Testing the Ability to Pass Link Equity - Getting Odd Results

[graywolf]

I'm testing a domains ability to pass link equity using made up 4 KWD phrase. The phrase had zero results when I started the test and it's extremely unlikely to be affected by any user metrics, its four multi-syllable seldom used words that would never be used together. Using this 4 KWD testing phrase I link to an orphan page on a second domain, there are no sitemaps, Google is never going to know the receiving orphan page exists it's URL is made up of random letters and numbers

[recieving-domain.com/a9xstwe1q.html]

The only way to know the page exists is my test link. I have other orphan testing pages on this domain with no inbound links that never show up, so it's not grabbing data from my browser or any plugins or extensions. I've been using this testing method for a while, the sequence of events goes like this:

• The testing phrase ranks on the originating domain [example.com]
  
• A several days later the page on the receiving domain ranks for the phrase
  
• If the originating domain never ranks, I know the site has been banned by Google
  
• If the receiving page never ranks, I know the site has lost its ability to pass link equity
  

I purchased a domain at auction and was testing its ability to pass link equity and got some unexpected and odd results. When you go to Google proper, not the address bar, and type in the domain name [example.com] you get zero results.

When you go to Google and do a site: search [site:example.com] you also get zero results.

When you search for the 4 KWD phrase the receiving domain ranks, but not the originating domain. To me, this indicates the domain can pass link equity, but Google has given the originating domain some sort of penalty that prevents it from ranking. This condition has persisted for a few days, so it's not just some weird data anomaly.

I don't really have a good explanation for why Google would create a condition that would allow this happen.

[edited by: goodroi at 9:21 pm (utc) on Dec 4, 2017]

[phranque]

that sounds like noindex,follow behavior...

[graywolf]

the robots.txt file is empty, and there are no robots meta tags on the page.

[edited by: goodroi at 9:22 pm (utc) on Dec 4, 2017]

[seoskunk]

Hi Graywolf,

did register domains in google search console?

[tedois]

Is it still like that?

[martinibuster]

I purchased a domain at auction and was testing its ability to pass link equity and got some unexpected and odd results. When you go to Google proper, not the address bar, and type in the domain name [example.com] you get zero results.

When you go to Google and do a site: search [site:example.com] you also get zero results.

Those are SEO Results, not search algo results. Different algorithms. Thus, you can't extract meaning from them that reflects on the regular algorithm. Same with intitle, etc. Those are skewed and limited by design.

When you search for the 4 KWD phrase the receiving domain ranks, but not the originating domain.

Nonsense phrases and nonsense words are not processed by the regular algorithm either. I'm fairly certain that in most cases you're mostly looking at basic keyword matching.

This is most visible in small niches where there's a lack of data (lack of sites hosting the phrase, lack of sites receiving the phrase, lack of searchers searching with that phrase). The processes in the algorithm do not work when there is a lack of data. Like trying to hammer a pin. So it fall backs on regular keyword matching. This is true with real-world searches, too.

There is not enough user traffic to help Google's algorithm identify what you mean when you type these phrases. RankBrain is supposed to kick in but it doesn't always.

I can give you real-world examples in private that'll show this in action.

Good luck,
;)

Roger Montti

[graywolf]

The originating domain popped through for [example.com] and [site:example.com] yesterday, still a weird test result.

[edited by: goodroi at 9:22 pm (utc) on Dec 4, 2017]

[aristotle]

It might be interesting to do a similar test using a nofollow tag on the link.

In the past some members of WebMasterWorld have expressed suspicions about google's stated policy regarding the treatment of nofollowed links. We do know that GSC shows some nofollowed backlinks in its reports. The question is whether google's algorithm really does disregard them, as google says.

[martinibuster]

We do know that GSC shows some nofollowed backlinks in its reports.

That's been discussed and settled years ago. GSC shows all links. NoFollow links literally are links, thus they are shown.

In the entire history of SEO for Google, whether showing links in GSC or in the backlink searches, Google has been consistent about showing ALL links, including links that aren't helping a site rank.

So hopefully this catches you up to the current and actual state of SEO. This is very basic SEO knowledge and everyone should know it.

Good luck,
;)

Roger Montti

[aristotle]

Yes it's true that what links GSC shows doesn't prove anything. I didn't say that it does prove anything.

What I did say is that over the years several members of WebMasterWorld have doubted that the matter is settled, despite what google says. Unless google makes its algortithm public, there will always be questions about what data it might or might not use

[martinibuster]

...despite what google says...

The technical specification for nofollow is that it is a directive and that means the search robot will not follow it. There is no choice in the matter. Algorithms literally have nothing to do with it since the directive removes it from consideration.

Whatever else WebmasterWorld members feel this week or the next has no bearing on that technical characteristic about directives.

[aristotle]

There are ways for google's algorithm to use nofollow backlinks even if googlebot doesn't follow them.

-- The algorithm could use the "anchor text" of nofollow backlinks as an indicator of the content of the target page. The possible use of anchor text isn't part of the technical specification, so we have to take google's word on the matter.

-- The algorithm could include nofollow backlinks in its evaluation of the naturalness of a site's backlink profile. For example, if a site has 1000 backlinks, and all of them are dofollow, that would look suspicious. A natural-looking backlink profile would contain a mixture of both dofollow and nofollow backlinks.

[martinibuster]

-- The algorithm could use the "anchor text" of nofollow backlinks as an indicator of the content of the target page.

\

The link does not exist when there is a nofollow. Anchor text are just words on a page. Words on a page are unconnected to any site, except for the page it sits on.

The algorithm could include nofollow backlinks in its evaluation of the naturalness of a site's backlink profile.

1. No such algorithms have been studied, researched or patented.

2. The fact that nofollow drops the link from the link graph means those links do not exist to be evaluated.
"Essentially, using nofollow causes us to drop the target links from our overall graph of the web. "
https://support.google.com/webmasters/answer/96569?hl=en [support.google.com]

3. Current algorithms make that kind of analysis redundant.
Modern link analysis algorithms have made many statistical analyses redundant and not necessary. Here is where a lack of understanding of how modern link analysis presents a road block toward progressing further in this discussion.

Here's are more or less 5,000 words I wrote on the topic, in case you want to catch up.
https://www.searchenginejournal.com/googles-penguin-algorithm-really-research/185261/ [searchenginejournal.com]

[aristotle]

The link does not exist when there is a nofollow

That's the end of my part in this discussion. I've already explained the possibliities, so am not going to waste any more of my time

[seoskunk]

The link does not exist when there is a nofollow

Google is hungry for data, to assume a link does not exist in Google because it's nofollow would discount any user metrics associated with that link.
The Link exists to users
The Link exists in search console
The Link exists in user metrics

It may not pass anchor text or pagerank but the link still exists.

The algorithm could include nofollow backlinks in its evaluation of the naturalness of a site's backlink profile.

Absolutely true

[phranque]

1. No such algorithms have been studied, researched or patented.

2. The fact that nofollow drops the link from the link graph means those links do not exist to be evaluated.

1. if it was important to google's algorithm they would have filed a patent on this by now.
anything filed by google relating to search would have been discussed on seobythesea.com.

I've already explained the possibilities

that's a great academic exercise but have a i missed something in "the literature" (i.e. authoritative study/research or a patent)?

2. as i understand this, google can't analyze nofollow links in a backlink profile if they are missing from google's "overall graph of the web".
Use rel="nofollow" for specific links [support.google.com]:

How does Google handle nofollowed links?

In general, we don't follow them. This means that Google does not transfer PageRank or anchor text across these links. Essentially, using nofollow causes us to drop the target links from our overall graph of the web. However, the target pages may still appear in our index if other sites link to them without using nofollow, or if the URLs are submitted to Google in a Sitemap. Also, it's important to note that other search engines may handle nofollow in slightly different ways.

The algorithm could include nofollow backlinks in its evaluation of the naturalness of a site's backlink profile.

please cite something authoritative.

[seoskunk]

please cite something authoritative.

In general, maybe, er, possibly, sometimes, occasionally .... many people have speculated on how Google truly use nofollow and read between the lines of what is said by Google themselves. You want a citation to cast doubt on Google handling of nofollow... you already provided one