BAD NEWS: [Life|Business|SEO] Randomness Or Why You Should Buy More [Links|Domains] (Or Just Move to

alexodysglobal

Jr. VIP
Jr. VIP
Joined
Aug 13, 2014
Messages
320
Reaction score
705
Website
odys.global
Full article title below (in case it didn't preview well on the forum):

BAD NEWS: [Life|Business|SEO] Randomness Or Why You Should Buy More [Links|Domains] (Or Just Move to a Bigger City) In Order To [Succeed|Get Laid] More Often

(dedicated to all of you out there who [bought|built] 1 link, 1 domain, 1 SEO package, tried ranking 1 site, just once! and failed - therefore concluded SEO is dead, doesn't work and gave up; for good! (not to mention you also asked for a refund or even charged back the hard working service provider)

This post has 6925 words
Approximate reading time: 36+ min.

(Feel free to bash me in the comments if I have wasted 36+ minutes of your life. This won't happen again in the next 4 years)

---


In this [article|rant], a.k.a [table of contents]:

  1. Life randomness applies to SEO as well (What's in it for Google?).
  2. Why PBN domain testing is [obsolete|marketing hype] and barely proves anything (especially on a small [batch|range] of $250+ [aged|auction] carefully spam-checked domains lol).
  3. Why buying 1/2 [links|domains] to test a service will not yield [unbiassed|fair] results.
  4. Is exposing your site(s) to a larger variety of [links|domains] the only solution?
  5. Is SEO dead for [beginners|people with small budgets]?

---

First of all I would like to confess I'm no math guy and I did poor at math in school.

(So don't fancy me for using the glamorous word 'RANDOMNESS' so many times throughout this post. Please. It's simply the only word that can be realistically used to describe our [tests|observations]).

However, the truth is, the older one gets the more math starts making sense.

Everything around us is math, the universe is based on math and everything we know so far about it can be in a way or another explained mathematically.

(You may also get in trouble calculating the profit margins of your link selling service if you don't know basic math lol)

Nevertheless, that's not the point of this post.

What I'll try to bring to your attention today is some tricky randomness that Google has been fooling us with for quite some time now.

(It's been 6+ years since the very first Penguin in the spring of 2012. Now, a lot happened since then, but increased randomness is one thing we've been observing and would like to talk in depth about today. And randomness requires at least some elementary school understanding of math - nothing fancy though as even I was able to get 'the idea'.)

The point is, Google has been adding more and more variables to the game. And we know that outcomes become harder and harder to predict, almost to a point when it becomes absolutely impossible as we add more variables to the play.

Looking for an example?

Sure.

Here's more than one:

  • Every single SEO niche/SERP is absolutely unique.
  • Every single site on the web is unique.
  • Every single [expiring|expired|scraped|auction|aged] PBN domain is unique.
  • Every single link is unique.
  • Every single link placement is unique.
  • Every single page has different authority and a different link profile powering it.
  • Etc.

What works in certain SERPs isn't a 'fit for all' in many other, or even NO other.

Your Niche A ranking method won't work for Niche B.

Google has been adding more and more variables and generating more and more uniqueness, therefore randomness, of the blue, for us, [spammers|SEOs].

Pretty smart move, eh?

I've been referring to it as RRF (or Random Ranking Factor, the term coined by Terry Kyle couple of years ago) in my previous 2014 thread known as the PBN Q&A. It stirred quite a bit of controversy back in the days, but I haven't been able to find any proof of its non-existance ever since. Four years later - guess it's time for another post!

---

Now, before I begin going over each point in the table of contents, couple of disclaimers:

Disclaimer nr 1:

The title of this post is provocative on purpose and is meant to expose more eyeballs to the randomness in SEO, randomness we've been blaming everything on, reveal some data based on our internal empirical tests and from our observations throughout the years. It doesn't really mean I want you to buy more links or domains (or that you should) in order to rank the site you've been struggling with! We will find real life analogies for some ranking bottlenecks and tests some of you may have dealt with recently and prove it's a matter of both [process|methodology] and [chance|luck].

Disclaimer nr 2:

Links and domain sales of some providers here around the forum (and not only!) [may|may not] skyrocket after more and more people are exposed to this post (so don't share it with your friends and on SEO social media unless you want [link|domain] sellers to [thrive|go out of business]). I'm not to be held responsible for this lol. Everyone please do your own due dillegence analysis. Always be skeptical about someone's else 'wisdom' and don't take it as gospel lol

---

1. Life randomness applies to SEO as well
(What's in it for Google?).

SEO, just like our ordinary [life|business], has a ton of variables that can lead to either success or failure.

Starting the actual process is important (deciding, planning, buying a domain, building a site, marketing it, getting links, writing great content, increasing conversion rates, etc), but some really tiny and random events is what makes your venture a hit or miss.

And the truth is, if you analyze your success stories (in terms of successfully ranked affiliate sites, clients, etc) you can always observe something really unique about every single 'process' that you started at some point in the past. And that random unique thing is what usually counts most.

Let me be more explicit.

You cannot replicate, or explain, with 100% certainty some things in life.

Some events are totally unique, and random.

Like how and when your parents met.

Like the moment you first saw the [woman|man] that you [would|would not] spend the rest of your days with; an event that marked your whole life and transitioned it into a direction you have never thought of.

Like the moment you met your most important, filthy rich client in a night club and somehow ended up driving him home just because he was dead drunk and his driver quit the day before, an event that he later remembered and secured you a contract which brought your net worth up to a million in just a year...

So forth and so on.

These events are unique, random and totally unpredictable.

Sure thing, these are ‘cause and effect’ type of events and we are usually 'very good' at finding explanations for what happened (even though that explanation is totally [obsolete|biased]), like saying I stumbled upon that drunk billionaire at a night club just because I went to the night club in the first place.

Ok. True.

But how about all those other guys and gals that went to the same night club and didn't stumble upon the drunk billionaire just like you did?

So yeah, randomness in its purest and totally unpredictable form.

Or maybe you didn't go to enough night clubs yet?

Oh?

When it comes to SEO, things are no different.

Vilfred Pareto's 80/20 (or whatever asymmetric percentage you want) observation proves, in our case (business, life, etc) MOST success is derived from [rare|small] (and unpredictable I would add!) events. It applies to everything, including SEO.

How many of you accidentally stumbled upon a [link|domain|ranking trick] that played a big role in the further development and authority building of your affiliate site or SEO client campaign for Niche A?

But for some reason you couldn't replicate it further... to Niche B?

(Please note we are talking about the state of SEO as it is right now, in 2018, not based on what one could do spamming Google 24/7 prior to 2012 lol)

One can preach - get the best links, best aged domains, buy $5k worth guest posts on Forbes and HuffingtonPost type of sites and you will rank. You will rank because I did it! (after following a similar strategy and actually ranking).

Here's the blueprint.pdf!

Here's what I did!

Go for it!

But you never consider the [uniqueness|randomness|game of chance] of the 'SEO chemistry variables' that actually led to that result.

Some claim it's possible to replicate success.

And it is, to a certain extent, but nothing is guaranteed, even if you do exactly as the SEO 'guru' next door preaches.

When a method is posted here on the forum, all our human brain needs is reading a few success stories based on that method in order to find sufficient confirmation we [can|should] also do it.

But is that enough?

How about everyone else that took action, followed the exact steps and didn't succeed?

Do their results event matter?

Well, if you want to have a more scientific approach to SEO you should definitely consider them as evidence too. But those folks are silent evidence and rarely someone considers it.

It's just not right to rely on the confirmation bias in order to prove something, SEO included.

So, if we become a little bit more introspective and get the chance to meditate just a tiny bit on the most recent events in our life, events that shaped the present (or the 'future' as it looked like at that particular moment in the past) we notice events linked to randomness and chance, [events|outcomes] generated by a huge volume of variables.

Luck favors the prepared (Louis Pasteur: “Fortune favors the prepared mind.”), for sure.

But don't ignore those that prepared like crazy and failed, saying hey, they didn't work hard enough!

Or maybe they didn't try hard enough?

Or enough times?

Didn't visit enough 'night clubs'?

Didn't meet enough drunk billionares?

Just food for thought.

Once again, SEO is no different.

Those of you who have been doing it for years definitely noticed both a game of patterns (like the fact that one [needs|doesn't need] links to rank in Niche A), but some random, totally unique stuff as well when analyzing the [businesses|websites] that took off and generated really good revenue and those which never did.

So how can we actually replicate success in SEO, playing this game of links and content with so many unique variables?

Is the observation of random results the end of SEO as a business?

Yes - for most.

No - for those who understand or at least try to understand randomness and uniqueness in Google's algo and that there is no such thing as a 'fit for all solution' to your ranking problems.

But, at the end of the day...

... it definitely helps Google fight beginners that lack basic [business|risk] understanding and all those with very little budgets. In other words, by introducing more and more variables to their algo, Google is able to keep a ton of un-committed beginners out of the game.

And that should have helped them deal with the pre-2012 spam problems.

And it did.

Less 'fit for all ranking solutions' - less spam.

Btw, just a reminder...

All those of you who [bought|built] 1 domain, 1 link, gave just one try and failed, gave up, went home.

This post in dedicated to you.

...

If Google manages to stop 50% of wanna-be SEOs from making a career out of spamming their SERPs, they are doing a good job.

And I have no idea what the actual wanna-be SEOs failure percentage is, but I believe it's quite big.

Yeah I know, randomness is tricky.

Really tricky.

But worry not! There is light in the end of the tunnel.

And actually that light is extremely mainstream, but a very few have paid attention.

Will talk more about it throughout this post.

But next, let me try and expose the marketing hype behind domain 'testing' and also show you how unfair or biased your 'link service review' may be.

---

2. Why PBN domain testing is [obsolete|marketing hype] and barely proves anything (especially on a small [batch|range] of $250+ [aged|auction] carefully spam-checked domains lol)

Everyone and their grandparents and children are into PBN domain testing these days.

The PBN testing bandwagon!

"I will not risk linking these domains to my site unless they PASS the test."

OK!

It became very popular since initially the [goal|expectation] was to solve Google's ranking randomness problem by eliminating those domains that are either 'neutral' or 'toxic' as soon as these failed one or two initial linking tests. In other words, you wanted A+ material for your PBNs/links.

For those of you who are new to the domain testing concept, here's a little briefing into the methodology:

  1. You buy/find a good non-spammy PBN domain powered by decent links and metrics;
  2. You [install WP|deploy your site];
  3. You find a niche relevant authority URL naturally ranking on the 2nd or 3rd page of Google, preferably a URL that has no links (but more on the ‘no links’ part later..)
  4. You link to that URL and start monitoring the results for [weeks|months] to follow.

---

EXTREMELY IMPORTANT!
But RIGHT HERE at 4. is where the whole testing becomes flawed from the very beginning, since the results of your 'test' can and will depend on the type of anchors you use: exact, partial, naked, generic, brand, random semantically related words, etc.

Some of you who test PBN domains link to URLs that have no incoming links using an exact match anchor text. Wait, can't that actually hurt rankings? It can. What if you linked using a simple generic 'click here' in contextual proximity to the keyword you want it ranked for? What if you used a partial match? What if you simply linked the naked URL?

SO MANY VARIABLES...

The idea is that your test can be totally flawed from the very beginning if you get just one set of rigid variables like:

  1. Linking to URLs that rank page 2/3 and have no incoming links;
  2. Using exact match anchor text to link to that URL.

This is just one way of testing.

And if the domain fails your initial or secondary test - this doesn't mean it's a bad domain.

What if you did the same test, but with a different set of variables like:

  1. Linking to a URL that ranks naturally page 2/3 and has no incoming links to this date;
  2. Using the naked URL as anchor text for the link.

What if this gets you a positive? Or neutral? Or failed again?

What if you did the same test, but with yet another different set of rigid variables like:

  1. Linking to a URL that ranks naturally page 2/3 and has no incoming links to this date;
  2. Using the 'click here' generic anchor text as the link placed in contextual proximity of the keyword you want that test URL ranked higher for.

What if you get a positive now? What if neutral?

What if you did the same test, with the same 2nd variables as in the first 3 examples above, but for 1., instead of linking to a URL that naturally ranks on 2nd or 3rd page of Google for your selected keyword WITH NO LINKS, you decide to link to a URL that also naturally ranks on the 2nd or 3rd page of Google and that actually has a certain link profile, even though hasn't received any backlinks lately, say 1 year.

It totally messes up the test, adds many more variables into play.

Now the anchors used and the URL's final anchor text ratio % will play a different role, etc...

Again, so many variables which generate so much randomness...

And now, to make everything even more confusing...

Add the idea that every single niche is unique and that even sub-niches in the, let’s say, financial vertical, are different.

Your 'passed' financial niche related PBNs that you tested on general financial related news site URLs may not meet your expectations for other financial sub-niches like 'payday loans' or 'mortgage refinancing'.

Stop reading this for a second and just imagine the PLETHORA of variables involved in a real scientific test and how useless it becomes the moment you plan to link to any other sites except that 1 site in your niche that you've been ranking for months if not years now.

I repeat, taking into consideration the absolutely enormous complexity of variables, testing would make sense only if focusing on one single niche, one single money site, one single environment.

You need to know what you’re testing against, right?

You can then start buying domains and testing them based on the linking patterns and ranking behavior for a particular niche (that you've been working with some time now), keywords, competition etc.

And what works for you may not work for someone else.

Since your [money|client's] site is unique.

Also, a domain that 'passed' for your niche, may not pass when linking to your other niches or to someone else's website.

Do you see where this leads to?

---

Now, let's get back to the mainstream PBN testing methodology.

The aim is to identify and note down, for every 'tested' PBN domain, one of the three following results:

1. Positive/Passed (It moved the needle)
2. Negative/Failed (It hurt rankings)
3. Neutral (Nothing happened)

I have to admit we ourselves initially got very excited about it and immediately started our own tests.

But soon enough realized it simply was too good to be true.

Theory is one thing. Putting it into practice and coming up with actionable conclusions is totally different.

In fact, it's no different than tossing a golden coin and either throwing it away or keeping it based on what you get upon the impact with the ground:

  1. You keep it if: head.
  2. You toss it away to the garbage bin if: tails.

Would you risk that with a $250 worth golden coin?

Now some of you avid PBN domain testers and believers in this method will jump to conclusions and tell me that coin tossing has nothing to do with it, but before you do, let me reveal some of the data from our own tests and observations (also, please consider the 'EXTREMELY IMPORTANT' note above which tries to bring your attention to the multitude of variables which permanently generate even more random results, randomness which makes the idea of testing PBN domains, from the very beginning - flawed).

Now for the sake of the idea, I won't share screenshots of the test results we will reveal in this post simply because everyone does it these days. And most of those fancy screenshots are fake lol

In fact, proof via screenshots has become so unreliable that I prefer to describe it in numbers and words.

Ok, so when tossing a coin, it's impossible to predict the outcome.

If you initially get head, it doesn't mean the next one is tails, or viceversa.

So 50/50 chance.

When it comes to domains, we noticed a slightly similar pattern, even though not as pessimistic and random as with the coin example, but quite close. Close enough so to proclaim PBN testing a waste of your time and money and suggest you will most likely be 'fooled by randomness' (as Nassim Nicholas Taleb says.)

Most people do 1 or 2 tests before they [keep|get rid of] the PBN domain.

Some very few do even 3!

At least so they say (getting rid of the PBN domain... lol)

But are 2/3 tests enough?

As we discovered, no.

There are too many variables in today's Google to make 2/3 linking tests enough to prove a PBN is bad or good for linking. We call it 'Google chemistry' since just like 'Love chemistry' in real life, it happens absolutely random and for unknown and unpredictable reasons.

Imagine the multitude of variables that led to you stumbling into that heart breaking [man|woman] in a city like New York and actually falling in love and engaging into that chemistry. Too many variables to explain what and how it happened.

Same for SEO.

You've all seen random generic domains delivering more link juice and ranking power than powerful niche relevant ones. Not a rule/law, but it happens!

You've all seen money sites built on totally non-relevant expired domains that rank high! Not a rule/law, but it happens!

You've all seen all kinds of weird SEO combinations that Google approves of and some LEGIT ones that Google is not rewarding for some reason.

The answer? Randomness (At least the partial one).

Randomness generated by a 100000000 TB worth of variables.

The whole PBN testing drama started when link sellers realized that by 'testing' PBN domains and publicly stating they are building their networks exclusively using domains that 'passed the test' they can get more sales; and clients; and maybe finally afford that Lambo.

Testing supposed to be a kind of ‘insurance’ against failure, against bad results.

And if you're a link buyer, the offer sounds too good to miss.

Think about it, almost nothing to lose.

The guy is using PBN domains that are PROVEN TO MOVE THE NEEDLE!

"I have nothing to lose. Let me buy now. Please take my money – thanks!"

Yet, how many of you have bought links from sellers that 'test' their PBN domains (or claim doing so) and have gotten neutral or even toxic results at times?

If you have bought links in the last year and half, there is quite a high probability it happened to you as well.

And how could it?

I mean, these domains 'PASSED' one of the initial 2 tests, didn't they?

Well, they might have.

But let's try to dig a little deeper here.

First of all, 2 tests aren't enough.

That's why we based our testing methodology on 5 tests for each domain (initially we wanted to do 10 link tests per domain, but realized it would simply take too long and we may end up publishing this post in 2019 if not later lol).

Why 5 tests?

Well, everyone has been doing 2, saying if 1 passed the domain is ok, if 1 [failed|neutral], but 2nd passed, the domain is still ok; and then we thought, wait - we must do more. What if 1st passed, 2nd passed, but 3rd failed? What if 1st failed, 2nd failed and 3rd passed? What if 1, 2, 3 passed, 4th failed - so forth and so on. Does this mean the domain is good or bad? Confusing, isn't it?

Here are the results we got from testing an initial small batch of 5 aged domains:

Our methodology involved using aged, never dropped, action domains in the entertainment/movie niche:
  • The average REFD of each domain is: 100
  • The average TF of each domain is: 20
  • The average DA of each domain is: 22

We did not wait before placing the link. We hosted, deployed WP, posted the article and linked to our target URLs in the very first day of the test for all our domains. So we basically posted 5 articles each containing one link on every domain. 5x5 = 25 articles day 1. We then fired up the targeted keywords in our rank tracker and started monitoring the results.

Would the results have been different if we waited a month before linking? (+1 more variable to randomness that makes things even harder to test lol)

The first results started showing up after the first two weeks, but we let the test just do its thing for 66 days.

Test start date: 21th of May 2018
Test end date: 25th of July 2018

And to add more variables in the game, note that results could have been different if we would have let the test run for another week or so. Rankings aren’t 100% stable, are they?

So +1 more variable to the randomness here.

Yeah, crazy I know... lol

Next...

We've been linking to:
  • Internal URLs of big authority Film/Movie sites (so niche relevant URLs)
  • URLs naturally ranking on Google's page 2 or 3.
  • URLs that had no incoming links at all.

When linking, we've been using the generic 'click here' anchor text in contextual proximity to the main keyword we wanted that individual page ranked higher for. Example: We would do something like: "To see the full list of main characters in Logan 2017 movie, click here", the targeted and monitored keyword being the very low competition 'logan 2017 main characters'.

Please note the above is just an example. We will not reveal the actual domains, keywords or URLs linked to.

With that being said, please take your time to observe the results we got at the end of those 66 days!

1st PBN
- 1st test: passed!
- 2nd test: passed!
- 3rd test: failed!
- 4th test: passed!
- 5th test: neutral.

(60% passed, 20% failed, 20% neutral)

2nd PBN
- 1st test: passed!
- 2nd test: failed!
- 3rd test: failed!
- 4th test: neutral
- 5th test: neutral

(20% passed, 40% failed, 40% neutral)​

3rd PBN
- 1st test: failed!
- 2nd test: neutral
- 3rd test: neutral
- 4th test: passed!
- 5th test: passed!

(40% passed, 20% failed, 40% neutral)

4th PBN
- 1st test: neutral
- 2nd test: neutral
- 3rd test: passed!
- 4th test: passed!
- 5th test: passed!

(60% passed, 0% failed, 40% neutral)

5th PBN
- 1st test: failed!
- 2nd test: neutral
- 3rd test: neutral
- 4th test: neutral
- 5th test: neutral

(20% passed, 20% failed, 80% neutral)​

Observations:
  • If we would have stopped at 1 link test only, as 99% of PBN testers do, 2 domains would pass the test (the 1st and the 2nd: 2 out of 5 - 40%);

  • If we would have stopped at 2 link tests only, as the majority of PBN testers do, 2 domains would pass the test (the 1st and the 2nd: 2 out of 5 - 40%);

  • If we would have stopped at 3 link tests only, as SOME PBN testers occasionally do, 3 domains would pass the test (the 1st, the 2nd and the 4th: 3 out of 5 - 60%);

  • If we would have stopped at 4 link tests only, as no one we know does, 4 domains would pass the test (the 1st, the 2nd, the 3rd and the 4th: 4 out of 5 - 80%);

  • Doing 5 link tests in a row for all 5 PBN domains helped us identify the 'PASSED' message for 4 domains as well. So 4 out of 5 again, or 80% passed.

So if we would base the methodology on this 5 link tests/domain above, only the 5th domain should be tossed away.

We can use the other 4 to build our network and link our to our money sites or clients knowing that they will get at least some good results since we had quite a few of those ‘passed’ messages. Right? Well, apparently so, at least this is how most PBN testers find it reasonable to ‘consider as logic conclusion'.

But randomness and logic don’t always go along, therefore..

Think about the [link seller|PBN builder] that uses only domains that 'passed' after 1st initial and only one test. He probably thinks he struck gold, knowing that his clients will rank 100% guaranteed or their money back!!!

After all he used domains that passed the initial test!

But, once again, is that enough?

Here are some more observations:

  1. First test showed either passed, failed or neutral messages and the historical results did not guarantee a similar future one (Ex: if 1st passed, then 2nd or 3rd passed as well).

  2. The more domains we tested, the more passed messages we got.

  3. The more tests we performed, the more passed messages we got (from 40% passed from 1 link test for each domain to 80% passed after 5 link tests for each domain!)

Here is some food for thought:

  1. What if we performed a 6th link test for each domain?

  2. What if we performed up to 10 link tests for each domain? Would that have brought a potential 'passed' message for the 5th domain as well, at least once?

It might have.

At this point, some of you may look at the percentage of passed messages for each domain and say:

Alex, why risk it?

Just grab the 1st and the 4th domain and use them in your network.

You've got a 60% passed result on these. 3 link tests out of 5 passed. There must be something to it.

Ok, good point. But what if we performed more tests, say 7 for each, and the 6th and 7th message would be failed or neutral for these two domains that initially got 60% passed based on the 5 link tests?

Wouldn’t that lower the 60% 'passed' rate?

Do these now still qualify as 'passed' or not?

So, as you can see, one can take testing to a whole new level by adding more and more variables in. The more variables, the higher the chance we will get fooled by randomness.

Yes, fooled by randomness.

And taking into consideration that you don't link to just one or two money sites or clients from each PBN (unless you do), the historic success/failure or neutral results are nothing more than randomness, random results, and cannot predict future ones.

The variables are insane.

  • First of all, every single PBN domain is unique.

  • Every single link profile of a PBN domain is unique.

  • In this case, we've used aged entertainment/movie domains to do the test and linked to entertainment/movie URLs on high authority sites.

  • Nevertheless, each PBN domain is unique and if you take into consideration each one of them had an average of 100 referring domains, all unique, that adds to the randomness.

  • And if you take into consideration that each of those authority sites is unique in its own way, that adds to the randomness even more.

  • Now if you take the domains that passed the tests and link to your entertainment/movie client who has a totally different website in comparison with the URLs linked to in the initial tests, that adds even more to the randomness and to the unique variables.

So at the end of the day, our main observation is that Google rewards links based on some 'chemistry' between the source of the link and the target URL, chemistry which occurs as it finds appropriate, and more often than not, totally random.

And it's unique to every niche, every SERP, every linked to URL, every type of link, every link placement, every anchor text decision you make.

Please note historic results cannot be used as guarantee of future ones.

Otherwise 100% of the clients buying links from sellers that use 100% tested domains which passed the initial 1st or 2nd link tests should have 100% positive results.

But it never happens.

There is no such thing as a link seller with 100% success rate for every link pointed to a URL, even if dealing strictly with those that diligently 'test' their domains.

So..

  1. I'm very sorry to disappoint you, but buying 1 or 2 links from a seller that uses just 'tested' domains to build his PBN network and having high expectations just based on historic records is not reasonable and will most likely make you feel disappointed (unless you get lucky!!)

  2. What if you're a link seller building your network using $200/250+ premium, aged, auction, never dropped, never spammed, sparkling clean, extra virgin domains and like a good Samaritan that you are take the pledge and begin testing the domains before going public (you want happy clients and boner rankings screenshots! - don't you?), only to discover mixed results just like the ones in our 5 PBN domains test above.

Would you throw away those juicy $200/250+ premium domains just because they failed an initial or secondary so called 'test' and there was really nothing else that would make you suspicious about those domains except the results you got from the 'test'?

Food for thought.

So it gets tricky if you're a link seller/PBN builder.

And it's also tricky if you're a BHW forum lurker seeking free 'review copies', trying to get free links to your sites while posting 100000% honest reviews of your results with any partcular service.

Yet, you never take into consideration the randomness we just described above.

  • Your good results may be just that 'Google chemistry' therefore, pure good luck.

  • Your bad results may be just that 'Google chemistry' therefore, pure bad luck.

  • The fact that you initially got BAD or neutral results with 1/2 links from 1 seller doesn't mean they have bad domains or don't know what they are doing.

  • The fact that you initially got GOOD results with 1/2 links from the same seller doesn't mean all future results will be good as well and also doesn't mean they have any idea what they are doing.

Unless bad results come from really bad and poorly setup domains.

Unless you get a good SERP bump from a spammy network.

Which DOES happen!

Would you consider using discountUGGboots2013.org in your PBN network it if passed the initial 'test'?

LOL

--

3. Why buying 1/2 [links|domains] to test a service will not yield [unbiassed|fair] results.

Buying 1 link or 1 domain, testing it 'negatively' (getting the 'failed' initial link test message) and wondering why it didn't move the needle is like wondering why the first girl you dated in college didn't eventually become your wife, event which would have led, in your opinion, to the successful perfect family, etc.

or

Buying 1 link or 1 domain, testing it 'successfully' (getting the 'passed' initial link test message) is like marrying successfully the first girl you dated in college, enjoying a successful perfect family, and thinking that everyone else who are having trouble finding A+ material for their marriage are suckers.

When in reality you just got fooled by randomness...

In other words, [dumb|lack off] luck can play a big role both in life and SEO.

I've personally helped a bunch link sellers build excellent PBN networks only to laugh at all those negative reviews they get from people buying just a handful of links in order to 'test the waters' and not getting the expected results.

Google may not have approved of the 'chemistry' between a particular domain and your money site.

Google may not have approved of the 'chemistry' of the anchor text you decided SHOULD be used to link to your money site.

Google may not have approved of the 'chemistry' of the link per se, since your site requires some additional on-page work before it would, theoretically start receiving links/votes of confidence from other sites on the web.

And viceversa.

Google may have approved of the 'chemistry' of the link you got from the same network and you now consider yourself smart, know-it-all, you think you know what you're doing...

Again, getting fooled by randomness.

(Consider the plethora of variables leading to totally uncertain outcome and randomness in the PBN testing example which we talked about in the second chapter of this post.)

So what's the solution?

Is there any solution at all?

---

4. Is exposing your site(s) to a larger variety of [links|domains] the only solution?

Let me ask you this.

Where do you stand a higher chance of getting laid more often:

In 'New York City' or a small town like 'Hyder, Alaska' (a town with less than 100 residents where everyone knows each other basically)

??

Now, if you do understand the concept of being exposed to a higher probability of something, you should agree one stands a higher chance of getting laid more often in New York City, obviously.

A city with 8.5 million residents will offer you many more opportunities of meeting strangers that want the same thing, totally random, than a small town with less than 100 residents in the middle of nowhere, Alaska.

In other words, the more people you are exposed to, the higher the chance of fulfilling your appetite.

Same for SEO.

If you buy just a handful of domains or links and wonder why you got stuck with your SEO campaign, guess what? You need to move to a 'bigger city'!

  • You need to buy/try more links.
  • You need to buy/try more domains.
  • You need to get exposed to more [attempts|failures] at what you do.
  • You need to fail and try again.

Now I'm not going to go into the motivational type of gospel here, but you get the idea.

The larger variety of [links|domains], the higher the chance you will get a good combination.

While most people do get fooled by randomness unconsciously and think they [are invincible at SEO|just suck at SEO], we can actually allow the enormous volume of Google variables to fool us, consciously, positively, without trying to control things that are beyond our control (like PBN testing in an environment with endless variables).

And we can do this, in SEO, by picking one site, one niche, one environment, one vertical and working on it till you start seeing patterns.

Test more links, more link providers (don't buy just 1/2 links from each one to 'test the waters' - this will get you totally biased results), buy more domains, increase the nr of domains in your network.

I've spotted the following phenomena:

People with larger budgets are those who get better results.

In other words, rich get richer and poor get poorer.

But there is a certain logic behind this, even in a world with so many variables generating randomness at every corner.

When you have funds to test more links, more domains, more linking angles on your site, you actually stand a higher chance of understanding what's going on and what works for your particular niche.

Seller’s A PBNs' may suck for Joe, but will deliver awesome results for Bob since Bob was persistent enough to try not just 1/2 links like Joe, but actually bought 20-30 links.

And we can all agree that 20 vs 1 means 20 x higher chance of finding something that works for you.

According to New York Statistics and Demographics (US Census 2000), New York is proud to host 3.794.204 males (47.38%) and 4.214.074 females (52.62%).

(didn't mean to offend anyone that doesn't consider themselves either male or female, but this post aims to introduce you to randomness in SEO, not any other debate - so my apologies.)

As of the census of 2000, there were 97 people, 47 households, and 25 families residing in Hyder, Alaska.

Where do you stand a higher chance of getting laid after all?

Don't you think moving to a bigger city offers a higher chance of solving your loneliness problem?

Same for SEO.

Don't you think exposing your site to more links, more domains, larger variety of attempts may increase the probability of actually understanding what works for you, letting positive randomness play a role and actually getting
better results?

Do you understand now that buying 1/2 links or domains and saying the service sucks or the seller has no clue what he or she is doing is just a biased result and not an objective conclusion?

Maybe you should move to a bigger city after all.

More mating partners to accidentally stumble upon into.

More night clubs with drunk billionaires without a driver.

Etc...

---

5. Is SEO dead for [beginners|people with small budgets]?

Well, it's not just about SEO.

There's no much money or fat profit margins in businesses with low entry points.

If you could spend $100 on a domain, hosting account, several articles, couple of links and tranform this creation of yours into a passive income asset that would generate 3k/mo, everyone would be doing it.

Low entry point businesses fail in 90%+ of the cases.

Remember the 'make money blogging' craze? lol

SEO is no different.

Just like in any other business, you need to understand risk, you need to test the waters, you need to observe what works and who kind of activities do people that are able to replicate what works engage in, so forth and so on.

Those of you who expect to succeed buying 1/2 [links|domains] and actually get some traction got lucky.

Those of you who expect to succeed buying 1/2 [links|domains] and fail just got un-lucky.

In both cases, you got fooled by [Google|Life]'s randomness.

Don't have money to buy 100 links, build a 100-150 PBN network for your [e-commerce|affiliate] site?

How about you find a way to make that money first and then get back to SEO?

How about you find a way to finance your venture first?

Remember, low entry businesses fail at a very high rate, unless you get hit by dumb luck.

So yeah, SEO may be dead for beginners and new folks with just $100 USD in their wallet.

It is what it is.

But the good news is that you can make a commitment, today, and change that.

Think bigger than you usually do.

See what needs to be changed.

Put your skills to work, make some money, come back and buy those 100 PBN links or 100 aged domains.

Get exposed to higher chance.

Get more exposure to luck.

After all, luck favors the prepared.

Be smart.

'Move to a bigger city.'

And last but not least...

Try to understand the basics of randomness.

Or else you risk getting fooled by it.

Over and over again.

Until the next post in 2022,
Alex
 

ScribScribScrib

Elite Member
Joined
Mar 15, 2017
Messages
1,864
Reaction score
2,125
Website
www.blackhatworld.com
TLDR: To combat randomness, we perform five tests on our domains and use these results to filter out unwanted outcomes (such as a domain ostentatiously doing well in the testing phase and later on failing to impress on real sites) and better predict SERP movements after pointing domains at a site.

Also something about abundance being a commodity not many exploit.
 

BassTrackerBoats

Floor Mod, Super Mod, Exec Vip, Dude that helps.
Staff member
Moderator
Jr. VIP
Joined
Mar 10, 2010
Messages
32,122
Reaction score
71,487
Website
www.blackhatworld.com
TLDR: To combat randomness, we perform five tests on our domains and use these results to filter out unwanted outcomes (such as a domain ostentatiously doing well in the testing phase and later on failing to impress on real sites) and better predict SERP movements after pointing domains at a site.

Also something about abundance being a commodity not many exploit.

I'll have to see about a creating a rule and a warning for someone crapping on a thread with a dime store tldr.
 
Joined
Sep 24, 2014
Messages
48
Reaction score
53
Now I'm wondering: why not do the testing real-time? Is all the testing up front really necessary before dropping a link? Or maybe I'm not following this...

I guess I don't really understand the need to test like this in an environment that is unpredictable, as you point out.

To me it seems like starting off with a basic link package full of easy links is a good way to determine what is actually needed to improve rankings for a site. If basic links don't work then the next step would be to improve and add more content to the money site, before moving on to higher quality links.

If it takes a variety of links and everything is unpredictable then why bother testing like this? I'm probably missing the point here, so if anyone can explain that'd be great.

Either way great thought-provoking writeup and I think everyone can see how serious you are about PBN's.
 

PropertyGrow

Jr. VIP
Jr. VIP
Joined
Oct 4, 2015
Messages
651
Reaction score
194
Website
Contentful.One
This post is golden.

I think we should also build more money sites and more links to them. Gotta start multiple sites per niche and concentrate on the winners.
 

tammyblogger

Junior Member
Joined
Aug 6, 2013
Messages
153
Reaction score
67
Thanks so much for sharing this post and your insight and experience with this. That must have taken you ages to write up. What you say makes a lot of sense and I have felt this way for a long time about Google and the random factor. I am going to read it again later on and look forward to your next post in 2022 as someone already said will it be PBN DEAD LOL! Thanks once again for spending so much time writing all of that
 

hero76

Jr. VIP
Jr. VIP
Joined
Oct 20, 2016
Messages
1,525
Reaction score
447
Great artice OP, this took you a month to write or a few seconds of ctrl c, ctrl v :)
 

lightningblitz

Elite Member
Joined
Nov 30, 2014
Messages
1,923
Reaction score
4,676
Great post!

Touches on a couple things I see a lot. The obvious randomness (which makes SEO guides only worth so much - even mine) and the random link value.

When building links I aim for links that will not hurt my ranking (obvious junk, poorly spun text, etc) spring to mind. But all link building for me is now the Doctors Creed.

"First, Do No Harm."

Reading between the lines of your post can definately point a lot of people in the right direction.

And spot on that you need to invest to make it in any business. The trick is, like every other industry, you need both the expertise and funding.

In IM, you can scrape by with enough time, expertise and hard work. That is the appeal.

But unless you win the IM lotto, you will not make it without expertise. That means you need to learn first, or pay someone who knows.

But that is not always profitable.
 

alexodysglobal

Jr. VIP
Jr. VIP
Joined
Aug 13, 2014
Messages
320
Reaction score
705
Website
odys.global
Thanks for sharing.

See I was correct on Skype when I said it was good to see you back on here :)

@MisterF

It's you and a few others that silently motivated me to actually run this test and publish the results on BHW.

Thanks!


Great read! Thanks for sharing

@bc21sek

Hope it makes sense to you lol

Had multiple folks already telling me those who do REALLY need to read it - won't.

Oh well.


That’s a great post by a smart man. Very well written and I hope it will educate many people.

@Diplomat

More and more people are worshipping PBN testing these days without even understanding what they're doing.

I got tired of privately explaining what a waste of time and resources it actually is and decided to go live with the post.

Let's see if we can provide another angle for open minded SEOs.


Awesome read, definitely worth 30 mins to read...

@EuforijaKg

My gut feeling tells me 80-90% of the viewers won't read it, even though they are the ones who need to read it most

lol


TLDR: To combat randomness, we perform five tests on our domains and use these results to filter out unwanted outcomes (such as a domain ostentatiously doing well in the testing phase and later on failing to impress on real sites) and better predict SERP movements after pointing domains at a site.

Also something about abundance being a commodity not many exploit.

@ScribScribScrib

The whole idea of the post was the you cannot predict movements based on initial (historic) results.

If you think you can predict in an environment with so many variables, you are most likely fooled by randomness.

https://en.wikipedia.org/wiki/Fooled_by_Randomness


I'll have to see about a creating a rule and a warning for someone crapping on a thread with a dime store tldr.

@BassTrackerBoats

Guess C.O.A.T could be a new kind of internet sport :)


Now I'm wondering: why not do the testing real-time? Is all the testing up front really necessary before dropping a link? Or maybe I'm not following this...

I guess I don't really understand the need to test like this in an environment that is unpredictable, as you point out.

To me it seems like starting off with a basic link package full of easy links is a good way to determine what is actually needed to improve rankings for a site. If basic links don't work then the next step would be to improve and add more content to the money site, before moving on to higher quality links.

If it takes a variety of links and everything is unpredictable then why bother testing like this? I'm probably missing the point here, so if anyone can explain that'd be great.

Either way great thought-provoking writeup and I think everyone can see how serious you are about PBN's.

@Yarharharharhar

Exactly, testing and seeking to understand what works in real time would be one of the most reasonable approaches to SEO (not saying it's the only one!), one that actually makes sense, taking into consideration that past results don't help us predict future ones.

You are NOT missing the point.

You actually got it!


Thanks Alex for sharing your experiences with PBNs. 2022 ?? I imagine your post then will be PBN DEAD

@kanamon

When considering PBNs, think beyond scraped dropped leftover domains.

Ranking may still require links even in 2022, unless Google totally devalues their importance.

Who knows - but if they still keep linking as a good signal, it's still websites that you're going to get links from - either someone else's or your own :)


This post is golden.

I think we should also build more money sites and more links to them. Gotta start multiple sites per niche and concentrate on the winners.

@PropertyGrow

One of the most successfull people online I personally know focus on 1 or 2 niches/sites/businesses max and run all their tests, live, against those.


Thanks so much for sharing this post and your insight and experience with this. That must have taken you ages to write up. What you say makes a lot of sense and I have felt this way for a long time about Google and the random factor. I am going to read it again later on and look forward to your next post in 2022 as someone already said will it be PBN DEAD LOL! Thanks once again for spending so much time writing all of that

@tammyblogger

It took us 66 days to run the test and around a week to give birth to this post


Great artice OP, this took you a month to write or a few seconds of ctrl c, ctrl v :)

@hero76

Just answered your question in the reply above :)


Great post!

Touches on a couple things I see a lot. The obvious randomness (which makes SEO guides only worth so much - even mine) and the random link value.

When building links I aim for links that will not hurt my ranking (obvious junk, poorly spun text, etc) spring to mind. But all link building for me is now the Doctors Creed.

"First, Do No Harm."

Reading between the lines of your post can definately point a lot of people in the right direction.

And spot on that you need to invest to make it in any business. The trick is, like every other industry, you need both the expertise and funding.

In IM, you can scrape by with enough time, expertise and hard work. That is the appeal.

But unless you win the IM lotto, you will not make it without expertise. That means you need to learn first, or pay someone who knows.

But that is not always profitable.

@lightningblitz

Had a short friendly debate with a friend of mine on the topic of PBN/domain testing last night.

He asked me how can I know a bad domain (hacked, spammy, abused in the past, penalized) is bad and won't hurt my rankings, unless I test it.

My answer was: Why get it in the first place?

There are around 15 different spam-checking angles that one can perform prior to buying, backordering or bidding on a domain.

So make sure you're getting the best domains and stop wasting time on tests that fool us around.

We cannot predict the outcome based on previous results.

In an environment with so many variables, those who accidentally get good results from a domain, network, link, etc - won the lotto.

But the main point is that by being fooled by initial success/failure results, some SEOs risk tossing out or using some domains that they wouldn't have in a different scenario.

Alex
 

alexodysglobal

Jr. VIP
Jr. VIP
Joined
Aug 13, 2014
Messages
320
Reaction score
705
Website
odys.global
Very useful post

@John4life2017

Really hope it was indeed useful.

After all, I've asked more questions than offered answers.

As we're constantly dealing with incomplete information when it comes to SEO.

We just assume what things might be and then as time moves on and we start getting results, we compare our assumptions to the results and humbly accept them.

Cheers
 

antihero11

Regular Member
Joined
Mar 25, 2014
Messages
357
Reaction score
104
Hi again Alex
Thank you for your help. I've read everything carefully.
Just one question
If you increase your links from 1 to 20 for example. If we take into consideration randomness you also have a better chance of finding toxic domains. Isn't it?
 

alexodysglobal

Jr. VIP
Jr. VIP
Joined
Aug 13, 2014
Messages
320
Reaction score
705
Website
odys.global
Hi again Alex
Thank you for your help. I've read everything carefully.
Just one question
If you increase your links from 1 to 20 for example. If we take into consideration randomness you also have a better chance of finding toxic domains. Isn't it?

@antihero11

Toxicity generated from a super quality, legit, aged, niche relevant, 100% clean PBN domain is nothing more than the failed chemistry between the link and your target site in Google's assessment.

It doesn't mean the domain is bad or 'toxic'. Google just doesn't like the combination.

It could easily provide non-toxic/positive results in a different environment.

So unless one buys bad, abused, scraped, weak, ex-PBNs, hacked domains, on purpose, to test God knows what, the whole PBN testing becomes just a waste of time and resources.

Is your goal becoming better at finding 'toxic domains' or making money?


Amazing article OP, just goes to explain everything on the randomness principle, this is just a gem.

@BuildMoreLinks

Thanks for dropping by mate!

Feel free to point your clients to this article whenever they wonder if more links are required for their SEO campaigns LOL

Alex
 
Top