- Joined
- Nov 28, 2014
- Messages
- 16
- Likes
- 0
- Degree
- 0
Can anyone recommend a good place or person to get domains from for a PBN without getting raped. Looking for good metrics and decent non spammy backlink profile.
Help with buying domains for a PBN?
- Thread starter Paul Bryant
- Start date
Golan
BuSo Pro
- Joined
- Nov 13, 2014
- Messages
- 406
- Likes
- 206
- Degree
- 1
- Joined
- Oct 14, 2014
- Messages
- 87
- Likes
- 66
- Degree
- 0
If you're not against a bit of manual work, I've had alot of luck manually scanning sites using Xenu.
Input site into Xenu, watch for "no such host" on the output (export to tabulated data and insert into Excel for easy sorting), check which are not regged and bulk enter them into tools such as http://bulkdachecker.com/, and sort by DA. Further sort from there, using Majestic's free FF plugin, which lets you see metrics for an unlimited number of domains (one by one, however).
May be a lot of work, but I've had some luck doing this. Lots of DA20-35 domains, with TF hovering around 17-23. Slightly off topic, but thought i'd throw it in there.
Input site into Xenu, watch for "no such host" on the output (export to tabulated data and insert into Excel for easy sorting), check which are not regged and bulk enter them into tools such as http://bulkdachecker.com/, and sort by DA. Further sort from there, using Majestic's free FF plugin, which lets you see metrics for an unlimited number of domains (one by one, however).
May be a lot of work, but I've had some luck doing this. Lots of DA20-35 domains, with TF hovering around 17-23. Slightly off topic, but thought i'd throw it in there.
Golan
BuSo Pro
- Joined
- Nov 13, 2014
- Messages
- 406
- Likes
- 206
- Degree
- 1
Sure it's a lot of work. If you value your working time, you'll see in the end that the time you've spent for digging out one free domain like this isn't worth it; you could just buy such domain on the market.
If you are serious in this game, you must invest to automation.
- Joined
- Oct 14, 2014
- Messages
- 87
- Likes
- 66
- Degree
- 0
It's obvious to anyone that automation takes out the work you're doing. That wasn't really my point.
If you're not savvy enough or want to/have the money to spend to either build it yourself or have someone do it for you, my above post can be a solution that actually works.
If you're not savvy enough or want to/have the money to spend to either build it yourself or have someone do it for you, my above post can be a solution that actually works.
- Joined
- Nov 28, 2014
- Messages
- 16
- Likes
- 0
- Degree
- 0
Jäger,
What site do you suggest plugging in? Like godaddy you mean?
Golan,
Can someone like me (a novice) operate these automated programs?
What site do you suggest plugging in? Like godaddy you mean?
Golan,
Can someone like me (a novice) operate these automated programs?
- Joined
- Oct 14, 2014
- Messages
- 87
- Likes
- 66
- Degree
- 0
Ideally you'd plug in a site that has authority -- something like BBC, Techcrunch or similar (whatever niche or industry you're targetting). Problem is, lots know this and custom made tools will most likely do a better job at this than Xenu, so for me I go more niche and try to think of (and research) sites people probably won't be targetting.
And for the really low hanging fruit, I've actually had some luck plugging in good ol'e directories. Just make sure it's a quality directory and be ready to sift through a shit load of crap
- Joined
- Oct 10, 2014
- Messages
- 21
- Likes
- 11
- Degree
- 0
I've been using the Xenu with numerous different strategies for a very long time, but it's getting incredibly hard to find great domains. Everyone and his mother is using it right now and you will hardly find any diamonds right now.
Some might beg to differ, but I bet they only look at crap metrics like DA and (to a much lesser extent) TF. Sure, it's not hard to find DA> 20 domains, but if you actually look at the backlink profiles, you'll see they are mostly crap. Heck, I have an excel sheet with over 10,000 expired DA>20 .NL domains (the only ones I'm interested in), but they're not up to my standards (anymore).
There is actually a way to find really, really good domains which is overlooked by almost everyone. The reason is because they're mostly DA<10-15 domains and everyone automatically ditches them during a bulk DA shift. It only struck me a month ago, but in that time I've found hundreds of AWESOME domains. When I have more time in a couple of days I will write an tutorial on this site.
But let me say this right now: What if a domain has a backlink from a genuine PR 5 page with an OBL of -say- 5?! That's a genuine PR 4 (!!!) site right there, but the DA and TF might very well be under 10 because it only has 1-2 other backlinks. What if you start to disregard these crappy external metrics and find a direct way to scrape expired domains from these high PR pages?!
Some might beg to differ, but I bet they only look at crap metrics like DA and (to a much lesser extent) TF. Sure, it's not hard to find DA> 20 domains, but if you actually look at the backlink profiles, you'll see they are mostly crap. Heck, I have an excel sheet with over 10,000 expired DA>20 .NL domains (the only ones I'm interested in), but they're not up to my standards (anymore).
There is actually a way to find really, really good domains which is overlooked by almost everyone. The reason is because they're mostly DA<10-15 domains and everyone automatically ditches them during a bulk DA shift. It only struck me a month ago, but in that time I've found hundreds of AWESOME domains. When I have more time in a couple of days I will write an tutorial on this site.
But let me say this right now: What if a domain has a backlink from a genuine PR 5 page with an OBL of -say- 5?! That's a genuine PR 4 (!!!) site right there, but the DA and TF might very well be under 10 because it only has 1-2 other backlinks. What if you start to disregard these crappy external metrics and find a direct way to scrape expired domains from these high PR pages?!
Golan
BuSo Pro
- Joined
- Nov 13, 2014
- Messages
- 406
- Likes
- 206
- Degree
- 1
Sure, all metrics are shit. TF is the most accurate of all, but is gamable anyway. What matters are backlinks - their profile, anchors, clearness. No metric can show you this.
If we speak about .gov backlink - it matters whether it's from official article or some "subcontractors for our annual shit whatever".
If we speak about .edu backlink - it matters whether it's from an article about the latest research or official standard, or some student blog link dumpster with 1000's of other links, or even non moderated blog post with thousands viagra karen miller outlet shit.
If we speak about wiki backlink - it matters whether it's a link to official site or just one of hundreds relevant "external links" at the bottom.
If we speak about NYT or Forbes or other newspaper backlink - it matters whether it's editorial link in serious article, or some private guest opinion or a comment or whatever.
No metric takes this to account. But Google does. So must you.
Of course if you scraped a nice list of 100.000 - even just 100 - free domains, you need to filter them somehow preliminary, in order to know which ones even worth manual checking. Otherwise you can spend a month checking them all, in vain. This is basically the only real use of all SEO metrics. (For yourself. If you need this domains to sell links, of course you also need to show nice metrics to your buyers.)
I can share my process of this filtering.
For each free/dropping/etc domain list, I grab the following metrics:
1. PR
2. DA
3. TF
4. TF with www
5. Age (first archive.org appearance)
Then I sort it by:
1. Age ascending
2. DA descending, delete all values below 4-5
3. Both TF descending, delete all values below 4-5
4. PR descending, delete all values below 0
So I get this nice sheet:
You can say a lot even by looking at the numbers. For example, if DA is a lot higher than TF, it's a perfect sign of spam (as TF is much more spam resistible). But not always.
Also, unusually high DA could be a sign of redirect. These you just check by entering this domain to OSE.
Then, if you see high metrics but the domain is quite new, 2012+, it's most probably a manipulation. It should really good site with strongest backlinks to get grown to this positions in the last couple of years.
So, from this list I have only three domains that worth checking.
But the lower we are moving down the PR scale, the better domains there are.
You see? Here I would check almost every domain. Even one of 2013 may be good, who knows. Here the criteria is a name. If it's some EMD or PMD or other SEO shit like thebesthostingreview.com or cheapuggbots.com, obviously I pass. But if it's some real name, I mean one that a real person or business could give, something nice or brandable, I always check them.
When I put my eye to a domain because of it's high PR, I always go to archive.org and see what it was there in the last couple years, as this PR could be possibly a result of redirect. But if other metrics are also high and backlinks are good, I don't really care if there was redirect or not, if the PR is fake. I just don't waste my time for archive.org check in this case.
The real tzimmes, but also the real mince, starts downstairs. Tons of domains that ain't nobody got time fo that.
Here I use two criteria: age and name.
Age: the older the better.
Subjective factor: Google may give to older domains some authority, some slack. One can't prove or deny it; you can believe it or not; I do believe in that. More than just believe, I have tons of experience of successful rebuilding of old domains.
Objective factor: odds for clearness. If the domain is old enough, the chances it got to hands of spammers at some point of it's life, are low. Also it hardly would be registered and used from a start for some spam purposes. There were not spam schemes, link pyramids, other shit back in 1998, right?
Next: the name.
I always try to look thru the entire list, even of hundreds or thousands domains. I look for nice names, interesting abbreviations, etc. It takes time, but I think I already trained my eyes to catch these domains even on a high speed of scrolling.
And to finish with this, another my rule: No TF is low. People are looking for DA20+ and TF15+; but I saw really tons of sick domains with DA<15 and TF<10. Some firms, organizations, gov departments, int congresses. With sick .gov or wiki links. They just don't have many other general links, thus have low DA and TF. But I will always prefer them over to high-metric domain with dozens of "just regular" links.
Well, I hope this is all. Sorry for taking the thread
If we speak about .gov backlink - it matters whether it's from official article or some "subcontractors for our annual shit whatever".
If we speak about .edu backlink - it matters whether it's from an article about the latest research or official standard, or some student blog link dumpster with 1000's of other links, or even non moderated blog post with thousands viagra karen miller outlet shit.
If we speak about wiki backlink - it matters whether it's a link to official site or just one of hundreds relevant "external links" at the bottom.
If we speak about NYT or Forbes or other newspaper backlink - it matters whether it's editorial link in serious article, or some private guest opinion or a comment or whatever.
No metric takes this to account. But Google does. So must you.
Of course if you scraped a nice list of 100.000 - even just 100 - free domains, you need to filter them somehow preliminary, in order to know which ones even worth manual checking. Otherwise you can spend a month checking them all, in vain. This is basically the only real use of all SEO metrics. (For yourself. If you need this domains to sell links, of course you also need to show nice metrics to your buyers.)
I can share my process of this filtering.
For each free/dropping/etc domain list, I grab the following metrics:
1. PR
2. DA
3. TF
4. TF with www
5. Age (first archive.org appearance)
Then I sort it by:
1. Age ascending
2. DA descending, delete all values below 4-5
3. Both TF descending, delete all values below 4-5
4. PR descending, delete all values below 0
So I get this nice sheet:
You can say a lot even by looking at the numbers. For example, if DA is a lot higher than TF, it's a perfect sign of spam (as TF is much more spam resistible). But not always.
Also, unusually high DA could be a sign of redirect. These you just check by entering this domain to OSE.
Then, if you see high metrics but the domain is quite new, 2012+, it's most probably a manipulation. It should really good site with strongest backlinks to get grown to this positions in the last couple of years.
So, from this list I have only three domains that worth checking.
But the lower we are moving down the PR scale, the better domains there are.
You see? Here I would check almost every domain. Even one of 2013 may be good, who knows. Here the criteria is a name. If it's some EMD or PMD or other SEO shit like thebesthostingreview.com or cheapuggbots.com, obviously I pass. But if it's some real name, I mean one that a real person or business could give, something nice or brandable, I always check them.
When I put my eye to a domain because of it's high PR, I always go to archive.org and see what it was there in the last couple years, as this PR could be possibly a result of redirect. But if other metrics are also high and backlinks are good, I don't really care if there was redirect or not, if the PR is fake. I just don't waste my time for archive.org check in this case.
The real tzimmes, but also the real mince, starts downstairs. Tons of domains that ain't nobody got time fo that.
Here I use two criteria: age and name.
Age: the older the better.
Subjective factor: Google may give to older domains some authority, some slack. One can't prove or deny it; you can believe it or not; I do believe in that. More than just believe, I have tons of experience of successful rebuilding of old domains.
Objective factor: odds for clearness. If the domain is old enough, the chances it got to hands of spammers at some point of it's life, are low. Also it hardly would be registered and used from a start for some spam purposes. There were not spam schemes, link pyramids, other shit back in 1998, right?
Next: the name.
I always try to look thru the entire list, even of hundreds or thousands domains. I look for nice names, interesting abbreviations, etc. It takes time, but I think I already trained my eyes to catch these domains even on a high speed of scrolling.
And to finish with this, another my rule: No TF is low. People are looking for DA20+ and TF15+; but I saw really tons of sick domains with DA<15 and TF<10. Some firms, organizations, gov departments, int congresses. With sick .gov or wiki links. They just don't have many other general links, thus have low DA and TF. But I will always prefer them over to high-metric domain with dozens of "just regular" links.
Well, I hope this is all. Sorry for taking the thread
- Joined
- Oct 10, 2014
- Messages
- 21
- Likes
- 11
- Degree
- 0
I will within a couple of days, mate
For now I can say it's both very easy and obvious (when you understand the thinking process behing it), but also a way more time-consuming than running a seed list in Xenu and shifting for DA/PA/TF/CF
Because of the increasing popularity of using Xenu and these procedures -in my experience- this is now one of the only edges you have to find great domains.
- Joined
- Jan 29, 2015
- Messages
- 1,214
- Likes
- 1,139
- Degree
- 4
If you're really interested in the science of network building. I would quit looking to play short run games and focus your entire attention on what you can do to build valuable web assets at scale with out losing them.
The noise level on this strategy has reached a feverish pitch, even by flavor of the month standards.
Spend your time trying to figure out cheap ways to create and seed naturally sharable content that hauls in fresh links. You'll see exponents higher rates of return then playing silly games.
Everyone wants something for nothing, or huge rates of return on negligible effort. The only way you're going to see that is if you can get ahead of the curve. This train hits the station every few years. Wait for it to approach again before you jump on. Its on the way out now. That's just the nature of these things.
The noise level on this strategy has reached a feverish pitch, even by flavor of the month standards.
Spend your time trying to figure out cheap ways to create and seed naturally sharable content that hauls in fresh links. You'll see exponents higher rates of return then playing silly games.
Everyone wants something for nothing, or huge rates of return on negligible effort. The only way you're going to see that is if you can get ahead of the curve. This train hits the station every few years. Wait for it to approach again before you jump on. Its on the way out now. That's just the nature of these things.
- Joined
- Dec 12, 2014
- Messages
- 125
- Likes
- 54
- Degree
- 0
Pbns work and they have a good history, well besides that last google update.
I have never built one because I'm cheap and they cost to much money.
I would rather spend time and effort into building sites that generate income.
You can use the "pbn" cash on articles, grab a few good articles and submit them to other sites. You will get great links from authoritative sites and develop relationships with editors.
Then it's kind of like you have a pbn without the startup or maintenance fees.
I have never built one because I'm cheap and they cost to much money.
I would rather spend time and effort into building sites that generate income.
You can use the "pbn" cash on articles, grab a few good articles and submit them to other sites. You will get great links from authoritative sites and develop relationships with editors.
Then it's kind of like you have a pbn without the startup or maintenance fees.
- Joined
- Feb 27, 2015
- Messages
- 3
- Likes
- 1
- Degree
- 0
As someone with significant software engineering background, I've experimented with 80legs.com as a crawler and find that it works extremely well. Here's how my stack looks (or used to look I guess):
1. Enter seed URLs to customized 80legs crawler
2. 80legs crawler retrieves all links on the seed URL and runs them through "the gauntlet", which is a series of proprietary checks to see if the domain for the backlink is available or not, then caches the information to help speed up future lookups; this runs recursively until the 80legs run limit is reached.
3. All unregistered domains found through this process are then run through a separate script that pulls metrics information: Moz's DA and PA for now but I've considered also grabbing Majestic data since Moz is slipping, as well as a sample of the backlink profile.
4. All the domains as well as their metrics are added to a custom portal that displays everything. There's blacklist filters for the domains as well as backlinks so I can get rid of adult, pharm, gambling, etc. type links pretty much on autopilot, and metric filters so I only see domains with a minimum DA.
5. I then go through the cream of the crop manually through the portal to figure out if it's a domain worth picking up.
I say "used to look" because I am a cofounder of domainjawa.com which is where those domains were sold, but I don't really believe in PBNs / expired domains anymore - content marketing is definitely more effective, less risky, and delivers more value to society when done right.
That said, I do understand the role that a PBN may play in a larger SEO strategy so I'm not completely writing it off, but it does take significant resources and knowledge to pull off successfully. That means content marketing is probably the much better strategy to pursue for someone without those skills or requirements.
1. Enter seed URLs to customized 80legs crawler
2. 80legs crawler retrieves all links on the seed URL and runs them through "the gauntlet", which is a series of proprietary checks to see if the domain for the backlink is available or not, then caches the information to help speed up future lookups; this runs recursively until the 80legs run limit is reached.
3. All unregistered domains found through this process are then run through a separate script that pulls metrics information: Moz's DA and PA for now but I've considered also grabbing Majestic data since Moz is slipping, as well as a sample of the backlink profile.
4. All the domains as well as their metrics are added to a custom portal that displays everything. There's blacklist filters for the domains as well as backlinks so I can get rid of adult, pharm, gambling, etc. type links pretty much on autopilot, and metric filters so I only see domains with a minimum DA.
5. I then go through the cream of the crop manually through the portal to figure out if it's a domain worth picking up.
I say "used to look" because I am a cofounder of domainjawa.com which is where those domains were sold, but I don't really believe in PBNs / expired domains anymore - content marketing is definitely more effective, less risky, and delivers more value to society when done right.
That said, I do understand the role that a PBN may play in a larger SEO strategy so I'm not completely writing it off, but it does take significant resources and knowledge to pull off successfully. That means content marketing is probably the much better strategy to pursue for someone without those skills or requirements.
Golan
BuSo Pro
- Joined
- Nov 13, 2014
- Messages
- 406
- Likes
- 206
- Degree
- 1
Any good progger can build you the scraper. Then you need to pay either for API's or for good proxies.
- Joined
- Feb 6, 2015
- Messages
- 45
- Likes
- 14
- Degree
- 0
@Golan
Tried today scrapping google via scrappy(python).
I got blocked by captchas. I also tried to run my http requests over the TOR network, but it seems to me that most of the endpoints are blocked.
What is your scrapping architecture? Are you using free proxies?
I would appreciate your answer!
Tried today scrapping google via scrappy(python).
I got blocked by captchas. I also tried to run my http requests over the TOR network, but it seems to me that most of the endpoints are blocked.
What is your scrapping architecture? Are you using free proxies?
I would appreciate your answer!
- Joined
- Jan 29, 2015
- Messages
- 1,214
- Likes
- 1,139
- Degree
- 4
Stop beating dead horses.
Stop reinventing the wheel.
Stop trying to play games.
Stop reinventing the wheel.
Stop trying to play games.
Last edited: