- Nick J. Ocean
As search-focused growth scientists, we wanted to better understand and categorize the internet. We wanted to take the time to ask a few larger questions, and really dig deep into the Data to better understand where we fall in this larger spectrum of the Internet. So we did just that and asked ourselves a few questions:
How many authoritative sites are out there? How much of the internet are reputable publications vs. mommy bloggers vs. spam websites pushing viagra?
Moz and Ahrefs both use a series of metrics to evaluate domain authority. In the larger context of SEO, these are also two of the most important metrics to pay attention to as they most closely correspond with Google's believed value of a site. They are:
Moz Domain Authority (DA)
Ahrefs Domain Rating (DR)
Both range from 0 to 100, and the higher the rating obviously the better off your site is prescribed to be. The following diagram illustrates a distribution of these 1 million domains as they pertain to representative Domain Authority (Moz DA).
Note the 15-18 DA peak. It’s also interesting how the number of domains with high DA tapers off slowly as DA increases (log decay).
Lets now do the same analysis but this time see what this looks like if we’re using Ahrefs’ Domain Rating. Interestingly enough, when we do a comparative analysis of the same 1 million domains with respect to Domain Rating (Ahrefs DR), we see a different graph altogether.
This is a classic normal distribution curve with the middle of the distribution occurring at DR 40. We now start to see some obvious inconsistencies between the two schools of thought here, and it quickly becomes apparent that Domain Authority and Domain Rating are evaluated rather differently.
Moz vs Ahrefs: Quantifying Domain Authoritativeness
When we plot Moz Domain Authority and Ahrefs Domain Rating against each other, se can see a largely predictable trend between the two. It helps us to better understand the relationship they have with each other, but also points out some glaring standouts from the larger data set.
Focusing on the lower end of the Ahref Domain Rating Spectrum, we see some questionable clustering atop the Moz DA Spectrum. Let's take a closer look there.
By closely observing the lower quadrant of sites with lesser DR sites with DA 90, we realize that they are all subdomain.blogspot.tld domains. In this particular case, Ahrefs does a much better job of evaluating how authoritative these individual subdomains are, where Moz gives them all an incredibly high rating largely due to their main root domain's ranking. In one particular case (http://mic-to-mic.blogspot.com), we see a DA of 90 but a DR of only 40. With this in mind, let's now remove all blogspot domains and take another look at the same comparison.
Once we're properly removed the blogspot subdomains, we only see small outliers to the larger general trend as observed above. The Moz and Ahrefs scales converge around DA/AR 60. Generally speaking, for DA’s below 60, DR tends to give higher scores. The opposite also holds true for DA’s above 60, as DR tends to give lower scores.
When we look at the extremes within this graph, we also notice how empty the lower right quadrant is vs. the upper left. Why is it that Moz and Ahrefs rarely disagree in a situation where Ahrefs believes a domain is authoritative, but Ahrefs often disagrees with Moz over many of Moz’s highest DA domains?
A closer look at the upper left quadrant leaves us with a large pool of web companies, such as Weebly, SquareSpace, TypePad, and the rest. Some other notable mentions within this range are Amazonaws, Heroku, Github, and Cloudfront. This therefore leads us to believe that these websites are highly regarded by Moz, but within the Ahrefs system their ratings lose their values.
At this point we at least understand the correlation between Moz DA and Ahrefs DR much better than we did in the beginning, but it's quite obvious that no one metric properly explains the rankings better than any other.