Determining Domain Link & Age Related Trust

SEO Question: What's the best way to determine whether Google has history of a domain - and considers it an old domain?

SEO Answer: Andy Hagans once posted that a site which is getting crawled fairly regularly has at least some trust greater than nothing. The Google Toolbar gives you and outdated exceptionally rough estimate of authority, but beyond just having many pages indexed and cralwed regularly it is a bit more abstract to determine how well a domain is trusted, especially if you do not yet own it and are not able to manipulate its contents to perform testing.

What you can do is back solve for some clues of trust. For example, a domain that does not rank #1 (or near the top of the search results) for the keywords "mydomain" is probably not as well trusted as one that does.

If a domain name ranks for its core unique string then you can see if it ranks well for the keyword phrase "my domain". If it does then you can assume that it would have more trust than one that does not.

Beyond that you can see how well the domain ranks for unique text phrases on its pages or more general keywords. Essentially as you modify your searches (testing shorter or longer word phrase sets and/or wrapping them in quotes or dropping the quotes) you are just testing what chance the domain may have of ranking for various different phrases (and thus its potential ability to rank for other phrases).

Another way to look at a domain is to see how old the domain is in Archive.org. Google crawls the web more efficiently (and likely more aggressively) than smaller search services that are not based on running / being supported by such a large ad network. Thus if a domain has been indexed for a while in Archive.org then it most likely has also been indexed in Google for a while.

Another thing you want to look at in Archive.org is to see if the domain has had a period of inactivity, or if porn webmaster or a pay per click domainer owned the domain for a while. If the domain was inactive for a while, or spent a period of time being abused then it may have had some of its authority stripped at some time.

You can also query Yahoo! for linkdomain: or link: to see what some of the most important backlinks are for a domain. See if those links point at documents that still exist. See if those links point at documents for the same purpose that they originally did. See if the site still serves the same purpose it originally did.

In one of his SEO videos Matt Cutts stated that it was legitimate to redirect a site to a new location so long as the purpose of the site is the same as the original site was. If redirecting a site's authority is legitimate then one could assume that buying or selling a site to use it for its same original purpose is also legitimate as well. If you are going to leverage a site for off topic purposes though you increase the risk that many of the people linking at the site may pull some of their links, and you also increase your risk profile such that a competitor may out you or a search editor may want to remove your site from the search results.

Back to the backlinks...

With the current Google anchor text no longer matters anywhere near as much as it once did, although variation is still a plus.

When looking at the backlinks of a site you can see how long some of those links have been in place by looking through the Archive.org history of the page linking into your site. Links that have developed naturally over time or that have been in place for a long time may carry greater weight than brand new links.

One of the nice features of SEO for Firefox is that it can give you a quick glimpse of the link profile of a page or domain. While some people discount extra weight being placed on .edu and .gov links than other links I would not be so quick to discount that theory. Generally links that are harder to influence are going to be sources that search engines would want to trust more. Either through directly trusting them more on a per link basis, or by creating a version of the web graph which starts near (and places more emphasis on) some of the core trusted authority sites. Since the web started largely in .edu and .gov type environments and those types of pages are often fairly pure in nature and easy to link at it makes sense that their link voting power would be highly represented in the search results.

When valuing a domain you have to look beyond just the link and age related equity it already has built up. How self reinforcing are its key attributes? Would it be easy for someone else to steamroll over your key widget by throwing a bit of ajax on a similar tool? Does a newer competitor have a richer community driven environment that is picking up steam? Are your links next to impossible for others to get? How official or legitimate does the name sound? Will you be able to build it into something that can continue to gain traction and authority? Or is it going to be a site surviving on past popularity until it withers away?

There are lots of things to consider when valuing a domain. Small changes in ad positioning or monetization method can lead to doubling or tripling earnings. And you can also drastically increase the earnings and traffic potential of any site owned by a person who is not savvy to marketing, SEO, or business. Earnings is one important factor, but do not forget to consider the value you can add to a site when trying to determine what you can afford to pay for it.

Published: August 22, 2006 by Aaron Wall in Q & A

Comments

August 25, 2006 - 9:45pm

I like using the way back time machine tool. It can't necessarily tell you if the domain is good or not but you can look at the website that was there before and the HTML code.

August 22, 2006 - 3:36pm

Thanks for your views on TrustRank. TrustRank seems like the magic bullet of SEO - if you can gain trust and authority, that may be one of the most powerful factors influencing your SEO efforts. In my experience, it is also one of the most poorly documented SEO concepts.

I guess that is due to the fact that it is somewhat abstract - there is a shortage of 'official' details on authority and trust from the major SE's, and noone seems quite sure how it works.

Once again, thanks for the great post!

August 22, 2006 - 4:55pm

What about site that are getting crawled regularly but have not been updated in archive.org in quite some time?

August 22, 2006 - 8:40pm

Mike - I believe the purpose of the article was to discern the TrustRank of a site that is not your own, how would you know how often an SE crawled a site if you couldn't review the logs yourself?

August 23, 2006 - 12:46pm

wangarific: for a one-off enquiry, probably not, but if you have a deeper interest in the domain and time on your hands, theres no reason that you couldn't check the google cache regularly to graph the crawl times, or close approximation.

August 23, 2006 - 4:51pm

Aaron ... very good information here. I commented about this post on my blog and added some offline advice as well .... one of them is checking the trademark. But I guess this doesn't really effect the SEO process.

August 23, 2006 - 6:55pm

I went to Archive.org, at your suggestion, to see what it said about my main business website, www.bernsteincrisismanagement.com, which is usually ranked #2 or #3 on Google under my key search terms -- "crisis management" and "crisis public relations." Strangely, it seemed to stop archiving my site -- which is updated heavily at least twice monthly when I post my latest newsletter -- as of Spring 2005. Do you have any idea why?

August 23, 2006 - 7:47pm

wangarific: I was basically saying what Jonathan Bernstein posted.

Add new comment

(If you're a human, don't change the following field)
Your first name.
(If you're a human, don't change the following field)
Your first name.
(If you're a human, don't change the following field)
Your first name.