A single link from a root TF site of 28 vs. 6 links from average root TF sites. This is a continuation of Test #488 where we had the hypothesis that a link from a TF 30 site would beat 3 links from TF 10 sites.
In that test, the hypothesis proved correct and the link from the higher TF site outperformed 3 links from TF 10 sites. A lot of SEOs will look at the quantity of links a competitor has and think, ok, I have to get a certain amount of links to win.
So they count links. We proved in the last test, that THAT was not the case, and quality beats quantity. But, just how many links does it take?
Transcription Below
I'm starting off instead of Clint because it's the one I did. This test is 509: how many links will it take to beat this root TrustFlow of 28? Now, this is an extension of test 488 where we were putting up three TrustFlow 10 sites against – it was actually a TrustFlow 30 at the time and we wanted to see if it would outrank it and it did not. So, what we learned from that is that quality, of course, beats out quantity. If you're a link counter that might not be the best use of your time. We look for quality.
Now, we want to find out how many links it is going to take to overcome that TrustFlow 28 now and that's for the root domain. I'll get into root and URL in just a moment. The test was set up pretty straightforward. At the beginning, they were all on the Recreation topic and I think most of them are now. We linked three of that TrustFlow one was a 10, 10, and a 12 to one page.
Then, the one link of the 30 to the other page to see what happened. Well, the one link with the 30 beat those three links. Since then I've added three more links and the average TrustFlow – here's the root metrics: the top one with the red around it is the site with the link that is beating out all the links from this site. Now, that doesn't mean each one of those bottom ones is going to a separate page. All of these are going to the same page. So, this site is beating the page that all of these are linking to it.
I just want to show you that the links are all contextual, they're all at the top of the page. They all look like this. They are sticky posts so they've not gone down whatsoever. What's interesting though is that the URL metrics are a lot higher than the root metrics.
If you can see it, the top site here the URL metrics is a TrustFlow 40. That's really extremely high and it's beating out these 19, 20, 24, 13, 13, and 19. So, I found that to be really interesting especially – and I'll talk about that back in a moment but, the results are that a single link is still beating out all those other links. Here is that graphic for that.
In my analysis, it surprised me 1) that the TrustFlow metrics had gone up since the beginning of this test because these are all PBNs and these are all PBNs that are five years old. They were created in a linked network in 2015. That explains their diminished root metrics, like the MOZ and Ahrefs metrics as well.
Although, I typically purchase for TrustFlow. I have liked TrustFlow way better than the other metrics and I never even paid attention to MOZ before but I bought a MOZ Pro account just so I could analyze this a little bit more and I could dive deeper. I spent the two days actually on this trying to figure out why these sites are not beating that one site.
So, I have this chart here where I put everything in an Excel file. You may not see it but you can see this one site down here it beats the top site and most of the categories in all three analytical tools. So, this is interesting and maybe some people can dive into this and give me more of an explanation of what's going on here.
We're going to still continue to link to that one page and see how many it takes to overcome this TrustFlow of 40, PA of 20, UR of 16. The root metrics aren't that great although it's a TrustFlow 28 DA of 10 and a DR of 3. The DR 3 still beat out the other sites. As I said, these are old PBNs.
I was really digging into this and the screenshots here, like in MOZ, I had different metrics from when I took them one at a time and when I did them one at a time that's what's in this chart, the Excel files screenshot, but when I would do the compare their metrics changed a lot.
So, I found that to be really interesting. Like, the total links that it says coming into the site are in the thousands and none of the tools – even when I did MOZ when I did it singularly, did not say it had these thousands of links coming into the site so I felt like that was really weird. Also, the internal followed links – and I'm not really sure if that means links coming into the internal pages from an external source or internal links that they're following in the site. Clint, maybe you know?
CLINT: Those are internal links in the site that are set to dofollow.
DORI: Okay. I know that these are old PBNs but they don't get posted to very often because these sites are not really in a network that's being used. So, I thought the thousands which are almost the total links, it's counting the internal follow links and the external links to get the total links. Is that what it's doing here?
CLINT: Yes.
DORI: That's interesting and I think it would also be confusing. It confused me because I'm not a MOZ user. It wasn't obvious to me so I thought that was a little bit strange.
That's the root domain. Then, I have a screenshot for the exact page which I, typically, want to rely on more because Google ranks pages. They don't rank URLs. The root domain, the DA, and the DR really doesn't mean all that much. All that is an internal calculation of how they think these sites would rank in the search engine.
You can see over here that one site has a domain authority of 14 whereas the one that's winning is only a domain authority of 10. So, it should be beating that out. In any case, I'd like to look at the exact page. That's why I like to look at the URL metrics instead of the root domain metrics. I thought that was interesting and also the internal links followed on those URLs. I know that there are not hundreds of links on that homepage. There's a maximum of 10 articles with three links in them so that would be 30.
So, there's that. I also thought maybe there's some internal linking playing here because I know that both MOZ and Ahrefs will use internal linking to score for the root domain and actually for the URL but they don't show that to us. In any case, here is a screenshot of that site with the TrustFlow of 40 it only has one link coming into it and that link has a TrustFlow of 13.
That's really like, "What? How can that be?" I know there are not these pages because it's a PBN that is linking to the homepage except in the navigation. Could navigation play that important part? Have people been linking to these sites? I don't know. That was a big question mark for me and I started getting a little wary about Majestic here, too. Somebody could say, "Well, look at the historic. Maybe it's getting from that." But when you look at the historic view it's a TrustFlow of zero so that's not it.
This is the root again, and it shows like there are some great links coming into the root, into that homepage. Is that giving it the TrustFlow 40? I don't know. There's a 33 and a 23, a 6 and a 14. Then, there's some other lower ones and then it went to zero because this site doesn't have that many links going into it anymore.
I spent a lot of time trying to figure this out like where is that coming from and I haven't been able to really yet so I'm going to do some more investigation. The bottom line is this: we still don't have enough links to beat that one single site and I'm going to continue to add more. I started to lose faith in the tools all of them because there was nothing that really showed me why this was happening and why that page is ranking #1. So, I will keep digging and more will be revealed.
We do know quality beats quantity but why is that site – it had that much quality? Clint, do you have any thoughts on that?
CLINT: I have always had the same issue with all of the tools just because they're so easily manipulated. UR rank is easily manipulated. PA and DA were always easily manipulated with just trackbacks and ping-backs. You can do that with that.
Then, TrustFlow was the trusted metric that I was using filters through especially when we were looking at link targets but when you see that, and I've seen tons of instances where that number just doesn't make sense. Like, if I had one link and that link had a TrustFlow of 90 then you rewarded my page with the TrustFlow of 40 I got it.
But if I have three links and they're all like yours: 30s, 19s, 0s, and 0s that makes zero sense of why my TrustFlow would be that. Is it taking into account the relevancy of the anchor text? Those are things that we don't know and I doubt MOZ or Majestic or Ahrefs would ever put out the true "This is how our database is manipulated in order to come up with these numbers”.
DORI: Yes, I looked at all the anchor text in these sites and decided not to put them in the test because it would just make it long and they were pretty much benign. Just a lot of brand links, no exact matches, no heavily like that so there wasn't anything that stood out to me, in other words, when I was looking at that.
CLINT: Lauren had a good question too. Is the internal links part of that TrustFlow measurement? It very well could be and I think looking at the counts and knowing that they're probably all WordPress sites – you can correct me if I'm wrong, and I'm not sure if you have the data archives, the author archives, the category archives, and the tag archives turned off. If those aren't turned off then that explains your high count of internal linking.
DORI: Then, that would be a way to manipulate it. I believe the archives may be on some of the sites. I don't know about these sites but I remember seeing them on some other sites. That is actually something I can go check. That would be really great. That's why I mentioned maybe the navigational links were giving that site-wide stuff back to the homepage but still... I do know that internal linking counts a lot.
That is something that Clint and I have talked about and it's not really talked all about all that much in the SEO industry, that how internal linking plays and how important it is. But when you think that it only has really two links coming into the site, where is that juice coming from? Except if all those pages are producing links that are going to give it juice.
Now, we do know from earlier tests that two or three internal links are counted as external links. So that, Lauren, you could be dead on there that that really could be what's happening. I'd like to test that more to really get a better understanding of why and how that site got such a high TrustFlow.
CLINT: All right. One thing I like about your test is it proved my theory that domain rank is trumped by URL rank for Ahrefs. I stopped following domain rank a long time ago and I only use UR when I'm doing my filters now and that has been working like gangbusters for me. The numbers that you pulled manually the UR rank tracks with the ranking positions of the pages so that's pretty cool.
Go back to your screenshot there. If you look at UR, see it's 16, 1, 24, 14, 27 that's an outlier, in my opinion. That one right there, that 16, the highest UR is ranking.
DORI: Yes, it would make sense if all of these pages were shooting links to different pages and this was the one that won, the 16, and the 23 would be an outlier but all of these links are going to that single page and I checked it to make sure too that they were indeed active on top of the page and I did find one – there were supposed to be seven links going in, that did not have the link. It just didn't have the link so I pulled that out. That's not even part of the conversation right now.
So, all of these sites are pointing to one page. That 16 is beating all of these. Same with that TrustFlow too, the highest TrustFlow. Maybe in an of couple days, it could change. We'll see. I'm going to continue to add links to it and we'll see next month where it stands and I'll look at that internal linking and see if those archives and all that stuff's up on the site.
CLINT: Another factor is also how often those companies are updating their link database. I know MOZ is probably like six months, almost a year now, of them actually updating their database to make it work right because before it was really small. Now, it's like 6. Something million because people will call them out on it.
DORI: [Majestic] they are crawling pretty often. I mean, you see some of these pages were crawled July 28th. This one, may 13th was the last time it was crawled. This one was July 3rd, July 2nd.
CLINT: What I would do there, just for my own education, is just run that in Ahrefs and see when their last crawl dates were to see how that all washes out. So, who's updating faster et cetera because maybe you're missing stuff that is not being reported yet.
DORI: That's true. Although, that one site is still ranking and that Google updates pretty often.
CLINT: One more question would be on the pages that are being tested, are they all the exact same in regards to the title tag, H1s, URL, and keyword density in total?
DORI: Yes, they're mass-produced test pages by my team. We create all our test pages. They're lorem ipsum. They have the same exact amount of everything. The lorem ipsum isn't identical on the pages but the word count is and the keyword count is as well but that's something I can go back and make sure. I can check.
CLINT: The only reason I bring that up is that Ted and I have tanked tests because we had things in there. We were testing a really weak factor and it was getting out and I'm outdone by the stronger factors. URL, title tag, and H1s are typically the stronger ones that we're messing upon.
DORI: That's a good point. I will go back and look at that. Thanks.
But dang! I want to figure TrustFlow out, the metrics, the tools to get a really clear view of the links that we're buying. Just everything.
CLINT: Well, I did something with Ahrefs and then Ahrefs threw a monkey wrench because we were using DR 40. You get the DR 40 links and you can rank all day long.
Then, they decided to change the way their algorithm works and that algorithm ~ initial change dropped the DR and the UR for pretty much every site on the web to like zeros and really low cap stuff, and people got pissed at them and said something and so they updated it again to get everyone's numbers like a happy medium for everyone to stop complaining, is essentially what they did. Now, DR is pretty much useless.
DORI: Yes, because we use that as a barometer as an SEO. As long as we know what it is it's like that's our barometer. As you said, I got a DR 40 and I was good all day long. Then, when they changed that barometer it's like, "Oh god, no. We gotta start over." So, I get that.