SEO Ranking Factors 2015 Summary
From the SearchMetrics Ranking Factors Report
Every year SearchMetrics releases a data-driven report on the factors that determine how well a site ranks. I’ve gone through the report and pulled out results I felt were most important, as a sort of cheat sheet to refer to for myself. I decided to publish it here in case others find it useful as well.
You can get the original report here, and I highly recommend you download it and give it a read. Even if you read this entire cheat sheet, there’s still a lot more data to discover in the original report.
On Page
Meta Descriptions
Meta Descriptions were present on almost all (96%) sites ranking in the top 30 for the search results used in this study
A compelling meta description doesn’t directly affect your ranking, but it influences click-through rate from the SERPs, which does affect rankings.
Header Tags
H1 and H2 (header tags) were present for 70%+ of sites ranking in the top 30. Sites that are better optimized with meta tags and header tags create a better user experience, and lead to better click-through rates (as it provides a better structure and understanding of the page being visited).
Technical
Keyword in Domain Name
Having a keyword in your domain name continues to decrease in ranking importance, as the increased focus on building a strong brand replaces keywords in relevancy.
HTTPS
Having a secure domain (via HTTPS certificates) is new, compared with other ranking factors, but one that continues to be more and more important. This matters more to niches selling products or with sensitive user data.
Ratio Of Home Pages
There is a trend that less homepages are ranking in the SERPs, in favor of landing pages or subdomains (presumably as Google is trying to serve the page with the most relevant answer and is getting better at identifying this).
Ratio of Subdomain
In contrast to the ratio of home pages, there is a negative correlation of a site ranking well in Google, and that site being a subdomain (a 1:4 ratio of subdomains to domains appearing in the top spots in the SERPs).
.COM Domain Rankings
Excluding Wikipedia (a .org), 81% of the domains in the top 30 for all tracked SERPs are .com, 84% of which can be found in the top 10. This is an increase from previous years. Top Level Domains are not generally considered a ranking factor.
Speed and Website Site
Bigger websites (presumably with more content and media) are ranking higher than their smaller counterparts (which is a trend increasing from last year). At the same time, these bigger sites are also faster.
It’s very clear that sites with faster site load speed have better rankings than those with slower load speeds.
In the above image, you can see that the sites especially in the top 5 loaded in less time than those in positions 5–30, and the top 10 loaded much faster than those outside the top 10.
URL Length
Perhaps relating back to the increased number of landing pages vs. root domains ranking in the top spots, the length of URLS in the top 10 tend to average between 30 and 50 characters, while the average length of spots 11–30 averaged around 50 characters. This is an increase over last year.
The first position has the lowest URL length of all, likely because brands ranking in this position are being ranked for their home page.
User Experience
Number of Internal Links
The number of internal links a site has is consistently higher in websites ranking in the top 10 than outside the top 10. The number of internal links has also increased from 2014 to 2015.
Internal links can easily become a risk if over-optimized, but done right, internal links help both the user and the search engine spiders navigate/crawl your site.
Number of Images
Another metric that ties in to increased site size is the increased presence of images on sites ranking in the top 30. Again, this is a trend that is increasing over previous years. Site ranking in the top 10 had more images present, on average, than sites ranking in the top 30.
Having images creates a rich user experience and can decrease bounce rate, which will help rankings.
Presence of Unordered Lists
Sites that ranker higher tend to have better organized content, so it’s not really a surprise that among sites ranking in the #2 position, 50% use unordered lists, while just 40% of sites use unordered lists ranking in position #30.
Max Bullets in List
The more bullet points a list contained, the higher the ranking. Pretty straightforward, and likely points to the length and depth of content present on top ranking sites.
Click-Through Rate
It’s not news that sites that rank higher up in the SERPs get clicked on more than the positions below them. Sites with well-written and compelling meta data can improve rankings by being clicked on more frequently than the sites the rank near.
Sites in the top spot get clicked on an average of 32% of the time, while sites in psotion 10 get 3% of the clicks.
Time On Site
This is a metric that directly measures the user experience (in general, a longer time on site represents having been served a relevant page for their search query), and thus impacts rank, as Google wants to deliver the most relevant results possible. On average, the sites within the top 3 had a significantly longer time spent on site than the 4–10 positions did.
Bounce Rate
Similar to time on site, bounce rate is a metric that measures user experience, and impacts search result positions. If a site is visited through a Google search result, and the user quickly hits the “back” button to be taken back to the search, that is a negative signal and indicates a high bounce rate.
Content
Word Count
The differences between the word count among pages ranking in the top 30 have decreased, suggesting that longer, richer content has become a standard. The average word count in the top 30 sites for 2014 was 902, while the average word count in top 30 sites for 2015 is 1140.
The sites in the bottom half of the top 10 have a significantly higher word count than the first half of the top 10, perhaps indicating that sites with lower authority are ranking highly due to in-depth content, vs. stronger brands that can outrank those sites with less content.
Keywords in Body
Similar to word count, the top 5 sites have a significantly lower number of keywords present compared with the second half of the top 10. This could correspond to the word count (as more words represent an opportunity for more keyword mentions), and similarly, the sites ranking in the top 5 are likely strong brands with more authority that needs less keywords present.
That said, having in-depth relevant content with industry terms (and not keyword stuffing) is the best advice for ranking well with content.
Keyword In External Links
External links that contain keywords are less popular than they were last year. This makes sense — why would a site owner link to a competitor with the keyword in the anchor text? Possibly trying to increase rankings with co-citation?
Compared with last year, top ranking sites link externally with keywords much less than sites did last year.
Social Signals
This section examined the correlation of highly ranking domains with social signals.
On average, websites ranking in the first position had twice as many Facebook signals than pages ranking second, and it sharply declined from there. This could be due to brands ranking in the top spot (as was the case with many of the searches used as data), and big brands investing more in social media than smaller sites ranking after the brand.
Google +
Top ranking sites (again, many brands) had an average of 12,000 +1s, while the rest of the top 30 had an average of much less than 1,000 +1s.
Similarly with Twitter, the top ranked sites had an average of 2300 tweets and retweets, compared with < 500 for positions 3–30.
Pinterest — a much newer social media site, had significantly less social signals present, but the pattern of the top sites having a lot of pins by a factor of more than 6.
Backlinks
Number of Backlinks
Backlinks continue to be a very important factor in determining a site’s rankings, though this is a factor that will likely continue to decrease in years to come.
There is a clear correlation with the number of backlinks a site has and it’s place in the top 30, with the top 5 spots having significantly more backlinks than the next 25 spots.
Referring Domains
While the number of referring domains for sites in the top 15 has increased slightly over last year, the first two spots in the SERPs generally had less referring domains than the rest of the top 30. Again, there is overall a clear correlation among number of referring domains and ranking position.
Backlinks From news Sites
Trusted and authoritative backlinks from news sites aren’t easily acquired for the majority of sites. There is a very big gap between the number of links from news sites in the top 10, and the number of links for news sites for positions 11–30. This is a metric that has increased from last year.
This could also be a commentary on the freshness of a link being valuable.
NoFollow Backlink Ratio
The amount of nofollow links in a backlink profile has increased from 2014 to 2015. Sites ranking in the top 10 had more backlinks than sites ranking outside the top 10.
—
I hope this cheat sheet was helpful. Let me know if you have any questions and I’ll be happy to respond. You can contact me at my site: www.visiblehq.com
