4 ms·

To me your comment demonstrates why it's really hard to figure out anything worthwhile about SEO. You wrote that code almost 8 years ago[1]. Google and other s

by avar 11y ago

To me your comment demonstrates why it's really hard to figure out anything worthwhile about SEO. You wrote that code almost 8 years ago[1]. Google and other search engines have changed a lot since then. Who knows if this has any impact today? Actually did you even A/B test it at the time? Or just turn it on along with a bunch of other changes? It's hard to tell, and there's so much SEO (mis)information out there based on old anecdotes, and all the players doing A/B testing for this sort of thing on a big scale keep their data to themselves. 1. https://github.com/reddit/reddit/commit/353ad2a https://github.com/reddit/reddit/commit/353ad2a

jedberg 11y ago
We did not A/B test it, but we didn't release many changes at the same time either, so it's hard to say for sure, but we were pretty sure it made a difference. Also, SEO is not supposed to be easy, Google purposely makes it hard. They don't want you to be successful (with good reason). They've always said, make good content that easy for a human to find and Google will find it too. That's probably the best SEO advice you can get.
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 avar 11y ago

jedberg 11y ago

We did not A/B test it, but we didn't release many changes at the same time either, so it's hard to say for sure, but we were pretty sure it made a difference. Also, SEO is not supposed to be easy, Google purposely makes it hard. They don't want you to be successful (with good reason). They've always said, make good content that easy for a human to find and Google will find it too. That's probably the best SEO advice you can get.

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.