Google’s John Mueller responded to a question about how Google treats outbound links from a site that result in a link penalty. His response suggests the situation might not work out the way many think.
An SEO asked on Bluesky if a site with what he described as a “link penalty” could affect the value of outbound links. The question is somewhat vague because a link penalty can mean different things.
- Has the site purchased or created low-quality inbound links?
- Did the site sell links?
- Was the site involved in some sort of link building program?
Despite the vagueness of the question, it hides a legitimate concern, namely whether getting links from a site that has lost its ranking could also transmit harmful signals to other sites.
They request:
“Hey @johnmu.com, hypothetically speaking. If a site has a link penalty, are outbound links from that site devalued? Or do they have the ability to transmit bad signals… aka bad neighbors?”
There are a number of link-related algorithms that I have written about in the past. And as often happens in SEO, other SEOs will take what I wrote and paraphrase it without mentioning my article. Then someone else will paraphrase that and after a few generations, strange ideas circulate.
Bad Signals AKA Link Cooties
If you really want to dig deeper into link-related algorithms, I wrote a long and complete article titled What is Google’s Penguin Algorithm. Most of the research papers discussed in this article have never been written by anyone until I wrote about them. I strongly encourage you to read this article, but only if you are willing to commit to delving deeper into the topic.
Another involves an algorithm that starts with a set of trusted sites, then the further away a site is from that starting set, the more likely that site is to be spam. It’s approximately distance ranking linkranking links. No one had ever written about this link distance ranking patent until I first wrote about it. Over the years, other SEOs have written about this after reading my article, and although they don’t link to my article, they are mostly paraphrasing what I wrote. You know how I can tell these SEOs copied my article? They use the term “link distance ranking”, a term I made up. Yeah! This phrase does not exist in the patent. I made it up, lol.
The other fundamental article I wrote concerns Google Link Graph and how it plays into web page rankings. Everything I write is easy to understand and is based on research papers and patents which I link to so you can read them yourself.
The idea behind research papers and patents is that there are ways to use link relationships between sites to identify what a site is about, but also whether it is in a spammy neighborhood, meaning low-quality content and/or manipulated links.
The articles on Link Graphs and Link Distance Ranking Algorithms are those that relate to the question asked about outbound links transmitting a negative signal. The problem is that these algorithms are not intended to transmit a negative signal. They are based on the intuition that good sites link to other good sites, and that spammy sites tend to link to other spammy sites. No outbound link cootie is passed from one site to another.
So what probably happened was one SEO copied my article and then added something to it, and fifty others did the same thing, and the big takeaway ended up being about outbound link cooties. And that’s how we got to this point where someone asks Mueller if the sites are transmitting “bad signals” (link cooties) to the sites they link to.
Google may ignore links from problematic sites
Google’s John Mueller was apparently confused about the issue, but he confirmed that Google simply ignores low-quality links. In other words, there are no “cootie links” passed from one site to another.
Mueller responded:
“I’m not sure what you mean by ‘has a link penalty,’ but in general, if our systems recognize that a site is linking in a way that isn’t very useful or consistent with our policies, we may end up ignoring all links coming from that site. For some sites, it’s just not worth the effort to research link value.”
Mueller’s response suggests that Google does not necessarily consider links from problematic sites to be harmful, but may instead choose to ignore them altogether. This means that instead of transmitting value or negative signals, these links can simply be excluded.
This does not mean that links are not used to identify spammy sites. This simply means that spam is not something that is passed from one site to another.
Ignoring links is not the same as transmitting negative signals
The distinction between ignoring connections is important because it separates two different ideas that are easily confused.
- The first is that a link may lose value or be discounted.
- The other is that a link may actively transmit negative signals.
Mueller’s explanation aligns with the idea that Google simply ignores low-quality links. In this case, the links do not contribute positively, but they also do not broadcast a negative signal to other sites. They are simply ignored.
And that sort of fits into the idea of something else that I was the first to write about, the Reduced link graph. A link graph is essentially a map of the web created from all the link relationships from one page to another. If you remove all ignored links from this link graph, all spammy sites will be removed. This is the collapsed link graph.
Mueller cited two interesting factors for ignoring the links: helpfulness and not being aligned with their policies. This part of the utility is interesting, also quite vague, but it makes sense.
Points to remember:
- Links from low-quality problematic sites can be ignored
- Links do not transmit “bad signals”
- Spreading Negative Signals Probably Not a Thing
- Google’s systems appear to prioritize usefulness and policy alignment when evaluating links
- If you write an article based on one of mine, link to it. :)\
Featured image by Shutterstock/minifilm
