Hacker News Reader

Comment by mcantor | Hacker News Reader

[citation needed]

Seriously though, despite how anecdotal this page is, it gave me some food for thought. Does Google really care about how far in the future I have registered my domains? I'd love a source on that.

View on HN

You can see why this would work by imagining that you are a Google search engineer that's trying to lower the rank of spammy sites so that quality-ranked content appears higher.

* Spammers tend to have tons of crap sites, so they have less money to spend on registering all of their URLs for a long time.

* "Good site owners" tend to have few high quality sites, so they can afford to spend more on registering their domains for longer.

Thus, you would probably want to use the time-until-URL-expiration as a "ranking factor" while sorting results on search pages.

Using the expiration date to determine domain legitimacy is in one of Google's patents: http://www.searchengineguide.com/bill-hartzer/does-the-lengt...

I've updated the post with a link to that article. Thanks for the inquiry.

Great link!

Thanks for the post. It's one of the better summaries of SEO practices around.

A citation would require a source with knowledge of Google's algorithm. Some of the best research that I've found on SEO is SEOmozs Search Engine Ranking Factors: http://www.seomoz.org/article/search-ranking-factors

It's based on a survey of SEO professionals/experts, which leaves a bit to be desired, but I still think the conclusions are on point.

Why is page load time listed on that page? I thought I read recently that's a new thing google is doing?

Google had hinted before this year that they would use speed. This article is from December, but perhaps there's something earlier: http://www.webpronews.com/topnews/2009/11/13/google-page-spe...

Still, Google said it will only affect 1% of searches.

Hey, it's Willy! Thanks for that link, I put it at the bottom of the post. Very interesting!

The problem is that all of these techniques are based on speculation and/or empirical data, but without solid measurements, especially since google's ranking system is a black box. Personally I have encountered 2 opposing opinions on this subject, one suggesting that google does take the duration of the registered domain under consideration, based on some patent that google had filed about it's ranking system, and the other one was from some supposed employee who claimed it didn't, due to the fact that some registrars withhold this information and thus no comparison could be made between them and domains on those that publicized it.

I can't claim to know Google's actual procedures, but if you have a system with many thousands of candidate signals, and you train it to achieve certain 'quality' rankings via a largely automated process, it's possible that even the designers of the system would not know how every signal, in every combination, affects the rankings -- without researching very specific conjectures one at a time.

To wildly hypothesize, what if length-of-registration tends to indicate more-beloved results in most .COM domains, but has no impact on .ORG and .GOV domains, and actually indicates less-beloved results in .COM domains that consist of generic common words (those loved by professional domainers who then lengthen their registrations strategically)? A learning system could discover these conditional relationships over time -- but they'd resist easy explanation by a employee summarizing factors via informal channels.

Google thinks about all sorts of things that have never been related to reality, or have ceased to be accurate because they are being heavily gamed. Then, to adapt to SEO gaming, they start to ignore things that were once or are potentially useful. I loathe the entire system.

No kidding. And remember, if you're doing good in search, they can destroy you overnight. I have a nonprofit wiki where people can share code, and about a month ago my search traffic dropped 95% for no reason. I've never done SEO on this site -- it doesn't even have ads or links for christ's sake. I submitted a reinclusion request which was "processed" - but nothing has changed.

The Google gave, and the Google hath taken away.

Agreed. There is some really shitty content that gets ranked highly on Google. We still need a better search engine.

It is there because Google outlined it in a Patent: http://bit.ly/akczok -- see [0099]

I honestly think that they do not really use this signal though, and if they do, it's so far down in their algorithm that most searches will never use it. I've had domains that are only registered for 1 year that rank just fine.

This is extremely anecdotal... but in my experience, it's all the stuff that actually works. And until Google releases a definitive guide to how their search algorithm works, I think anecdote is all we have to go by.

Google is a registrar and they look for patterns in how domains are registered. We know in their efforts to "fight" spam they certainly have some measurement in place that looks at domains that are registered for 1 year versus domains that are registered for more than 1 year.

I would think there is so much money in "nailing the SEO problem" that there are probably folks out there whose full time job is to simply try things out and figure out what works experimentally.

Reply on news.ycombinator.com