While these changes were being rolled out, we closely monitored the reaction of the community and saw a lot of positive feedback:
But these kinds of changes never go entirely smoothly, so there were quite a few people who disliked this update.
When analyzing the negative feedback, we noticed that it was revolving around a few common concerns:
“I have a legit “authority” website that ranks for competitive terms and gets good search traffic. I just got the new DR of 8 and it is total BS.”
“I have links from Forbes and many other “authority” websites, how come my DR dropped to 5??”
“While I support the DR change, you’re now clustering too many websites in DR 1–5 range, making it impossible to differentiate between them when doing link prospecting.”
“How do I explain my clients their new DR? They will think that I did a bad job building links for them.”
“Now that I better understand how you calculate DR, I can’t seem to find good use cases for it in my day to day SEO work.”
I’ll gladly address all of these things and help you persuade your clients that there’s absolutely nothing to worry about (as long as their traffic and rankings are going up).
But to do that, we’ll need to go back to the very basics…
What is Domain Rating? (hint: it’s not what many people think it is)
Like I just mentioned, a lot of people were connecting their DR score to the keywords that their website ranks for, the search traffic it gets and some kind of “authority” that it has.
So, when they saw a sudden drop in DR (while all these other metrics stayed the same), they could only conclude that it was a “miscalculation” on our end:
Well, please be advised that Ahrefs’ Domain Rating is NOT a measure of “website authority” and it is not calculated off your search traffic or rankings.
To put things simply…
Domain Rating is a metric that shows the “link popularity” of your website compared to all other websites in the world on a scale from 0 to 100.
No more, no less.
And I used quotation marks around the term “link popularity” because it refers to a rather specific way that we calculate it here at Ahrefs.
I did explain that calculation in a previous article, but here it is again for good measure:
As you can tell, there’s nothing about traffic, rankings or even “link spam.” The calculation is quite straightforward and easy to understand.
So please don’t draw lines between the DR of a website and anything else but the websites that are linking to it.
A few more common questions that I need to address:
Q: “I have a link from forbes.com which has a super high DR of 93. How come my own DR didn’t move a bit?”
A: “As explained above, the amount of ‘DR-juice’ that Forbes will pass to you is divided by the number of all websites that forbes.com is linking out to (with dofollow links). And they’re linking out to nearly 100k other websites. So you’re better off growing your DR via a link from a smaller website that doesn’t link out to others so much.”
Q: “But aren’t the pages that a domain is linking from relevant to the calculation? Does it matter for Ahrefs’ DR if I get a link from Forbes.com homepage or some article deep in their archives?”
A: “Unfortunately, this does not make any difference for DR. To calculate value that is being passed from a specific page to a specific page we have another metic in Ahrefs — URL Rating (UR).”
Q: But why then don’t you calculate DR based on URL Ranks of all referring pages to a given domain?
A: Well, at the present moment, to calculate DR we look at how domains are interlinked. And there are 174.6 Million domains in Ahrefs’ database, which are being used in this calculation. As for pages, we have 189.5 BILLION of them in our database, so you can probably tell how using them in DR calculation would ramp up it’s complexity (and cost).
Q: “Why then don’t you calculate DR of a domain as a sum of URs of its own pages?”
A: “That is truly a good idea and it should be much easier and cheaper to calculate than the previous one. We did try it before, but stumbled upon a few roadblocks that didn’t have any simple solution. So we had to put that idea on hold.
What is DR good for? How to use DR for SEO?
Even though search traffic is not taken into account when calculating Domain Rating, the two correlate very well.
Which is of course expected, because websites with tons of search traffic tend to get a lot of backlinks. And vice versa.
But, at the same time, it is perfectly normal for DR10 website to have more search traffic than DR20 website. That happens because DR is not the only contributing factor to your search traffic.
A lot depends on:
- The total search traffic potential of the industry each website belongs to;
- The search traffic potential of the topics that each website is covering;
- The level of competition for these topics in each niche;
- The quality of content and on-page optimisations;
- General website quality (design, speed, UI/UX, etc);
So, as I said earlier, Ahrefs’ DR is purely a measure of a websites’ “link popularity” (as measured by Ahrefs), but it can be very useful when you look at it alongside some other sitewide factors.
And if we’re talking about DR as an isolated metric, here’s what it is good for:
1. Benchmarking against the rest of the world
Our own website (ahrefs.com) is ranked at position #3,923 according to Ahrefs Rank.
This means that we’re in the top 4000 websites in the entire World by our “link popularity” (as calculated by Ahrefs).
When we plot Ahrefs Rank on a scale from 0 to 100, we get a two-digit Domain Rating of 85 for ahrefs.com.
Both AR of #3,923 and DR85 paint a nice picture of how ahrefs.com compares to the rest of the websites in the world by its “link popularity”. But, at the same time, these numbers are not immediately actionable.
2. Benchmarking against your competitors
Our main competitor (popularity-wise) is moz.com, which has an Ahrefs Rank of #576 and Domain Rating (DR) of 91.
Now our own numbers don’t seem that impressive, right?
Comparing your DR/AR to other websites in your niche can give you a lot of additional insight about the “link popularity” of your own website. When we compare ahrefs.com to moz.com, we realize that there’s still a ton of untapped link opportunities for us.
In other words, DR is a good metric to discover competitors who’re doing better than you. Then, you can piggyback from their backlink profile (which is easy to do via the Link Intersect tool).
And if you’re in a position of the leader, you can keep an eye on that DR gap and monitor if your competitors are catching up with you.
3. Link prospecting
When doing link building, you need a way to vet your link targets and estimate the “value” that you’re going to get in exchange for your efforts.
DR naturally seems like a good metric, since it is meant to show the relative “link popularity” of a given website. This should then translate into the amount of “value” that you will get if they link to you.
But there are quite a few caveats in using DR (as well as any other sitewide “authority” metric) for link prospecting.
So we’ll discuss it in a separate section of this article.
Why did we change DR scores?
Because YOU asked us to.
It was very common for our support team to get questions like:
“My website is DR45, and my competitor is DR40. Why can’t I outrank them?”
“I just got a link from DR50 website, why don’t I see any increase in rankings?”
“I have 100+ backlinks from DR30-50 sites, why my search traffic isn’t growing?”
These kinds of questions made us realize that:
- A website with DR over 30 was perceived as being “strong”.
- A difference between DR40 and DR45 was perceived as quite dramatic.
- A website’s DR of 50+ was supposed to positively affect it’s rankings/traffic.
- A link from DR50+ website was considered a “strong” link.
But none of these things were true (with the old DR scale).
We figured that there were two big issues that were misleading our customers:
- The DR scale that we used to have didn’t work the way people thought it worked;
- There was lack of information about how we calculate DR and what we advise it should be used for.
As you can tell, we’re doing our best to fix both of these things.
Introducing the new DR scale
Like I said, the old scale wasn’t very accurate. The main issue was that the majority of the ~170 million domains in our database were clustered in DR20-50 range. This made “small” websites look “big.”
How often did you see a domain with DR of less than 10?
I bet you didn’t see any at all.
A brand new website with a handful of backlinks immediately gained DR30+, making its owner quite happy and satisfied (especially if he/she was a link seller).
But we wanted to have an objective DR scale; one that would accurately represent the “link popularity” of all websites on the World Wide Web.
So with the release of the new DR calculation, the majority of websites (that were previously clustered at DR20-50 range) were moved to DR0-30 range. This is where they objectively belong compared to the likes of Twitter, Facebook, Google, etc.
Here’s how ~170 million domains in our database are distributed on 0–100 DR scale right now:
As you can tell, ~83.9 percent of all domains are now clustered in DR0‑5 range. Because this is where they belong, according to their backlink profile.
So let’s put this graph on a logarithmic scale to make it easier for you to make sense of all DR ranges:
And here are the numerical values of how many domains we now have in each DR range (and their average number of dofollow ref.domains):
- DR 0–5: 146,455,043 domains (1 d.ref.domains)
- DR 6–10: 10,322,431 domains (30 d.ref.domains)
- DR 11–15: 5,171,577 domains (44 d.ref.domains)
- DR 16–20: 3,085,066 domains (57 d.ref.domains)
- DR 21–25: 2,499,437 domains (85 d.ref.domains)
- DR 26–30: 2,493,952 domains (137 d.ref.domains)
- DR 31–35: 1,377,630 domains (547 d.ref.domains)
- DR 36–40: 861,493 domains (314 d.ref.domains)
- DR 41–45: 494,599 domains (263 d.ref.domains)
- DR 46–50: 313,324 domains (352 d.ref.domains)
- DR 51–55: 268,988 domains (444 d.ref.domains)
- DR 56–60: 152,985 domains (708 d.ref.domains)
- DR 61–65: 93,021 domains (956 d.ref.domains)
- DR 66–70: 64,713 domains (1,374 d.ref.domains)
- DR 71–75: 64,898 domains (2,789 d.ref.domains)
- DR 76–80: 21,209 domains (6,761 d.ref.domains)
- DR 81–85: 7,368 domains (16,425 d.ref.domains)
- DR 86–90: 3,175 domains (36,811 d.ref.domains)
- DR 91–95: 967 domains (637,650 d.ref.domains)
- DR 96–100: 22 domains (8,891,919 d.ref.domains)
IMPORTANT: please keep in mind that DR is not calculated from the raw number of dofollow referring domains alone (see details above). So these numbers of d.ref.domains per cluster are just raw approximations. They only serve as a point of reference and not the actual number that is required to reach certain DR score.
IMPORTANT #2: please read what is written in the box above and please don’t ask us why your DR doesn’t match your dofollow ref.domains (or vice-versa). That’s because the DR calculation is a bit more complicated than just raw number of dofollow ref.domains.
How to tell if the new DR is better than the old one?
Let’s compare the difference in DR between ahrefs.com and bloggerjet.com (my no-longer-updated personal blog) before and after the change:
Before the change, the difference between the two was only 15 points. And I really don’t think my small personal blog deserves to be this close to Ahrefs (which is in the top 4000 websites in the world according to Ahrefs Rank).
So now the difference is 30 points (on a logarithmic scale!) which seems to be much more fair.
Q: We had DR50 and then dropped to DR18, and now we’ve settled at DR43. What does it all mean?
A: Like I said, there was an update and then there was a “patch.” Based on the feedback from the community it felt that we have “punished” smaller sites a little too hard. So we went back to the drawing board and figured how we could use 0 to 100 scale a bit more effectively.
What about 146 million websites in DR0‑5 range?
Like I said earlier, with the new DR scale ~83,9% of domains in Ahrefs’ database are now in DR0‑5 range.
To be precise:
- DR0: 104,218,641 domains
- DR1: 18,767,526 domains
- DR2: 9,145,844 domains
- DR3: 6,175,538 domains
- DR4: 4,515,899 domains
- DR5: 3,631,595 domains
That seems like a fair granularity, considering that most of these website hardly have any good backlinks.
But some of our customers requested that we give even more granularity and add decimals to this DR range.
Unfortunately, this kind of change would require a lot of effort from our team. And we believe that this time and resources would be better spent creating some other Ahrefs’ tools and features. So we had to put this on hold for now.
However, as a temporary solution, we still managed to add decimals to Site Explorer “Overview” report:
And, besides, you can always reference Ahrefs Rank when analyzing smaller sites, which offers much greater granularity.
The new DR has nothing to do with “Link SPAM”
In my original announcement of the DR update, I mentioned a few spammy websites and what would happen to their DR after the update.
This example “tricked” a lot of people into thinking that our DR update was targeted at spammy websites only. And, as long as your website is legit, you won’t see any drop.
Later in that same article I had an entire section dedicated to “DR vs SPAM”, where I said that:
But I guess a lot of people didn’t read this far.
So that confusion is totally my fault, sorry.
IMPORTANT: to avoid any further confusion, let me outline the part of DR calculation that we have changed in this update:
In other words, we didn’t change the way we count links from domain to domain or how we assign “raw” DR values (and build Ahrefs Rank off them).
What changed is the way we plot these “raw” DR values onto 0–100 scale. Because the previous DR scale had some glaring issues and wasn’t really objective.
And while it is true that with the new DR score, it would be quite hard to pump your DR to 40+ via cheap spammy links, growing your website to DR40+ via white hat links won’t be much easier.
In other words, spammers can still “pump” DR with low quality links, but it is much harder now, because of a better DR scale.
How to explain the DR drop to your SEO clients
If you were using Ahrefs’ DR metric as part of your SEO reports to clients, you might have a hard time explaining the change (especially the drop).
So let us do this for you (just point your clients towards this part of the article).
If you’re seeing a drop in your Domain Rating (DR), it doesn’t mean that your SEO agency has been doing a bad job building links to your website.
As I explained in the section above, the DR re-calculation wasn’t meant to “uncover” any questionable link building tactics. The goal was to put every website in the world in the right place on a 0–100 scale according to its “link popularity” (as calculated by Ahrefs).
If you see a big change in your DR, it doesn’t mean that your SEO agency was doing a very bad (or very good job) with link building. It only means that your previous DR wasn’t quite fair.
That said, this DR update affected absolutely all ~174M websites in our database (some went up, others down).
If you want to better understand whether your SEO agency is doing a good job with link building, we suggest you look at things like:
- Total search traffic to your website (over time);
- Total number of keywords you rank for in top100 (over time);
- Your ranking positions for a bunch of most relevant keywords that you want to rank for (over time);
- The “quality” of websites that the links are coming from;
- The “quality” of pages/content that the links are coming from.
If all the above look good (and you don’t have to be an SEO expert to analyze them), then your SEO agency is doing a good job.
At the end of the day, you’re not paying your SEO agency to grow the Ahrefs’ DR metric of your website. You’re paying them to bring more paying customers to your business via traffic from Google.
How to use Domain Rating for link prospecting
The problem with using DR (or pretty much any other sitewide metric) for link prospecting is that Google never said anything about the benefits of acquiring a link from a “popular” website.
Google’s PageRank algorithm works by passing value from a page to a page, not from domain to domain (or from a domain to a page).
So let’s do a quick poll:
If you ask me, I voted for the link from a higher-UR article, but lower-DR domain.
So does this mean that using DR for link prospecting doesn’t make sense?
Well, yes and no.
DR is a near perfect metric to help you vet the “link popularity” of your link prospects (given the size of Ahrefs’ backlink index and the frequency of updates).
High “link popularity” basically means that there’s a decent chance that a page on that website will generate some backlinks and have high-UR.
But “chance” ≠ “guarantee”.
I mean, there’s always a chance that a page on a low-DR website will generate backlinks and get high-UR. As shown in my poll above.
So, we do recommend that you use DR for linking prospecting. But we don’t recommend you to use it in isolation from other factors.
Some things you want to consider alongside DR are:
- What does their backlink profile look like? (DR can still be inflated, remember?);
- How many pages do they have? (The more pages they have, the less internal link value each of the pages has);
- How many websites are they linking out to? (And how many outbound links do they have per page, on average);
- What’s the overall quality of the website and their content?
- Will this website be around in 5 years? Will it grow or stay the same?
Will there be more changes to DR (or any other Ahrefs’ metrics) in future?
A lot depends on whether YOU need us to change anything about the current DR (or any other one of Ahrefs’ metrics).
Like I said earlier, we changed DR not because we wanted to, but because we had to—it was your feedback and your questions that prompted us to do this.
And I already feel that you will keep pushing us forward, no matter what:
I can’t say if we’re going to make more changes to DR in future or not, but what I can tell you is that we’re experimenting with a few other interesting metrics that could potentially be a perfect complement to DR as it is today.
You might have noticed that we didn’t quite “nail” the DR update on our first attempt.
After releasing v2.0 we got some feedback from our customers (with examples, arguments and reasoning). This allowed us to create a “patch” and release v2.1.
And we’re quite happy that it went how it went. This made a lot of people pay attention to our Domain Rating metric and helped our team make it better than we initially planned.
If you’ve been Ahrefs customer for a while, you may know that we’re very receptive to feedback from our users. So if you see any way we could make Ahrefs tools/data even better, you can use this handy little service to submit your feature requests.
PS: I wonder if we just made Ahrefs’ DR “the most thoroughly explained metric in SEO”? 🙂