How to read a results page before you commit a quarter to it
You are about to target a keyword and want to know whether you can realistically win it.
Short answer
Open the query in a private window and read the actual results. Count how many genuinely answer it, how many rank on authority alone, how much of the page is taken by features before the first organic result, and what format the top results share. That reading predicts winnability better than any difficulty score, because it describes this results page rather than an average.
Why the score is not enough
A difficulty number compresses a whole results page into one figure and hides the thing that decides your chances: whether the pages ranking are actually good. A score of 40 means nothing next to the observation that four of the top ten are thin pages from big domains, which is the most beatable configuration there is.
Count what is above the first organic result
Ads, an AI answer, a featured snippet, People Also Ask, images, a video carousel, a map pack. Everything above position one is space you cannot occupy and attention you will not receive. A query where organic starts below the fold is worth less than its volume suggests, sometimes far less.
Read the intent the page has settled on
The results tell you what the engine believes the query means. If every result is a tool and you plan an article, you will not rank however good the article is, and the reverse holds. Match the format the page has already chosen or pick a different query.
Look for the weak result
Almost every results page has one: a page from a strong domain that does not really answer the query, an old post, a thin category page. That is your opening. Winnability is rarely about beating the best result; it is about being clearly better than the worst one that is ranking.
Check the dates and the depth
If the top results are three years old and unmaintained, the topic has drifted and nobody has updated. If they are all recent, comprehensive and specific, you are looking at a contested query and you should know that before, not after.
Note who is missing
If no result is from a site your size, that is information. If several are, that is better information: it means the query is winnable at your authority level and you can see exactly what it took.
Write down the verdict
One line: what the query means, what format wins, what the weakest ranking page is, and what your page would have to do to be clearly better. If you cannot write that line, you are not ready to build the page.
Questions
- Should I use incognito?
- Yes, and ideally check from the country you care about. Personalisation and location change results enough that reading your own logged-in results can mislead you about what a stranger sees.
- How many results should I actually open?
- The top five, properly. Skimming ten titles tells you less than reading five pages, because the question is whether they are any good, and titles do not answer that.
- What if the results are all huge brands?
- Then either find a more specific version of the query or accept a long timeline. A specific query with three weak results is worth more to a small site than a broad one with ten strong ones, even at a tenth of the volume.
Measured, not asserted
This is how the winnable queries on ihatepdf.cv were chosen. The site reached 100,542 users at an average position of 9.07 and a Domain Rating of 22, which is a modest profile: every query it won was one where the ranking pages were beatable, and the one obvious head term it chased anyway returned 45 clicks in total.
Free tool for this: SERP Audit. No account, nothing uploaded.
Where this goes deeper
Every number on this page comes from one complete dataset: one product taken from zero to 100K+ users on search alone, with nothing spent on advertising. The full argument is Chapter 4 and Chapter 27 of the book. Five chapters are free to read.