Rich-text Reply

Optimizely for website redesign - best practices

mzak 01-18-15

Optimizely for website redesign - best practices

Hi,

 

we are releasing a new version of our website and we wanted to use optimizely to compare it against the old version. From this post I understand that the recommended way to do it would be via redirects, eg. users coming to mywebsite.com would be redirected to mywebsite.com/version2.

 

I would assume that for some users 'mywebsite.com/version2' url might look less reliable than 'mywebsite.com'. From your experience do you know what kind of urls would minize this effect (eg is /new better than /version2)? And anyway, do you perhaps have data on how strong this 'less reliably looking url' effect might be?

 

Best

 

Maciej

Level 1

Joel_Balmer 01-19-15
 

Re: Optimizely for website redesign - best practices

Hi Maciej!

Yes you're right, running a redirect will be the most appropriate way to run this experiment (in terms of performance, efficiency, simplicity, etc).

And yes it's possible that the name of the URL can influence visitor behaviour (if you were interested, you could potentially redirect to an identical version of the site, but on a different URL, to test this out)!

From my experience with redirect tests, customers use an innocuous directly such as 'example.com/home'. But equally often I will see users redirecting to the same URL with a query parameter (for this, you must make sure your URL targeting conditions are correct, insofar as it excludea the URL you are redirecting to). This query parameter can be something similar to the above 'readable' name, or can be a code that makes sense to you i.e example.com/?nv2=true.

I hope that gives you some ideas!