The value of split-testing

[Norri]

This screenshot clearly demonstrates the value of split-testing. I increased my results by 42%! That's awesome!

Now that I have a winner, I can take the winner of this test and tweak it some more to squeeze out a few more conversions.

This particular test was for an opt-in list where my goal is to get 2,000 subscribers in under 12 months. I was on-track before the test. Can you imagine how much I've sped things up by doing proper testing? A few percentage points difference might not seem like much but look at the numbers: 50 conversions versus 71. I'll take the 71 thank you very much!

Somebody wise once said that marketing IS testing. If you're not testing, you're not marketing. You're gambling. Ever since I read that, I've been hooked and it continues to pay off, each and every time

[Rod]

I'm glad to hear it has worked out for you

[Mark Atkinson]

What do you use to manage your split-testing, Norri?

[Norri]

Google Website Optimizer. They're phasing it out in favour of Google Analytics Experiments but either works well for straight-forward split-testing. If you want to do multivariate testing, GWO is the way to go still.

I've used something else in the past that wasn't free but can't remember it's name right now :P

[dfsa]

Nearly 2% variance. That is good. What was your cost input between Original and the Variation1?

[Norri]

To be honest I wasn't tracking that :P I pay about 7c a click for AdWords clicks on this particular campaign. A fluke really but one I intend on taking full advantage of!