People often ask what metrics I use when looking for domains.
Here’s my long-winded answer, in order of the tools I use.
I use Moz metrics, because even though out of the big 3 they have by far the smallest index, they do have the best metrics. I look at both PA and DA, firstly at DA to see the overall domain strength (DA20+ is worth developing, I personally focus more on 30+ though they are harder to find). Then I look at PA to see where I should launch the homepage of the site (www vs nonwww vs subdomain vs subpage), and how I should setup my redirects (A standard launch note would be “Redirect non-www to www, install 404 redirect and verify the /blog/index.html is redirected”). I used to sell links based on PA but because the link gets recycled into archive eventually, I’ve found DA is actually the better long-term metric.
I use Ahrefs for two thing, the tag cloud on the overview page, and the top pages page. The tag cloud is great because you can quickly identify spam based on a glance at the anchors as well as the percentages (anything with over 50% is spam). Top pages is just the easiest way to check which pages require redirects /wp-installs, whether you need to create subdomains, or whether there was a lot of subdomain spam.
ARCHIVE.ORG Fantastic tool. Parked pages are fine. Often sites will be put up as autoblogs and left and never link built, and these tread the line. If there’s no evidence in both OSE and Ahrefs of any suspicious links, I will often still pick these up. The difference is I will launch them and wait for indexation before developing them. I completely ignore pagerank. I completely ignore Majestic.
Bonus Tip! When you find a clean domain, analyze its backlinks to find more gems. That means a quick scrapebox + xenu search followed by a bulk domain check and a netpeak check for PA/DA.
52 thoughts on “Metrics to look for and tips on expired domain hunting!”
Why do you ignore Majestic entirely? I have a majestic account, but not an ahrefs account… curious how you determine which is better and why.
Ahrefs tag cloud is crucial for me. Think about it a bit and you’ll figure out why.
ahrefs can provide a sitewide overview of all the anchor text crawled in the aged domain. I find it to be very accurate, even better than MOZ or Majestic to identify spam.
great info as always but I don’t ignore Majestic altogether, infact I use it to filter out bad domains with High DA. Sometimes the domain DA would be high but that is due to mainly spammy links like comment spam. Having a Trust Flow / DA combination nicely filters domain like these for me.
I find you throw out the baby with the bathwater a little too often. At this point there are no metrics that quite clearly identify spam, you need to look at anchors followed by archive.org. See my reply to Kurt on how Ahrefs can help with this…
Do you have a script to check domain availability on a large scale or any tips. Scanning several thousand domains 50 at a time is very time consuming.
Thanks in advance
I do, though it’s custom script that begins by checking whois and it has difficulty with soft tlds like .org. You shouldn’t be doing 50 at a time though, dynadot does 1000 at a time, name.com does 2000.
Hey Blaine, read my comment below on Zennoposter – well worth a look for bulk checking domains. If you purchase it you can run multi threaded queries on sites such as GoDaddy, NameSilo with batches of hundreds at a time if you setup right. The program has a learning curve and I have a couple of templates if you fancy a play with the demo.
Nice read, I just passed this onto a colegalue who was doing a little research on that. And he actually bought me lunch because I found it for him smile Therefore let me rephrase that: Thank you for lunch!
Does this mean you no longer look at MozRank and MozTrust?
If you do, do you apply the same principle of domain vs. page, where the domain metric is better for long-term sustainability?
Thanks so much for all of this info, you are the Ron Jeremy of expired domains.
Nope no longer look at it as PA/DA essentially show that anyway (PA is mainly mR and DA is mainly mT plus LRDs).
I guess I’ll take that as a compliment…
hi hayden. Thanks 1st of all for this gr8 post.. I following your way of finding gr8 expired domains… But i have few doubts to clarify :)
1. Do u see this method getting saturated as many ppl doing the same thing
2. Are there still gems that can be found? PR4 and above..
3. I found some domains which looked promising and had good links the the irony is that the rdd is too low ..Most of them are below 10..
But Ur method is so addictive that i have forgotten to sleep hahah..
Thanks a lot :)
1 – Not nearly as saturated as say mining drop lists which has gotten CRAZY STUPID SATURATED.
2 – Absolutely. Through my crawler I find 5-10 domains an hour.
3 – As long as it’s not below 5 or so I’d say it’s ok…
Ya sorry about the sleep. I did give fair warning though!
Michael Bely says:
Thanks for your brief summary of how you look for expired domains.
It is interesting to know why you ignore Majestic metrics.
I’ve found Citation Flow and Trsut Flow quite useful to see if the domain has a weak backlink profile.
If CF/TF is not good enough (less than, say, 10) this is a red flag for me.
So what is your thought why you ignore CT/TF?
And the second question please regarding using scrapebox + xenu (your bonus tip).
There is a check option in Scrapebox that allows to check domains availability and the whole work can be done in SB without using Xenu.
So the whole process is like this
1. scrape the initial list of URLS.
2. extract the external links from these URLs (you get another list of URLS).
3. trim the URLS of the external links to a root domain (you get the list of domains).
4. check if the domains are available for registration (it works not very precisely but it seems that it removes most registered domains from the list).
5. go to bulk domains check to see which domains are really available.
6. check their PA/DA etc.
* and if you have paid MOZ account you can check PA/DA in the scrapebox.
So the work above (including checking domains availability to filter out registered domains) is done in Scrapebox without using Xenu.
So the question is why do you prefer using Xenu for it?
I just answered the Majestic question a couple times in the comments, please see above.
To clarify, I don’t use Xenu. I dislike Xenu, the only benefit it has is being able to go 999 levels deep – if you want an entire site’s URLs it’s easier to use Xenu in some cases than use SB+Automator with various site:sitestructure searches.
I had no idea SB had a whoislookup. Very cool. This is why everyone should read comments :) Also please keep in mind that I have a custom scraper so I’m not really in practice with the SB method – I tried to give some hints in this post but can’t go out and say exactly what I’m doing with the scraper.
Michael Bely says:
Thanks Hayden for your reply.
Yes, SB can check if a domain is available or not, but it is not precise. I use it as a filter before I go and bulk check the domain manually.
P.S; Usually I do read comments. I just wrote mine when there were a couple of comments and no your answers :)
Good info in comments.
Thanks to Hayden’s advice over the last few years I have built up a very strong private blog network with many gems in the PA 40+ range. It is good to know the focus is now DA as PA often differs for www vs non-www. DA is a much cleaner metric. Recently I put to use a commercial grade crawler (which was previously built for something else) for finding good expired domains and there is still a lot of gold out there as I am finding lots of domains available in the DA 20+ range. Keep looking as there are lots of gems out there and I have some extras to sell if you are having a hard time.
Totally agree Lynn.
I’ve had a lot of success using majestic. Can you share why you don’t like it?
I don’t like trustrank/trustflow as they are just not as accurate as PA/DA. I prefer Ahrefs because of the anchor cloud. It’s not that I don’t like Majestic, it’s just that I don’t see a reason to use it when I’m already paying for the other 2 services.
Thanks for sharing these tips. I have a PBN-related question. How important do you think it is to have different whois info for your registrar accounts. For example, I have about 15 PBN sites on a single godaddy account with the same whois info. I have another 10 domains on another godaddy account. It doesnt seem like these sites are passing as much linkjuice as they should.
What is your experience on how important it is to have different whois information? I’m talking in terms of how much linkjuice is passed. Do you think google sees 15 pbn sites with same whois and only giving me linkjuice for one or two of them, much like what would happen if you tried linking to the same money site multiple times from the same website. Thanks.
It is important – vary it up.
Michael Bely says:
what do you think about the opinion that due to the fact that Google is the registrar itself (and therefore it has access to whois information) it is only a matter of time when Google will penalize the PBN regardless of their IP uniqueness?
Just because they are a registrar doesn’t mean they can see all your registrar-specific details (ie billing etc). Use varied registrant info.
Michael Bely says:
awesome post hayden.. But i have some trouble finding gr8 domains..some how for me its getting a bit harder to find the gems..Dont u guys feel this method will get saturated as more and more ppl do this.. And secondly when selecting a domain do u check for drops. Because when ever i find a good domain, they have atleast 2-3 drops:( But the majority ones that i find have mainly has 2 to 3 referring domains with a cf and tf above 10..
Drops don’t bother me, as long as they’re spam-free. I always launch and wait for indexation before building them out.
Is there an automated way to find a niche related expired domain? I mean if my main money site is about ‘car insurance’ then I want expired sites that are somewhat related to that topic and not a DA40 dog food domain, right?
Use Scrapebox to find insurance related keywords and pray.
Greatly appreciate you sharing all of your knowledge with us. It has helped me tremendously.
I know you said you don’t really look at MozTrust Root Domain but if you found the Root Domain to have a MozTrust of 4 or higher even if DA/PA was low would you still buy it?
There are still tons of domains out there.
I just found 175 of them with DA of 20+. 1 40+, 5 30+ and the rest were all in between 20 and 30.
Also are you planning on updating your pricing sheet as I would like to start selling some of these domains?
Send me an email with a list as I may want to buy some / can possibly help you sell some.
I did a good bit of digging but even though I throw out everything <PR5 in SB, the free domains I end up with only have a PA of around 20-25 and DA of 10-15, max. Those with DA 20+ are all spammed to death. Do you just have to be lucky?
Also, do you pick up sites that have nothing else but 1 PR5 link, for example? Or is that just too weak?
I completely ignore PR now.
Glad you mentioned previous parked domains are okay, I was just wondering that.
Hayden I’m finding a lot of domains via Moz with under 15 Total Links and usually 2-6 Unique Linking Root domains.
In particular I found a site with only 2 unique referring domains but those 2 sites are pumping pr 5, pr 4, pr 4, pr 3 links.
Would you go for stuff like this?
No, I completely ignore PR.
If you see any hint of spam, or something like a site repurposed with an Asian language, would you still consider picking it up if it’s still indexed in Google and doesn’t appear to be penalized?
Scott Davis says:
We don’t buy anything with a hint of spam.
Thanks Hayden for helping us noobs out!
I found some domains available with PA around 35, DA around 28. Still, they have fewer than 15 LRDs. Still worth adding to the PBN?
Scott Davis says:
It’s not ideal but it’s up to you.
So if I got this right, it looks like your simply looking for domains with good anchor profiles and high DA/PA with emphasis on DA.
What is the next most important criteria in an expired domain for the purposes of a PBN:
Relevancy to the money site it will be pointed at?
The amount of back links it has?
If you watch the webinar on building out domains you’ll see relevancy can be stretched a bit, but if you can find one actually relevant to your money site then perfect.
Thanks Hayden for sharing your latest insight/experience with the metrics you look at.
Some people asked about automated domain availability check. I have built my own custom script to do this with Zennoposter. It’s an absolutely awesome tool for automation. It basically enables technically savvy users that have a BASIC understanding of programming concepts to automate all their repetative online tasks.
I built an automated domain availability check with it (i just put in thousands of domains, let it run in the background, check back later and the gold in spread out for me accross the floor :D). But, the tool is great for many other things also that can benefit from automation.
It’s not for everyone though, it does have a learning curve, especially if you don’t consider yourself technically savvy. But the benefits from learning will last a lifetime, since this tool is timeless and will maintain usefulness for life.
Hey Hayden this have been driving me batshit crazy because I cant figure out what to do.
I mined a domain that has about 60% of its links pointing to “www.domain.com/en” (a couple pr 6 edu links) and the other 40% at “www.domain.com”.
Should I create the site in cpanel as “www.domain.com/en” and redirect everything else, or create it as “www.domain.com” and redirect the /en?…Or create a seperate page for the /en?
Also is a plugin sufficient for 301 or is going in .htaccess better?
Scott Davis says:
We pick the url with the best metrics and redirect the other urls. Also a plugin is fine.
Hayden, I read where adding the following to the .htaccess file
will redirect dead backlinks:
ErrorDocument 401 /
ErrorDocument 404 /
Is this true, and where in the file do you add it?
Also, you said that you wait for a site to be re-indexed before building it out. I bought this expired domain (above) in Dec. 2013. I plan to do some social and Youtube postings to get it indexed.
How would I know if it has already been indexed? Thanks, Nate
Scott Davis says:
Search site:websitename.tld in google to see if it’s indexed. Also you can just use a wp plugin to redirect 404 pages.
Matt B. says:
I found you through Spencer and now you’ve been riding with me on my commute an hour each way for the last few days (that’s a sort of creepy Beyond Podcast reference).
I’m trying to build another network after dropping a couple grand on the relatively small one that I have now.
You’re right, there’s a HUGE Scrapebox learning curve.
Do you have any resources you’d recommend to help speed up that curve?
Hope to be closer to where you’re at by year’s end. :)
There are a few scrapebox tutorials on this blog. Otherwise the scrapebox youtube videos?
Let’s say I’m mining and finding sites with metrics like PA 20+ but sometimes DA is around 18 but not many links.
Example I found an official health site not spammy back links but only 5 total back links, 3 from 1 domain. The room domain of the back link is PR 8 but the inner page that the site is getting a link from has no PR.
I know you say that you ignore PR but the main reason why I ask is because this domain that I want to purchase only has like 5 total back links.
If it truly only has 5 total backlinks I’d skip it. Use Ahrefs to determine this, or if you use Moz, make sure to get the number from the “Compare Link Metrics” tab under “Root Domain”.
At the risk of sounding like a noob, what do you think of expireddomains.net. I have found and developed a few domains there with decent traffic. Is it the fact of competition and too much exposure? They have filtering mechanisms that make it simple.
Also I understand the reason no one is really emphasizing PR any more but it still has merit. What is a really good explaination of why it is passed over.
Thanks for the video
Would you use this method for finding expired domains for your private link network, and/or for your main site?