For those of you who have been living under a rock, domain authority is a popular concept that claims that a domain name that has lots of pages, gets lots of visitors, has lots of links, and is frequently linked to, is indexed and an authoritative domain and gets special privileges from Google. These privileges supposedly include:
- Your pages getting indexed faster
- Your pages will get a rankings boost
Some of the bigger names in SEO subscribe to the concept of domain authority. Rand Fishken, a big proponent of domain authority, had this to say about it (login required to view):
Search engines DO judge domains. Try putting the same piece of content on Wikipedia and joeshouseofpages.com, point the same links at each, then see which one ranks better. Domain association is a very, very powerful thing.
Unfortunately, Rand and the others who adhere to the myth of domain authority are not looking at the big picture. Here’s what they’re missing:
- Search engines don’t rank domains or websitesHave you ever done a Google search and seen an entire website returned as the search results? Of course not. They return a list of individual pages that most closely match your search query. That’s because search engines rank individual web pages (technically they rank individual documents) and not domains or websites. Each individual web page is judged on its own merit. Other pages on a domain can affect its ranking, primarily through linking, but that’s no different then links from pages on external domains.
- Domain authority has nothing to do with relevanceRanking pages is about one thing: relevance. The goal of every search engine is to return the most relevant search results for any given query. No matter how much trust or authority a page may have if it isn’t relevant to a user’s search term that trust and authority means nothing. Putting crappy content on Wikipedia doesn’t suddenly make it less crappy or even good. It just means it is on Wikipedia.
A real world example would be hiring Johnny Cochran to mow my lawn. Sure, he’s a great lawyer (and dead, too, but let’s look past that for this example) and I trust him to get me out of a murder conviction. But that doesn’t mean he’s going to do a good job of mowing my lawn. Sure, he’s good at practicing law and should come to mind for questions pertaining to it, but that doesn’t mean he’s going to be good at everything he tries to do. Basically, he isn’t an authority on everything.
- Spam is spam regardless of what domain it is on. If trust and authority do exist they serve to assist Google is determining whether content should be included in their index or not. Basically, it is used to help determine if content is spam or not. A page on a domain that has high authority or trust might be less likely to raise a red flag versus a page on a domain that has had lots of content removed from the index for various reasons (DMCA requests, duplicate content, spam). But if that content is problematic in some way, regardless of whether it is on wikipedia.com or spamfest.com, spam is spam and will be dealt with in accordance to Google’s policies.
- There are no arbitrary bonuses in search. Google already has methods for determining relevance (ever hear of PageRank?). They don’t need to give an arbitrary bonus to domains. It’s no coincidence that the same people who claim domain authority is important also support the myth that .edu and .gov domains are more important then other TLDs. They just love the concept of arbitrary bonuses. Using concepts like domain authority and TLD authority and other arbitrary bonuses is exactly how search engines like Altavista worked. Or should I say, they didn’t work?
- Most of the criteria for having domain authority are useless or redundant. Let’s look at what it takes to have domain authority and see why they just don’t add up:
- An authoritative domain has lots of pages. This one is pure nonsense. Quantity of content means nothing. It doesn’t indicate quality of anything. It just means the website has a lot of pages.
- An authoritative domain has lots of links. This is easily debunked with one word: PageRank. Google already has a means for valuing links. Why add another arbitrary one?
- An authoritative domain gets lots of visitors. There is no way for Google, or anyone, to know how many visitors a web page or domain gets. Yes, they can use Google Analytics if they wanted to (but they don’t) but even that would only work if the domain was using it on every page. You can be sure that few, if any, authoritative domains are actually using Google Analytics. (Yes, I know they released Google Trends for websites, but this data is still unverified and as accurate or useful as any other site that offers traffic information for websites. In other words, don’t count on it being accurate. And you need extremely accurate data to use this in a search algorithm).
- An authoritative domain is frequently indexedThis is more of a byproduct of link popularity, thus being redundant, or frequently updated content, and thus not an indicator of domain authority.
So why do websites like Wikipedia do so well in Google searches? Like I told Rand (login required to view):
It’s easy to explain why Wikipedia performs so well vs joeshouseofpages.com: internal linking. Wikipedia does an outstanding job of interlinking its internal pages which only serves to increase each page’s individual strength as we all know internal links are just as important as external links. So Wikipedia will beat the average website because they do a masterful job of promoting their internal pages. Even those obscure little ones will do well thanks to the link juice being spread around by their excellent internal linking structure.
Could domain authority be used as a tie-breaker if all things were absolutely equal? Sure. Why not? But as we all know it never comes down to two pages having everything being equal. There are countless factors affecting a page’s ranking and two pages, except under extremely controlled circumstances, never are even close to each other in ranking factors. So the whole “all things being equal” scenario just never happens.
So, if domain authority exists, remember what it’s real purpose is: spam control. Google already has other means for ranking pages.
I have no idea what you are saying, but it sounds important and knowledgeable.
I agree that internal links are important, but I’d probably stop short of saying they are just as important as external links. Wikipedia does inter-link well, but they have also built up a strong brand that many other people reference. Goodness, even their article on deuterium has 822 in-bound links.
Steph, what John’s saying is that you don’t have to have a trusted domain like cnn.com or wikipedia.com to improve your rankings. Your domain name can be theworstdomainnameever.com and if the content is good, you have a strong internal linking structure (pages within your site linking to each other – and I don’t just mean the menu either), and you have great links pointing to your site from other Web sites, you can do well. (Though that is not a guarantee that it WILL happen.)
@cochlea, internal links are some of the easiest links (with targeted anchor text) you can possibly get. Not only do they count as links as far as the search engines are concerned, but when used properly can also be considered to be helpful to the person using the Web site (since they don’t have to go to the menu just to get to the page in question). So not only do you get an SEO benefit from them, but also an accessibility and usability benefit as well.
Now what about those websites, like Wikipedia, who have what looks like upgraded listings in Google. If you type in Wikipedia you’ll see their title and description and then directly underneath that there is about six links to internal pages. I once read that people who had “authority” got that kind of listing but apparently not. 🙂
Pingback: Meta Tags Are Dead. Get Over It. :: John Conde .net