The concept of Link Diversity is often accepted without much thought. However, nothing should be accepted without critical consideration. This is not to suggest that acquiring the same type of link repeatedly is acceptable. It might be useful to examine the role link diversity plays in site ranking and whether it remains relevant to SEO in 2018.
Reality Check on Link Diversity
One SEO expert known for understanding search engines is Ammon Johns. Therefore, I turned to him for a perspective check.
I shared my doubts about link diversity with Ammon.
Other than identifying link manipulation, the key criteria for links is about relevance, not diversity.
Am I missing something?
Ammon Johns replied:
Not that I’m aware of, and I agree.
The whole diversity thing is based on a misunderstanding of the basic old principles of not having an identifiable pattern or footprint to manipulation.
I think it just got dumbed down over a decade, like the children’s campfire game of telephone.
Let’s face it, even the concept of relevancy is massively misunderstood and misquoted. The bigger the scale of the link network, the harder it becomes to hide.
I always describe the way Google looks at links like the Six Degrees of Kevin Bacon. It’s all about how many links from a cast-iron authority site they know can’t be bought or manipulated.
Like peeling a sticker, once you have a grip on a corner, the whole thing peels away.
To be honest, there are even times when I think the search engines aren’t worried about ownership diversity. When a company like Apple launches a new product or service and promotes it from their existing authority domains, I’m certain that it actually benefits rather than harms them.
Definition of Link Diversity
Link diversity is a strategy of acquiring inbound links from various types of pages (e.g., directories, articles, news). It involves obtaining links from different domains such as .edu, .com, .net, and so on. Lastly, it includes acquiring a mix of anchor text links, such as "do follow" links (which pass link equity like PageRank) and "no follow" links (which do not).
The idea is to obtain diverse inbound links to maintain a site’s backlink patterns within a normal range, preventing them from appearing spammy.
Link Diversity Makes Backlinks Look Natural
In several Facebook groups, people mentioned link diversity as a reason to obtain no follow links, believing a mix of link types makes a backlink profile appear natural. The idea is that natural-looking links help sites avoid detection by statistical analysis algorithms.
What is Statistical Link Analysis?
Statistical analysis was confirmed to be in use in 2005 at Pubcon New Orleans. It studies the characteristics of inbound and outbound links, such as the number of outbound links on a page or inbound links to pages within a site.
When characteristics are graphed, sites with varying percentages typically form a gentle curve, except for spam sites, which cluster outside this curve. These clusters are called outliers because they lie outside of normal ranges.
An important research paper titled "Spam, Damn Spam, and Statistics" illustrates how statistical analysis can spot link patterns. This analysis led to the idea of obtaining a “natural” looking backlink profile.
Because statistical link analysis can identify linking relationship patterns, the SEO industry suggested acquiring diverse inbound links. Instead of exclusively using link directories, they recommended adding links from other sources to avoid standing out from normal link profiles. This may be where the link diversity concept originated.
Is it Important to Look Natural?
The “natural look” concept stems from the 2005 revelation that Google uses statistical analysis to determine if a link profile is statistically normal or different.
Spam sites had different link profiles that were easily spotted. The idea is that manipulated links would be harder to catch if the profile matched average link profiles.
Thus, a high amount of keyword-optimized anchor text would be balanced by a larger amount of brand name anchor text, known as an anchor text ratio. This describes the difference between keyword-optimized and non-optimized anchor text amounts.
Modern Link Analysis
Modern link analysis uses a seed set of trusted sites, calculating distances from these sites to others ranking for specific keywords.
In 2005-era statistical analysis, spammers could hide bad links among non-optimized ones, appearing statistically normal.
With modern analysis, such tactics don’t work. The bad links can’t hide under many regular non-optimized ones. This method of detecting spam, known as Penguin, changes the definition of a "natural" link profile. Link diversity is irrelevant in this definition.
This is why I’ve concluded that link diversity should move to the back of SEO factor considerations, if not be discarded completely.
Google Link Diversity Research and Patents
Search engines prefer diversity mainly to identify inter-site relationships, helping to catch manipulative links.
If one person owns a network of a hundred sites, linking patterns and ownership data could identify relationships, detecting link selling and spamming.
Beyond identifying manipulation, linking criteria focus on removing irrelevant links’ influence, leaving only relevant links in the ranking process—a process of removing noise to identify a signal.
To my knowledge, aside from ownership data, no research uses “link diversity” as a ranking factor. Link analysis mainly removes noise from link data to leave the signal.
It’s safe to say no research or patents use link diversity as any ranking signal.
Should Link Diversity Be Dismissed?
In my view, link diversity can be largely ignored. This doesn’t suggest poor link practices, like excessive directory links, are acceptable.
There are two reasons to dismiss link diversity. Firstly, it’s become an excuse for poor link building. Those advocating low-quality links defend them by citing link diversity.
Secondly, search engines have evolved since 2004. Insisting on diverse inbound links is unnecessary due to modern algorithm operations.
Does it matter if a site has a mix of links from .edu, .org, and .com? Researching top-ranking sites’ backlinks shows this diversity doesn’t exist.
