Does referring domain traffic correlate with rankings?

Last updated: October 29, 2021

When building links it makes everyone’s life easier to have clearly defined boundaries of the requirements. Third party metrics have only made this easier with most sellers using Ahref’s Domain Rating or Moz’s Domain Authority as their authority metric of choice. 

But are these arbitrary metric goals making you miss out on links that could be more valuable for your website? 

More SEOs are looking past just authority and using a variety of metrics to assess their links, probably the most popular one beyond authority is organic traffic (as estimated by Ahrefs of SEMRush). With a large portion of the link building market expecting all their links to have a minimum of 1,000 estimated traffic. 

There is some debate whether the metric of estimated traffic is one that you should be looking at when link building and I’m the first to say that while this article won’t paint the full picture as there are so many factors that go into what makes a good backlink I thought it would be interesting to look at it from an analytical perspective. 

Ahrefs recently did a data study showing that 90.63% of content receives no traffic while this is done on a page by page basis and they don’t look at the domain overall. I would hazard a guess that over 90% of the internet will never receive the commonly used 1,000 traffic per month cut off point. 

Does this mean that you shouldn’t build links from these websites when doing link building campaigns? Let’s take a look. 

In this debate there are some people on either side and I want to make it clear that I’m not looking to put words into their mouths and that they’re unlikely talking in absolute terms and realise there are many more factors than just traffic that come into play there are some people on either side of the court. 

For Traffic

Matt Diggity is probably the first person who comes to mind when it comes to making sure that domains have traffic, specifically the domain having over 1,000 estimated traffic according to Ahrefs. 

It’s his number 1 criteria in his link vetting video. He goes on to explain that Google often says that they ignore backlinks and that only a small percentage of websites actually receive traffic (9.37%), they’re unlikely to ignore these links. 

My Take

My take on this is that it makes sense, Google is already rewarding these sites with traffic, which they certainly don’t do for every website. These sites that are being rewarded with traffic are likely to be seen as higher quality websites in Google’s eyes. That being said I really don’t believe that all traffic is created equally. Matt later goes on to talk about website relevance here, so it’s clear that he’s not just looking at 1,000 traffic but 1,000 traffic is his clearly defined cut off point. 

Update: In Matt’s latest video he actually says 500+ but 1,000 is better. So I guess this isn’t such an absolute cut off point anymore. 

Against Traffic

Grind Stone is a very smart SEO who has been in the industry for a long time now, he’s probably most well known for his fantastic SEO case study 2019, which I still think is a must read, even though it’s a bit dated. 

This case study helped to clear up some things which we had been seeing around that time and heavily influenced my talk at Search Birmingham in 2019 (the video was after the event and I was very ill so excuse the coughing if you decide to watch).  

Here’s his Tweet where he talks about traffic being a poor link building metric to use, as you can see below. As you can tell, he’s pretty outspoken and not afraid to say it how he sees it, maybe not the most eloquent but he gets his point across 😅.

In this tweet I believe Grind Stone is talking about selecting a page for a link opposed to selecting a domain to post on (think link insertion versus guest post) so the context is a little different. 

As Grind Stone mentions if you’re ranking in position 5 you’re likely not getting much traffic, the click through rate of position 5 is around 2.97% according to Advanced Web Ranking. Meaning that if there’s 1,000 monthly searches for the keyword cluster and you’re in position 5 you’re only going to get around 30 clicks per month. 

My Take

My take on this is that it also makes sense, especially when you’re choosing link insertions versus doing a fresh guest post. If the page is ranking well for terms related to your topic then it’s probably a good link to have. We recently built a link insertion for a client that ranked for reviews of the product they were selling, the URL stats were hardly impressive but we saw a nice boost in rankings on the page shortly afterwards.

But I think that Grind Stone’s tweet can be expanded further to a domain level. Just because a site isn’t receiving significant traffic according to Ahrefs doesn’t mean that it’s not a good link for you to build. If it’s ranking in the top 30 for some industry relevant, semi-competitive terms I would still happily take it, even if Ahrefs says it’s receiving no traffic. Google is still showing that it considers that site to have some authority in the space by ranking it there, out of the thousands of websites in the subject it’s ranking that specific website in the top few pages.

An example of this would be Truth Naturals, if I was in the CBD niche I would happily take a link off of this website (even if it was on a fresh page like a guest post). 

From a metric standpoint, it’s ok, nothing special DR 28 is an ok link but nothing crazy powerful. It does have some nice authority links (some of them being due to a 301 from http://www.fayethenutritionist.com/). 

But a lot of people would completely overlook it due to it’s organic traffic only being 4, but if you look at the keywords that Ahrefs has it ranking for you’ll see that there are some competitive keywords like “buy cbd oil uk” that it’s ranking 32nd. 

If you search buy cbd oil uk there are 26,700,000 results in Google’s index, out of these 26,700,000 it has Truth Naturals in position 32 so they must see it as some sort of authority in the CBD space.

Even if we change the search to “buy cbd oil uk” there’s 378,000 pages in Google’s index which have an exact match of the phrase. 

Popular Links That Don’t Receive Traffic

There are a few types of links that are being heavily utilised in SEO for their power that often don’t receive traffic, this seems counterintuitive to the idea that traffic itself is a heavily weighted link building signal.

Private Blog Networks

Private blog networks (PBNs) are still one of the most powerful links that you can build and in the majority of cases these sites have no traffic at all. 

If you’re not aware of what a PBN is, it’s a website that is owned by you that uses a previously registered domain which has some historic authority from backlinks. You then rebuild the website which allows you to link to your money site.

The main reason these networks work so well is because of their referring domains to outbound link ratio. To give a very over simplified example, if the website has 100 referring domains let’s say the website has 100 power points, if the website links out to 1,000 external websites each external website will receive 0.1 power points. Whereas, with a PBN you heavily restrict the amount of external links so you might link to just 10 external websites, meaning that each website receives 10 power points. There’s of course a lot more that goes into this equation but this is just to explain quickly why they’re so effective.

Cozab works in a lot of grey niches like cannabis and casino where PBNs are being heavily utilised, some people have tried to adapt and have their network get traffic to their articles but I would say that these are the exception to the rule. But the power being passed by these links is still huge despite the lack of traffic.

Redirects

301 redirects have been being used for a long time now to pass the authority from one website to another. These can be used in more traditional ways like when a website rebrands and they want to redirect all the URLs of the previous domain to the new website. But in SEO there has been an increasing popularity of buying expired or auction domains in the same way that you would for a PBN to 301 to your money site. 

Just like PBNs when you buy these domains they don’t have traffic but when they’re redirected many SEOs are still seeing positive results from them, again suggesting that traffic might not be a heavily weighted link building signal. 

I would say that 2019 was the peak of this tactic and we have since seen a decline in the amount of people using these redirects effectively but there’s certainly still a fair amount of people doing so. 

One of the most popular examples of this from 2019 was Nutshell Nutrition. At its peak traffic I’m certain this site was earning over $50,000 per month in affiliate commissions as I know some of the niches that they were ranking for very well. 

At the times of the redirects, these websites weren’t receiving any traffic according to Ahrefs, but yet they still passed on huge power to allow the website to break over 100,000 estimated traffic in very competitive SERPs. 

Tier 2 Traffic Rebuttal

Some might argue that all of these links that don’t receive traffic have links pointing at them that receive traffic and they derive the benefits from this but this is a bit beyond the scope of what we’re looking at here. We’re just trying to see if (estimated) organic traffic is a good metric to use in your link building efforts or whether we’re looking at the wrong signals. 

Methodology

The methodology is quite simple overall, we’re going to look at the top 3 pages of a few different SERPs and see whether the average percentage of dofollow referring domains (to the domain) over 1,000 traffic is higher on page one than page two and three. Sorry, I can appreciate that it is quite a mouthful to try to explain so I’ve broken it down here.

  • So we first looked at the amount of dofollow referring domains pointing to the domain (the reason that dofollow only were chosen is that these are the ones that pass the most significant amount power)
  • We then found out the percentage of these referring domains that are over 1,000 traffic
  • This gives us our raw figure of percentage of dofollow referring domains over 1,000 traffic that each domain has
  • We will then find the average percentage of dofollow referring domains over 1,000 traffic that each domain has but on a page by page basis
  • We will then compare and see if there is any relationship between the percentage of referring domains over 1,000 traffic and ranking

Chosen Terms

We are going to look at the following terms: 

  • “best online casino” in the UK SERP as this is a SERP that I have experience in gambling SERPs and know how competitive the UK gambling market is. So I’m really interested to see how the results come out on this one.
  • “CBD gummies” in the US SERP, I think we all know how popular the CBD market has been with a huge number of SEOs trying to catch some of the market, this has led to it being super competitive. Gummies are one of the most popular products, which is why I chose this keyword.
  • “best web hosting” in the US SERP, this is another highly competitive SERP. Web hosting websites pay commissions sometimes over $200 per client referred so a lot of people want to rank for this lucrative term.

If traffic to referring domains is a powerful ranking signal we would expect to see that there is a clear relationship with page 1 having a significantly higher average percentage of dofollow referring domains over 1,000 traffic compared to pages 2 and 3. 

Results

Best Online Casinos

As you can see here the percentage actually increases which means that the top ranking URLs actually have a lower overall percentage of their referring domains having over 1,000 traffic. 

Which would suggest that in this SERP that the organic traffic of your referring domains is a weak ranking signal as the higher ranking pages have less. 

CBD Gummies

With this SERP the results are much less clearly defined with page one having a higher percentage over 1,000 traffic than page two but having significantly less than page three. This suggest that there is no relationship between between the average percentage of dofollow referring over 1,000 traffic and higher rankings in this SERP. 

Best Web Hosting

This SERP is much closer, than the other SERPs, with no real significant difference between pages 1 through to 3. The range is only 2%, meaning that all three pages have a very similar percentage of referring domains with over 1,000 traffic. 

Results Takeaway

There really doesn’t seem to be a clear correlation or relationship between the percentage of your backlink profile that has over 1,000 traffic and rankings. In two of the SERPs we looked at, Page 1 had the lowest percentage out of the three pages. I’m not going to come out and suggest that you aim for the lowest percentage to outrank your competitors but I think if traffic was a huge ranking signal that we would have seen the sites on page 1 having a higher percentage overall, which means that there are likely lots of other factors at play. 

Flaws Of Methodology

SEO =/= Links

While links and authority are a key part of SEO overall, they’re only a part of SEO. 

If a website doesn’t have their technical or on-page to a high standard then links will only get them so far. I tend to explain to all my clients that in SEO you are generally as strong as your weakest link. If your website is rife with technical and on-page issues then you should look to fix these as well as link building for the best results.

This only looked at the off-page aspect and didn’t take into account any of the other parts of SEO. So it is limited in that regard. 

Estimated Traffic =/= Traffic

This is probably the biggest elephant in the room, just because Ahrefs / SEMRush or whatever SEO tool you choose to use to estimate traffic, their numbers aren’t accurate. In fact they can be wildly misleading, but they are happy to admit this.  

Here’s a test site that I have, for the past month Ahrefs has reported around 3-4,000 traffic (before dropping like a stone recently).  

Here’s how the analytics looks for the last 90 days, as you can see the traffic has been around 100 per month, nowhere near the reported 3-4,000. 

Some sites they will massively over estimate, others they will massively underestimate. So at the cut off point of 1,000 used in this methodology there will be websites that show as above 1,000 in Ahrefs which are actually under and ones which are under according to Ahrefs which are actually over. 

No Consideration Of Traffic Quality

Not all traffic is created equally but I would argue that this is more a link relevance comment. If a website has 1,000 niche relevant traffic versus 1,000 support keyword spam traffic then I would hazard a guess that the relevant traffic would pass much more value than the celebrity spam traffic. 

There’s no quick and effective way for us to take this into consideration in this article, which is why we’re pointing it out as a flaw. 

Correlation =/= Causation

I feel like as SEOs we can quite often get caught up in correlations especially given how SEO tools that cater to this nowadays. Just because something correlates doesn’t mean that it is the cause. 

There are a lot of other factors at play when it comes to link building that aren’t being taken into consideration in this analysis. The results might be skewed by these factors that aren’t being controlled or measured. 

This is kind of the nature of SEO when there are so many ranking factors it’s impossible to control them all. For example, some of the websites selected might be being held back by their on-site, which if fixed maybe they wouldn’t have lost traffic.

Data Set Size

This was only performed on a very small cohort of sites, I would love to see someone who has the time and is able to assess a bigger data set do so. 

So much is rolling out on a SERP by SERP basis, what I see in some niches is drastically different to others. So I’m sure this data doesn’t apply in some other niches not included in this article. 

The aim of this piece was to start a discussion into the metrics that we use to evaluate links when doing outreach campaigns. It’s very easy to fall into the trap of Goodhart’s Law, which states “When a measure becomes a target, it ceases to be a good measure.” So during our link building efforts we need to make sure we’re measuring what matters most. 

Why I Will Continue To Use Traffic For (Most) Of Our Outreach

While there isn’t a clear answer from this article whether traffic is a link building metric that should be used to determine the strength or quality of a link it’s still a metric that I will continue to use in my outreach campaigns. 

The first reason is I think that the traffic metric is a lot harder for sites to manipulate than authority metrics like Domain Authority or Domain Rating. There are certainly sites that have questionable traffic generating keywords but it’s still harder to rank for these than manipulate the authority metrics. 

The second reason is the reason that I explained earlier, that Google doesn’t reward every site with traffic and therefore, at least to some extent Google must think they’re a domain worthy of receiving traffic. 

But that being said I will often ignore traffic on our smaller sniper style outreach campaigns, where I am looking to really focus on link relevance. 

The example of Truth Naturals mentioned earlier is a perfect example of this, a domain which is hyper relevant to CBD but has no traffic, but is showing promising signs. 

These sniper style outreach campaigns start with a much smaller list, that we will generate by scraping the top 50 results from specific keywords to create our outreach list. The keywords will be hyper relevant to the subject area of the website. 

Luckily you don’t have to choose what types of links you’re building, ones only with 1,000 traffic or ones with under 1,000 traffic. Why not build both? 

It’s worth noting that while relevance is certainly a large part of link building, in each niche there are only so many websites that are directly relevant to your website and of those there’s only a percentage that you will ever be able to get a link on. In competitive niches even if you had all of these links it wouldn’t be enough to rank from an authority perspective because competitors will also have these links plus more general links. 

In the casino niche where we’re very active, if you were just to use casino only domains for your link building you wouldn’t have a chance. All the large players have these links but they also have links from various other niches such as gaming, sports, technology and news. Meaning they will completely out do you in overall authority. 

Takeaways

  • There doesn’t seem to be a strong relationship between ranking on page one the percentage of dofollow referring domains over 1,000 traffic.
  • Using 1,000 traffic as a baseline isn’t necessarily a bad thing and might help to reduce low quality backlinks but there needs to be more to your link building decisions than looking at authority metrics and estimated traffic.
  • SEO is not just link building, and referring domains with traffic isn’t a magic bullet that’s going to instantly move you to page 1. There are hundreds of factors at play and links are just one of them.
  • You don’t have to choose between links over 1,000 traffic or links under 1,000 traffic. You can choose both.
  • When link building, just because a site is under 1,000 estimated traffic you don’t have to discount it, there can be other redeeming factors. Look at other factors like relevance, what keywords it’s ranking for (even if it’s not getting traffic), the backlink profile and the topics the site covers. 

SHARE THIS ARTICLE

OTHER BLOG POSTS