5 ms·

I'm pretty confident that no one A/B tests URLs for SEO purposes. How would you A/B test this anyway? You can't exactly serve Google different URLs and see what

by rbinv 11y ago

I'm pretty confident that no one A/B tests URLs for SEO purposes. How would you A/B test this anyway? You can't exactly serve Google different URLs and see what works (in fact, this would ruin both approaches).

jasongill 11y ago
You're definitely wrong about that - there are whole businesses built on building "case studies" that try to decipher the inner workings of Google's algorithm. These days most tests are run by buying two domains that are a combination of random letters/numbers, setting up two nearly identical sites, and doing identical linkbuilding to both - then seeing which one ranks higher. It's not exactly scholastic research quality, but repeated enough times it gives you an idea of what's working "right now".
- tobltobs 11y ago
  This method wouldn't work. To make both results comparable you would have to put the same content on both sites, which would result in one of both pages getting hit by the duplicate content problem.
  - jasongill 11y ago
    It actually works great; the "duplicate content problem" only impacts pages on the same domain, not identical content across multiple sites. It's possible to make a lot of money by taking the content of mild authority sites and putting it on a high authority domain - can outrank the source sites in short order.
jedberg 11y ago
We could have only done the special URLs for 1/2 the content and then see if the inbound requests from Google were more heavily weighted towards the ones with the special URLs. Not a perfect test, but with enough links, it would have been a good indicator.
encoderer 11y ago
In large localized sites you do it by geography. (E.g. Craigslist, Trulia, Auto Trader, etc) You analyze your traffic, find two geographies that are statistical neighbors and that have sufficient traffic. One is your control, the other is your experiment. Run a test with 20 control and experiment locations.
- tobltobs 11y ago
  > find two geographies that are statistical neighbors You would probably have to clone the earth to find that.

by rbinv 11y ago

jasongill 11y ago

You're definitely wrong about that - there are whole businesses built on building "case studies" that try to decipher the inner workings of Google's algorithm. These days most tests are run by buying two domains that are a combination of random letters/numbers, setting up two nearly identical sites, and doing identical linkbuilding to both - then seeing which one ranks higher. It's not exactly scholastic research quality, but repeated enough times it gives you an idea of what's working "right now".

tobltobs 11y ago
This method wouldn't work. To make both results comparable you would have to put the same content on both sites, which would result in one of both pages getting hit by the duplicate content problem.
- jasongill 11y ago
  It actually works great; the "duplicate content problem" only impacts pages on the same domain, not identical content across multiple sites. It's possible to make a lot of money by taking the content of mild authority sites and putting it on a high authority domain - can outrank the source sites in short order.

jedberg 11y ago

We could have only done the special URLs for 1/2 the content and then see if the inbound requests from Google were more heavily weighted towards the ones with the special URLs. Not a perfect test, but with enough links, it would have been a good indicator.

encoderer 11y ago

In large localized sites you do it by geography. (E.g. Craigslist, Trulia, Auto Trader, etc) You analyze your traffic, find two geographies that are statistical neighbors and that have sufficient traffic. One is your control, the other is your experiment. Run a test with 20 control and experiment locations.

tobltobs 11y ago
> find two geographies that are statistical neighbors You would probably have to clone the earth to find that.