What to actually do with Semrush's 2024 ranking factors
We read Semrush's 2024 Ranking Factors Study so you can stop chasing correlations and start shipping the work that moves rankings.
Semrush published its 2024 Ranking Factors Study, and most summaries of it read like a list of correlations you already suspected. We wanted the other thing, the part that changes what you open on Monday morning. So this is our reading of their data, not a recap of their headings.
Semrush ran the numbers. We ran the “so what”. Below is how we translate their findings into a to-do list, with a clear note every time we are stepping past what the study strictly measured.
The methodology is theirs and worth stating plainly. Semrush analysed over 16,000 English keywords and their SERPs, using a Spearman correlation metric to gauge how strongly each factor tracked with position. Correlation, not causation. We keep that caveat in front of us for the whole piece.
01What does the study say matters most for rankings?
Content relevancy and quality show the strongest correlation with higher rankings in Semrush’s 2024 study. That is their finding, and it is the one we would bet a quarter’s roadmap on. The practical read is unglamorous. Cover the topic properly, answer the question a searcher actually has, and stop counting keywords.
We see the same thing in client work. Pages that win are the ones that finish the job, not the ones stuffed with the target phrase. Semrush’s data suggests content freshness and raw keyword count are weaker signals than most teams treat them as.
Write the page that ends the search. That is the whole content strategy in one line.
Our take, and here we are extending past the study: treat “quality” as a testable claim, not a vibe. Does the page answer the follow-up questions? Does it name the entities a knowledgeable reader would expect? If not, you have a content gap dressed up as a finished page.
02How many backlinks do you actually need?
Backlinks from unique referring domains still correlate with stronger page and domain performance in Semrush’s analysis. Volume helps, but their reading is that a smaller set of quality links can build topical authority and pull traffic. We agree, and we would go further.
For most of the sites we advise, chasing hundreds of links is the wrong fight. A handful of genuinely relevant domains, earned because the content deserves them, does more than a bought pile ever will.
- 01Unique referring domains over total link count.
- 02Relevance to the topic over raw domain authority.
- 03Links you earned with content over links you bought.
03Where should you spend your ranking effort first?
Improving pages that already rank returns the most, according to how we read Semrush’s URL and domain findings. The study highlights that small position gains compound, moving from position 2 to position 1 can lift organic traffic sharply. So the first move is not a new page. It is your near-winners.
Pull the pages sitting at positions 2 through 5. Those are the ones one solid revision away from real traffic. New content matters, but the fastest returns hide in what you have already published.
04Do exact-match keywords in titles still help?
Exact-match keywords in titles and meta descriptions show minimal or no correlation with higher rankings in Semrush’s study. This is the finding we wish more teams internalised. Stop writing titles for a matching algorithm that has moved on.
Their data points a different way: include terms in titles and descriptions that are naturally relevant to the content and to what the searcher wants. We write titles for the human scanning a results page. The keyword shows up because the page is genuinely about the thing, not because a rule demanded it.
05Does user experience affect rankings directly?
Semrush found minimal or no direct correlation between user experience metrics and higher rankings. That does not mean UX is a waste of time, and their study says as much. These metrics shape what visitors decide to do once they arrive.
Here is our position. A slow, awkward page can still rank, but it will bleed the conversions you actually care about. So we monitor technical performance because it protects the business outcome, not because we expect it to be a ranking lever on its own.
06Why does brand matter for SEO?
Brand building matters because a significant share of visitors to top-ranking pages arrive directly or through branded searches, per Semrush’s user-signal findings. That reframes brand from a “marketing thing” into an SEO input. People who already know you are part of why your best pages rank.
Our read: awareness work and content work are the same project. Every time someone searches your name instead of a generic query, you have earned a signal no link can buy.
- 01Finish your near-winning pages before writing new ones.
- 02Judge content by whether it ends the search, not by keyword density.
- 03Earn a few relevant links instead of chasing volume.
- 04Write titles for humans; drop the exact-match habit.
- 05Treat UX as conversion protection and brand as a ranking input.
07What is the honest limit of a study like this?
Correlation studies describe what tends to appear alongside high rankings, not what causes them, and Semrush is upfront that individual factors are hard to isolate. We hold the same line. Their study is a strong map of the terrain, and it is still a map, not the territory.
That is why our advice above is framed as priorities, not guarantees. SEO is analysis plus judgement, not a checklist you tick and walk away from. Use the study to decide where to look first. Use your own data to decide what actually worked.
The full study is Semrush’s, and we recommend reading it in their own words. You can find the complete 2024 Ranking Factors Study on Semrush’s site. Everything above is our reading of their data, not a substitute for it.