Should you fix what’s ranking poorly or double down on what’s working?
- Published November 6, 2026
- Linked from 2 answers
- In a cluster of 51
Put the next hour where it returns the highest gain, which favors pages that already work unless a weak page has a specific cause you can remove. A page that ranks has proven demand and relevance, so a push, such as covering what the result above it covers or improving its title for clicks, builds on something real. A weak page is worth fixing when you can name why it underperforms and the cause is fixable; without that, work on it is a guess.
Compare expected returns
| Candidate | Worth it when | Not worth it when |
|---|---|---|
| A page ranking 4 to 10 for a valuable search | The results above it suggest what it lacks; a section they have and you do not is a hypothesis to test, not proof of the gap | It already matches them and the gap is the competitors' standing |
| A page ranking 11 to 20 | You can see a specific gap to close | You cannot say why it ranks there |
| A page that ranks but gets few clicks | Its title and snippet are weaker than those around it | It ranks for searches with little demand |
| A page with a technical fault holding it back | The fault is clear: noindex, wrong canonical, blocked resources | Nothing technical is wrong |
| A page with no impressions for its searches | The intent was misread and can be corrected | There is no demand for what it covers |
The rows with a clear “worth it” condition beat the others. Between a strong page with a visible next step and a weak page with an unknown problem, the strong page wins the time; between a strong page with nothing left to gain and a weak page with one obvious fault, the weak page does.
Do both, in order
The choice does not have to be all or nothing. Sort the candidates by expected return, the clicks or revenue each page could gain for the effort, and work down the list. Search Console’s Performance report gives the starting data: queries, positions and clicks for each page. To rank update candidates across the whole site, use which pages are worth updating first. Before polishing a page further, check when the cost of perfect SEO exceeds what ranking is worth.
Avoid two reflexes
Rescuing every weak page out of a sense of fairness spends time where the return is uncertain. Polishing only the winners until they have nothing left to gain ignores weak pages with easy fixes. Decide by the evidence for each page, and let the expected return set the order.
Add a comment