SEMrush Ranking Factors vs. MOZ Search Engine Ranking Factors (plus 7 key take aways)
Recently, SEMrush announced the release of their new SEO ranking factors report. I don't know about you, but I love reading these kinds of things. They're really interesting and give great information on ideas for tweaking how we do things.
Overall, I was surprised by some of the findings. The ascendance of UX and the demotion of keyword optimization was eye opening for me. Tying the user experience and engagement to SEO has long been a challenge for many people to grasp. Making flat statements like "good user experience depends on search optimization" go nowhere. It takes exploring what makes a good user experience and how that aligns perfectly with the goals of search engine optimization in order for most people to see the connections. But that is a longer conversation than most one hour meetings allow for. Now we have the data that SEO benefits from good UX (just as we know that UX benefits from good SEO).
The SEMrush summary slide (8). My annotations on the left.
7 Key Takeaways from SEMrush Ranking Factors Study 2.0
If you don't have time to download an read the full report, here are my key takeaways:
Back links are critical. The number and quality of the links are especially important, but keywords in the anchor are not as important.
Get your site on HTTPS. This is non-negotiable any longer.
Content length should be around 600 words per page. Long tail topics need about 20% more words.
Keyword placement is less important than expected. Still, make sure you have the title and body text optimized. This is especially true for short (more competitive) keywords. Everything else is a "nice to have".
SEO depends on UX. Good engagement signals a high quality website:
Bounce rate (50% or better)
time on site (210 sec. or better)
Pages per Session (3 or better)
A website is still just a bunch of pages. Each page needs to be optimized to get to #1. While we think of organic search as a channel that supports the whole website, the nuts and bolts (and the most impact) still reside at the individual page level.
Brand always beats topical. "When the user searches for keywords containing a brand name, the brand domain will appear higher in search results than Wikipedia or other popular resources."
What’s up with Direct?
There is one confusing result in this report: the direct entry connection. Direct is another word for "dark", meaning the referring data is missing. Normally this is assumed to a strong brand affinity indicator. However, that assumption may not be entirely correct. Some have postulated that increases here are related to the rise of mobile (since apps don't pass referring info). This one needs more exploration, in my mind, to be a usable. It seems to have a lot of opportunity to be an important insight if it can be unraveled a bit.
The gaps I'd like to see filled
There are some missing factors in this report. Perhaps the Random Forest method used in this study did not surface these as important, but it would be nice to see them addressed somehow. Factors that have been on the SEO radar but don't show up in this report:
mobile vs. desktop
H1 as an on page element
social and sharing
How does this compare to MOZ?
How does this compare to another very popular industry perspective--the MOZ Search Ranking Factors report? (note the most recent version is 2015)
MOZ is a combination of expert opinion (150 people) and search ranking data (17,600 keyword results) while SEMrush is a straight up review of data (600,000 keyword results). MOZ has an excellent reputation as one of the industry's best resources for learning and research. Their MOZ tools are used by SEO's everywhere. How does their research compare?
A screenshot from the MOZ report. My annotations on the left.
It looks like MOZ gives more credit to linking and keyword usage and less credit to UX. This may be because Google's Panda algorithm has continued to gain importance since the MOZ study was released in 2015. Or it may be a simple difference of opinion between authorities. Either way, the gaps are good to know.
Things they do agree on--to do good SEO you must focus on links, UX and keyword optimization. Pretty straight forward. Let's get to work!