Why I value Domain Authority over Trust Flow

[macdonjo3]

My video on how to manipulate any metric has been seen by thousands.

Below, we are looking at a handful of domains, sorted by trust flow. These are TF20+ domains.

I watch countless people run after these TF20 domains like they have some massive value.

This is from a list of 33,000 domains. They haven't been filtered, tis is just the top TF domains.

How can a TF 26 domain with only 2 referring domains be so valuable?

So, which metric is showing how much link juice a domain is pushing?! Domain authority!

This is EXACTLY why I check the domains for my PBN in the following way:
- first I take my list of news featured expired domains
- then I sort them by domain authority descending
- then I check the anchor keywords and archive.org history for spam

So, why are so many people obsessed with trust flow? Because they are lazy, and don't want to check it for spam!

The only advantages that trust flow gives are:
- the Majestic crawler is bigger and more comprehensive than Moz's
- it factors spam and low quality links into the metric

So, now you're about to buy expired domains, you're going to look for high TF and high DA...

Well TOO BAD, those domains are worth a ton, and easily sell in the 3 figure range on GoDaddy auctions.

If you want to find good domains for a good price, ditch trust flow.

Metrics overall, should only be used to pick through massive amounts of domains, but never used as a primary factor.

The number of referring domains, mixed with high authority links, and manually checked for spam, IS KING.

Why referring domains? Because, don't put all your eggs in one basket.

If your 5 of your links get removed, is it going to take a hit on your backlink profile.

Here are high DA domains, check out those referring domains...

 

[Aatrox]

Yeah, you kinda make a good point there. I check both, but then settle on big TF. Noticed this "situation" when I was a building a PBN to sell here. I found quite a few 20TF domains with only ONE referring domains. The reffering domain had only 50+TF and gave all other sites 20TF on the spot. I wanted to buy that, but hell, that was a big risk. If that link was removed, bye-bye 20TF. So I let them be.

On the other note, Moz is slow and their tool is not working properly for me in Chrome at least. Always getting errors after trying to get metrics on a few domains in a row.

Damn, gotta rethink my strategy here. Manually checking for links is still the best thing to do. Thanks for the good info.

 

[ThopHayt]

Makes sense to me, authority is key. Anybody can spam blast a domain. I'd take a domain with a Fox News and Huff Post links and little else over a domain with thousands of crappy GSA links.

 

[MySeoEmail]

So what is the magic number of DA you look for ? What if you have a domain with a DA of 30 and a PA of 5, is that a good one?

 

[Ch3Mik]

Majestic is innacurate for low RD domains (a 3 RD domain can easily get TF30)
Moz is innacurate for low RD domains and you could be missing some gems (they don't crawl shit and less after last November's update)


There are high chances that > DA 20+ domains being found on scrapes have been spammed. Because how Moz metrics work.
There are chances that RD > 20 domains have TF below 10, because how Majestic metrics work.


In short, both metrics suck for our common PBN domains, Ahrefs+Majestic backlink manual check is the way to go

 

[macdonjo3]

So what is the magic number of DA you look for ? What if you have a domain with a DA of 30 and a PA of 5, is that a good one?

I can't remember the exact answer but PA on expired domains is deceiving. I believe it is built to drop to 1 as soon as the domain is dropped. However, DA is maintained a slowly gets smaller. You'd have to read up on the Moz forums about that.

However, about DA that I look for... I don't look for a certain number, but I just sort by it, then I look at the top ones in my list. I don't have a specific yes or no requirement.

 

[archon10]

blah blah nothing to do with Google. If you want to rank in Google, you should value Google metrics.

 

[asap1]

All of these metrics are good in their own way.

But TF is still the best metric to filter domains and see which are worth manually checking.

When I was scraping for domains I would first filter the list by DA, only taking 15+ DA domains then I would filter by TF. Only checking domains with 15+ TF.

When you have a 30k-40k list of expired domains this is the best method I came up with.

Also, 99% of high DA expired domains are SPAM. Anyone that has ever scraped for domains will know the different of finding good 15 DA and 20 DA domains is HUGE!

So yes the high DA metric looks pretty but the link profile is not.

 

[Comrade Luigi]

blah blah nothing to do with Google. If you want to rank in Google, you should value Google metrics.

Are you retarded? There are no Google metrics. PR hasn't been updated since 2013. They keep that to themselves. How the fuck are you going to get the Google metrics when they don't disclose it? This is why many people used metric based analysis as its a substitute to PR.

This thread is interesting. Thanks for sharing. What is the "ideal" domain's backlink profile for you? There are many domains that I scrape that contain say a single backlink from Economist and the rest of the links aren't as good. Or the referring domains are like 7 but the only thing that makes the expired domain worth registering is the authority backlink. What is the average expired domain looks like in terms of the kind of authority backlinks that it has? What is your minimum threshold for registering a expired domain?

 

[Aty]

I have seen great domains with poor metrics, and the other way around too. That's why I don't rely on metrics.

 

[fc-dh]

DA is a crappy metric that is easy to manipulate with sape or 301 or whatever, it is way to inaccurate.

I don't trust none of the metrics but if i had to choose it would be TF, (hell i'll take outdated PR over DA) i have seen gazillions DA 30+ domains with only two links or less...

But in the end it is all about the backlinks, that is why you should manually check your domains instead of being depended of these metrics, you can use them (very) loosely as an indexation, but you always need to do the real selecting yourself...

I have found a lot of great domains that had great authority backlinks and shitty TF/DA metrics and people just let go because they didn't look further, well great for me as they are doing just fine in my PBN's and as monerysites

 

[wokaka]

metrics are just rough reference. it's like the very first layer of filtering. in the end, you still need to manually check their backlink profile through ahrefs.

 

[marshallron]

After page rank no longer maters I think the only make sense is DA, so your theory makes 100% sense.

 

[fc-dh]

After page rank no longer maters I think the only make sense is DA, so your theory makes 100% sense.

Why does DA make sense to you?

 

[umerjutt00]

Well DA is a more bullshit factor than TF. I personally donot prefer any metrics at all but If I had to choose, I would choose TF. Moz scarpers can't find shit then how come their metric can be any trust worthy? On the other hand, majestic is pretty good at finding links.

At the end of the day, you have to rely on the backlinks instead of the metrics. If a domain have good backlink profile, I would get it regardless of the metrics.

 

[sam440]

I also look for Referring Domain and check in wayback machine. I don't believe in TF, CF and other shits.

 

[bignose]

I still find TF much more resistible to spam than DA.

And lately i respect Ahrefs Domain Rank more and more, they are very accurate.

 

[nimdekvan]

I use both (sort for TF first and see if the CF is too low, if not then sort another with DA) and it is fine especially when digging into backlinks profile.

 

[DentalMidget]

This needs to be resurrected for own benefit

 

[safeSEOboost]

This needs to be resurrected for own benefit

Yes as this thread is from last year. It seems many people do not like moz metrics, and prefer majestic.
But DA is still a useful metric.