1. This website uses cookies to improve service and provide a tailored user experience. By using this site, you agree to this use. See our Cookie Policy.
    Dismiss Notice

How to Choose Great SEO Domains - Our Experience from Many Years in Business

Discussion in 'Black Hat SEO' started by bignose, Aug 20, 2018.

  1. bignose

    bignose Jr. VIP Jr. VIP

    Joined:
    Dec 6, 2009
    Messages:
    977
    Likes Received:
    180
    Gender:
    Male
    Occupation:
    Domaining
    Home Page:
    Hello guys.

    So, first of all: How to find good domains?

    Good domain is defined by its backlinks. Only backlinks define whether the domain is good or bad.

    So we check backlinks and the site history. And not any metrics.

    Rule #1 of checking domains:

    All Metrics Don’t Matter. No Metric Matters. Only Backlinks Do.


    [​IMG]

    And also the Site History. Whether the site has some spam history, some PBN history, some redirect history, some doorway history, and so on.

    But the most important - backlinks.

    So, - how do we check backlinks? We take a domain, and check its backlinks. It's a manual work. There is no automatic way to do it. We can automate, robotize different stages of the process, but not actual manual checking of domain's backlinks. This is the most important part. And if you delegate this task to robots, to scripts, to automatic programs and software, - here you will fail. Because only a human eye is able to evaluate whether the backlinks are good or not.

    But all the preparation work you can entrust, delegate to bots. So here we use SEO metrics. SEO Metrics are the mere sign of – whether it’s worth to spend time checking the domain’s backlinks. Nothing more than that!

    [​IMG]

    The more we rely on SEO metrics, - the less we are accurate, the less good domains we can find, and the more good domains we miss.

    And if we don't trust SEO metrics at all, we miss zero. But this would be very time consuming. To check all 100K daily drops one by one, would take probably a week and your sanity.

    So we take some SEO filters, some metric filters, and using those filters we can decide whether it's worth actually to take the certain domain and to check it. And the tighter filter we have, - the more chance we have that we can miss some good domain and even some great domain. Because often time good domains just don't have good SEO metrics.

    Bottom line, we need to find some balance - how much time we can dedicate to domain checking, daily.

    So I choose, for example, to spend two hours daily, and to miss a certain amount of good domains daily. But I don't regret it, because I still need time for other tasks and jobs, I need time for my life, for my family.

    This is basically the first and the most important rule we all should remember. Often times we have these revelations of SEO Gurus, long reads, long blueprints, long PDF's, and they can even take huge money for those, and they say for example: "You need to take domains with DA 22+, or 15+, or whatever, and a quantity of .edu backlinks should be minimum 3. Or 10. And then your site will be at the 1st position in SERP. In Google." And then I can exactly say when some of these SEO Gurus reveals, or sales, his new revelations. Because we are starting to get many requests of the same kind: "We need domains with DA 25+ and .edu backlinks 10+." Why?? What do people want to get with domains chosen by metrics? - Nothing. They don't know. They just read some SEO Guru stuff.

    Don't be like these people! Remember: Only Backlinks Matter.


    So what we should do - is to take our domains from whatever list and to check their backlinks in Ahrefs.

    And how do we choose them? By names? To go domain by domain through 100K names? As I said, it could take a week.

    So we use some SEO metrics. And here, SEO metrics are our tool, and nothing more. Our tool for filtering domains, to free our time spent on domain research.

    There are lot of SEO metrics. Like Ahrefs, Majestic, Moz, SimilarWeb, etc. And there are never ending arguments - which SEO metrics are better, are more relevant for defining whether a domain is good or not.

    We find that DA, Domain Authority by Moz, is quite OK for filtering the results; maybe it's not very relevant to the real domain power, but it's actually quite good for cutting off bad domains.

    I personally love DR, Domain Rank by Ahrefs. It's very good. But - there are lot of good domains that have low DR, so we cannot really rely on this metric.


    We can take endless, limitless number of metrics - we can take all metrics by Moz, all metrics by Ahrefs, all metrics by Majestic, and then we have SimilarWeb, we have Alexa, all other stuff; and if we have like 50 different metrics and put them into the Excel sheet, we have a mess, we have a huge mess that we just cannot operate.

    So we need somehow to limit these metrics, because if we have like tens metrics, then our eyes, our sight can't really take something from there. From another hand, one metric would clearly be not enough, we need to have some broader picture. Just imagine, if we had only DA for all of these. You would have to open every domain and just guess.

    [​IMG]
    VS.

    [​IMG]

    So: The more data you have, the more time you save. But not too much. Not too little, but not too big data.

    So we have our own set of reliable metrics. We take our domain list and get those metrics for these domains. And then we compile an Excel sheet with this data.

    [​IMG]

    I don't tell you how to get these metrics, because that's what everybody decides for themselves. The obvious way is to go and purchase all the subscriptions of all these services - Ahrefs, Majestic, Moz, etc., - but it might be very costly, it can be more than a thousand dollars per month, and just not everybody's business can afford that. So there are different sites, different tools, different services that gather this data and give it to you for some subscription price or whatever.

    Which metrics do we have in our sheet? First of all, Moz DA. Moz metrics are not as relevant, not as powerful as they use to be few years ago, but still, as I said earlier, DA is a very good indicator, and it’s very good for cutting off bad results.

    Then we sort results, and in this dataset, I use cut-off by 20. DA 20 is my Floor, my lowest point where I want to spend my time on this domain. And I know I may miss here some domains, because there are still lot of domains with DA like in 15 to 20 range which might be good, but time I would spend on investigation those domains will be huge. So I decide for myself, that I don't take domains with DA lower than 20. Your personal floor may be different.

    [​IMG]

    Then in order to clear some clatter, we delete some insignificant values. So, another cut-off for me is 24. It's my personal number of choice, you can choose your own, of course, like 15 and the secondary cut-off 20. I use 20 and 24. So from 24 I just delete the values from the sheet. What does it give to me, it diminishes an amount of clutter. The sheet becomes easier to work with. But here, every domain which doesn't have DA, I know it's still 20 to 25. I just don't get them visible, so they don't bring distractions to my sight.

    [​IMG]

    Then we have Majestic. And this is our secret, which I can reveal now to you. Actually, everybody who have ever bought domains from us, Domains R Forever, knows it, that we use www for Trust Flow. And I don't think than anyone else, any other service, does it. And it's strange. If you read some SEO Gurus, they might say that TF-www doesn't matter, because it shows only some subdomain rank. And technically it's true, www is subdomain. So maybe in some sense they are right, for some SEO stuff it's like this, but for searching of domains TF-www definitely matters. Because if you take a domain and check its metrics, you can very often see that TF-www is much higher than TF, so you can say that most backlinks, while the site was live, lead to its www version, that's why it's higher.

    [​IMG]
    Examples of TF-www > TF

    So if you don't have this in your dataset, you will miss actually a lot of domains. I tell you: a lot. And here, with my method, you actually get them.

    So this is my second secret. The first secret is: Metrics Don’t Matter. And the second secret: Always use TF-www.

    Actually, I don't know any service that gives you this metric, so you should find it somehow.

    And here we also make some cut-off, I take for example 5. A couple years ago TF was much more relevant. If I remember well, it was about 18 or 20, so below 20 it was not worth. But then, Majestic somehow changed their algorithm again, and many good domains fell to 0-5 range. So actually, now 5 is not so bad.

    And TF-www we also make from 5.

    [​IMG]

    And then we have another great metric - indexing in Google. Many people are searching for indexed domains. It makes perfect sense, but. Many good domains out there are not indexed but have quite a good metrics, - many of these domains are de-indexed not because of some sanctions, or some bans by Google, but they are de-indexed just simply because they were sitting there idle without any site, for long time. Just because of that they've dropped from the index. And in most cases, when you take a good domain and build a good site on it or re-build a site on it, it gets back to index very fast. It's my secret number 3. In most cases. Not 100% cases, but most cases.

    [​IMG]

    So you should not worry much about domain being de-indexed. Of course, if some other signs are bad, for example in the domain history there was some redirect or stuff, maybe it's quite toxic. But if everything looks good, and it's de-indexed - take it. As I said, in most cases it'll jump back to index soon. If not, you always can file for reconsideration.

    It actually depends on what site you do build on this domain. On the quality of this site. There might be different kinds of sites. If you take this domain for building some good site, white hat site, informational site, and so on, with good content, useful for visitors, you can really not worry about zero in the Index cell. You can always send for reconsideration, you can say, "I'm not responsible at all for what was there with this site before me, I just bought this domain."

    But if you build something weaker, something worse, like for example PBN site, it will depend on the quality of that PBN. If you built your PBN on copypasted content, or scraped content, or machine generated content, you obviously cannot send for reconsideration. And if you build a site for example for doorway, you of course cannot send for reconsideration. And if you don't build a site at all, but use this domain for 301 redirect, - you just cannot take de-indexed domain.

    So, this is in a way paradoxical: The better site you build, the less you can worry about its indexing or de-indexing. And vice versa.

    But even if the domain is in index, you are not guaranteed in any way that it won't drop from index. So it's always a kind of lottery.

    And then we have a very important one: the Age. In our case, the age is the first appearance in the web archive. Archive.org works from '96, and few times I stumbled on sites from '95. But there are many sites that existed even before '96 or '95, and archive.org just don't have that their history.

    It's an important metric. It's not an important metric for your SEO, for Google results it doesn't matter. But it's an important metric for us, for our decision to look at the domain.

    [​IMG]

    Then we have a set of data from Ahrefs. First of all, as I mentioned, - Domain Rank. For me, it is the most relevant metric among all. But for, so to say, preliminary filtering DA serves better. Still DR is a very important landmark.

    Good domains are usually start from DR 20. But many good domains have lower DR, down to even 3. So what we do here, we cut off from 2, and up from 20 we put in color. Actually you can format your sheets as you like, the most important - to make your data visually clear and intuitive.

    [​IMG]

    And then we have another important one - RD, Referring Domains by Ahrefs. You can use Referring Domains index from Majestic or from Ahrefs, they show more or less the same. And here I use obvious cut-off on 20. Even 30-40 often times is not worth it. But sometimes you have domains with really good backlinks, like .gov or .edu backlinks, but they have only like 20 referral domains at all. I would take such domain any day. All the domains below 20 I delete.

    [​IMG]

    And then I also delete all these empty rows below, and now my slider is here, at the bottom. So it's now much easier to work with this sheet, you can use scrolling.

    [​IMG]

    And then we have some very secondary metrics. We can skip them, but they give some indication. This is Ahrefs .gov and .edu referring domains. They don't say us much, because all these .gov and .edu backlinks might be a junk. There might be university forum profiles, might be redirects, or whatever.

    Also, Ahrefs keywords. This is not an actual Google index, especially for drop domains these keywords are not relevant. But is a nice indicator anyway.

    [​IMG]

    And I don't take any other metrics. Many people love SimilarWeb. But SimilarWeb metric can be easily inflated. If you take really good domain, you see that SimilarWeb rank is low. But if we take just any domain with low SimilarWeb rank, the chances are the domain is just bad. It's easy to manipulate it. Much easier than with DA for example, and of course than with DR. The same with Alexa, and many other metrics.

    Also many people like to use CF/TF ratio. You can see that I don't use CF at all. I just don't like it, I never could see any meaning in it. But this CF/TF ratio can be really good. I just don't like to have it here in my datasheet, and also with our two variants of TF, with and without www, it can be really difficult, because now we would need to have also two different ratios, with and without www. And there will be lot of visual clutter here.

    Some people like to use another ratio - Ahrefs Domain Rank to URL Rank. It's also good idea, but this set is really enough for me.

    And here we have Alexa index. It's also can be easily manipulated, so if you have low Alexa for some domain, it doesn't mean that the domain is good. But when you see many good metrics, and Alexa is low as well, you can see that the chances the domain is good are very high.

    [​IMG]

    So I've explained you everything we have in our setup.

    Then I make the following sorting. First of all, Age from oldest, then Alexa from lowest, then TF from highest, and in the end DA from highest. But you of course may use any order you like. So now, I have my final file I work with.

    [​IMG]

    In the video, I show many examples of how I really work with investigating the domains. It would be difficult to tell it all here, so you might want to watch the video or to read the full transcript; here I'm only able to explain some main points.

    So the scheme is like this: we seek in our sheet for domains with consistent metrics;

    [​IMG]

    then we take these domains and check them in:
    • Ahrefs for backlinks
    • Ahrefs for anchors
    • Archive.org for toxic history
    • Google for what is there in the index

    DA. Very rare you can find really good drop with DA > 50. Because that just doesn't happen. And if that does happen, you cannot really grab such domains because of course Dropcatch or Snapnames will grab them, and we see there auctions for thousands dollars for such domains. But often times these DA are wrong, because of redirects. If the domain was redirected on the last stages of its life, then its DA might show DA of the target domain, and not of this domain.

    Of course, Ahrefs is very expensive to use. If you are on a low budget, you can use some alternatives, like Majestic, which I think is cheaper but for me is not so good, not so handy to work with, and the results output is much less intuitive. For example, in Ahrefs I see all domains sorted by DR and I immediately see if the domain is good or bad (of course then I still check archive etc.):

    [​IMG]

    And in Majestic it's difficult to understand what you are looking at:

    [​IMG]

    Or, if you are on really low budget, there are some free alternatives, like Openlinkprofiler or Linkpad, which you can start with. And of course then, when only you can scrape some budget, you should buy Ahrefs subscription.


    So this is how I check domains. The more domains you check, really every day, the more experience you get, because I can see if the domain is good or not only by looking at the referring domains, but it's an experience of years.


    A never ending discussion: is it okay to take domains that were used for PBN? As for me, if the domain is really good, I would take it for myself, for my projects, I don't mind PBN. My position: backlinks are great, backlinks are backlinks, backlinks are there, I don't care what one of the previous owners of the site, of the domain has built on it, what site, I don't care, I can use it for my projects. But many people don't like such domains. And I think it's a kind of common misbelief. Our vast experience tells that PBN history is a very insignificant factor for domain power.


    So, as I said, you can find more stories and domain cases in my video. Here I could only highlight some theory. I hope you enjoyed it, and I wish you all lot of success. Would be glad to answer your questions as well.
     
    • Thanks Thanks x 16
    Last edited by a moderator: Aug 20, 2018
  2. tg01

    tg01 Power Member

    Joined:
    Aug 23, 2007
    Messages:
    720
    Likes Received:
    229
    Gender:
    Male
    Thanks for the in-depth share, one that I’ll have to revisit!
     
    • Thanks Thanks x 1
  3. meenakshi1163

    meenakshi1163 Jr. VIP Jr. VIP

    Joined:
    Jul 30, 2018
    Messages:
    185
    Likes Received:
    31
    Gender:
    Female
    Occupation:
    Internet marketer
    Location:
    Internetland
    Great share. Like you said, it involves manual checking. I think this can be outsourced to VAs
     
  4. Nargil

    Nargil Jr. VIP Jr. VIP

    Joined:
    May 10, 2012
    Messages:
    6,956
    Likes Received:
    4,611
    Location:
    Europe
    Home Page:
    I agree with basically everything. Thumbs up especially for ignoring TF CF ratio.

    DA20+ is a way too massive filter. I consider it completely unreliable and I just took a look at my list and out of 2400 TF5+ RD10+ (and more) domains that I have and are surely clean, only 30 have DA of 20 or higher. Sure, such domains can be found, but not with the limited amount of resources. Most of them have been spammed to death already.

    And I also disagree with the fact that former PBNs can be used. I've seen some very toxic results from such domains. Also, there's always a very good chance of registering a penalized domain, that will end up to be merely a waste of money. For churn and burn projects, yeah I guess, why not. For anything mid - long term, nah.
     
    • Thanks Thanks x 4
  5. fullwall3t

    fullwall3t Elite Member

    Joined:
    Apr 22, 2015
    Messages:
    1,873
    Likes Received:
    356
    Thanks for the guide. What is your practice if you use dropped domain for money site?
    What do you do with 404s?
     
  6. bignose

    bignose Jr. VIP Jr. VIP

    Joined:
    Dec 6, 2009
    Messages:
    977
    Likes Received:
    180
    Gender:
    Male
    Occupation:
    Domaining
    Home Page:
    I highly doubt so. In my many years experience with different VA's, most VA'a are good for some strict, pattern, mechanical tasks, but always make things only worse whenever you start talking about any creative or analyzing work. VA is basically the same bot, only with slightly higher intelligence. Sorry if i sound politically incorrect, i'm just talking business wise.

    If you can train your inhouse staff - then yes. But that would be totally different story budget wise etc.

    Hi @Nargil , thank you for coming by to comment!

    I appreciate your experience, but i described here my experience. I've seen very good results with such domains, even with sites filled with CN/JP junk.

    Thank you for your input! I will definitely experiment with your filter set.

    For money site, first what you need is a good name, and the name domains are totally different market from SEO domains. Should be a brandable name, marketing wise, or at least EMD/PMD. Because if you have a domain with good backlinks and metrics but ugly name - the users will not come.

    If you happen to find such good name domain with previous backlinks - that's a huge bonus of course.

    Sorry, can you explain better what you mean, please?
     
    • Thanks Thanks x 2
  7. fullwall3t

    fullwall3t Elite Member

    Joined:
    Apr 22, 2015
    Messages:
    1,873
    Likes Received:
    356
    Well when the site has xxx backlinks, they lead to pages/post that don't exist anymore. Do you keep that way, or do you use redirect to your homepage?
     
  8. bignose

    bignose Jr. VIP Jr. VIP

    Joined:
    Dec 6, 2009
    Messages:
    977
    Likes Received:
    180
    Gender:
    Male
    Occupation:
    Domaining
    Home Page:
    Right. Now i get you. Thank you for the question.

    No, you should never redirect to homepage. Since a few years ago, G treats such redirects as "soft 404", whatever this could mean. This makes a huge footprint and this way you are loosing the whole power of backlinks.

    Indeed, if backlinks are what makes domain good, you should use them all.

    So for example, you get a domain with valuable backlinks to let's say 100 pages. Then you restore/re-create/build from scratch in an ideal case all 100 pages, but it depends on your resources, time, etc. Or let's take another example: the domain has valuable backlinks to 1,000 pages. You create on the site 100 pages, covering more or less all the most important topics from the old site.

    Then, you take the list of the most valuable backlinks, and make a sheet of redirect dispatch, sending each backlink (see by backlink url, anchor, linking page) to the most relevant page of the new site. Or to more or less relevant page, so that all new pages get some of the old backlink juice. That's all. Then you import this sheet to Redirection plugin if you use WP, or otherwise implement it somehow to .htaccess or whatever.

    Of course if there are only handful of backlinks, you simply make it all manually in Redirection plugin.

    Hope it helps.
     
    • Thanks Thanks x 2
  9. GregFromMoonsy

    GregFromMoonsy Power Member

    Joined:
    Aug 12, 2018
    Messages:
    571
    Likes Received:
    201
    Gender:
    Male
    Location:
    European circus
    Completely nonsense OP.

    Google use set of algorithms to evaluate backlinks, where the main role play Pagerank.

    If you don't know how it working, how you want to verify it manually ?
    Because site looks good ?
    Or maybe backlinks looks good ?

    Looks good NOT EQUAL is good.

    For this reason, in over 90% (or even much more) this method will fail.
     
    Last edited: Aug 22, 2018
  10. drstew

    drstew Jr. VIP Jr. VIP

    Joined:
    Nov 5, 2008
    Messages:
    1,158
    Likes Received:
    272
    Gender:
    Male
    Location:
    DownUnder
    Home Page:
    Nice post mate.

    Don't worry about metrics, but worry about www-TF ? For me, Majestic's multiple TF versions is BS, but Root TF is a cumulation of everything including all backlinks to sub-domains and folders so Root TF is the best gauge I think.
     
    • Thanks Thanks x 1
  11. privaterate

    privaterate Registered Member

    Joined:
    Jul 22, 2018
    Messages:
    67
    Likes Received:
    8
    Good share
    Thanks for the method
     
  12. AngelSeo

    AngelSeo Jr. VIP Jr. VIP

    Joined:
    Nov 3, 2017
    Messages:
    1,589
    Likes Received:
    880
    Gender:
    Male
    Actually 301 redirects of old pages works great - you simply need to use some common sense:
    1. Recreate the strongest pages (with most links) and do proper internal linking.
    2. Redirects all the rest (with links) to homepage, internal pages, sitemap.
    3. Enjoy.

    About the metrics for initial filtering - I found that 5+ DR for expired and 10+ for auctioned/dropcatch work the best for me. DR is simply measuring the strength of the do-follow backlinks according to Ahrefs (they ignore no-follow), so 5+ DR means that most of the time there are at least some good links. Like Nargil said 20+ DA leave way too much good domains out of the picture, although with last update MOZ become a bit more relevant. I ignore Majestic metrics completely (use them just to check the anchors for a bit extra info), but agree with you that www-TF is what you need to look at as well.
     
    • Thanks Thanks x 1
  13. PixelMathis

    PixelMathis Senior Member

    Joined:
    Jun 10, 2014
    Messages:
    1,195
    Likes Received:
    472
    Gender:
    Male
    Occupation:
    Mad Methodologist
    @bignose said it's important to look at a domain's backlinks manually as well as using tools like Ahrefs or Majestic. How do you vet a domain - by hacking into Google's backlink database and sorting backlinks by PR?.

    You haven't shared a single method in the forum and I constantly see you downplaying others' contributions. I'm sure you don't even own a Godaddy auctions account. Get off your high horse.
     
    • Thanks Thanks x 3
  14. PixelMathis

    PixelMathis Senior Member

    Joined:
    Jun 10, 2014
    Messages:
    1,195
    Likes Received:
    472
    Gender:
    Male
    Occupation:
    Mad Methodologist
    Have you ever recreated a formerly-PBN site as a money site? If so what did the results look like?
     
  15. AngelSeo

    AngelSeo Jr. VIP Jr. VIP

    Joined:
    Nov 3, 2017
    Messages:
    1,589
    Likes Received:
    880
    Gender:
    Male
    Nah, he share his method - he is looking at Semrush keywords and traffic historical data.
     
  16. PixelMathis

    PixelMathis Senior Member

    Joined:
    Jun 10, 2014
    Messages:
    1,195
    Likes Received:
    472
    Gender:
    Male
    Occupation:
    Mad Methodologist
    That's a pathetic way to vet a domain btw.
     
  17. AngelSeo

    AngelSeo Jr. VIP Jr. VIP

    Joined:
    Nov 3, 2017
    Messages:
    1,589
    Likes Received:
    880
    Gender:
    Male
    I know :D But hey if is working for him - great.
     
  18. PixelMathis

    PixelMathis Senior Member

    Joined:
    Jun 10, 2014
    Messages:
    1,195
    Likes Received:
    472
    Gender:
    Male
    Occupation:
    Mad Methodologist
    He can do whatever he wants to but at least he shouldn't crap all over on threads made by other members.
     
    • Thanks Thanks x 1
  19. bignose

    bignose Jr. VIP Jr. VIP

    Joined:
    Dec 6, 2009
    Messages:
    977
    Likes Received:
    180
    Gender:
    Male
    Occupation:
    Domaining
    Home Page:
    Right, you for sure know better how to interpret my experience for the last 8 years. Good luck bro.

    Thank you for your input!

    I don't pay much attention to Majestic either, however it gives a bit of a nice visual presence when you look at your sheet and try to catch domains with more or less consistent picture.

    Well i think a had a few; i would restore a site based on the existing backlinks/topics/anchors, then always would wait until G catches it fully, and only then start slowly adding some content and changing the site topic towards the new niche. Granted the new niche is in the same vertical. When you don't hurry and can allow yourself a few month for a smooth topic transition, most times i didn't face any troubles. Well couple of times this strategy failed, but you never know for sure anyway.
     
    • Thanks Thanks x 3
  20. godknowseverything

    godknowseverything Regular Member

    Joined:
    Oct 13, 2014
    Messages:
    479
    Likes Received:
    737
    Occupation:
    Coiner
    Location:
    Planet SEO
    1. If a domain has been redirects in the past (301/302), Is it okay to use them for a money site? @bignose @Nargil

    2. Also, I've read it's better to use a non-dropped domain for money site, What are you thoughts here?