make it better, and link the other one to it.
That is the whole argument of this series, and it starts here: the site already tells you what works. The work is in listening to it.
Next in Rank Higher