Which requirement do we have for a whole website redesign test?
here is my problem:
we have a fully redesign of our big webshop.
We want to test it, but we are having trouble testing it.
We wanted to test it via a test in google analytics. We achieved that, by getting two split designs via asigning a code to the user.
If the user was selected by google- the first click on our page assigned him to the new design in his session. So he made it through the site with getting the new design within this session. Our problem is, that we are not able to see the clickpath and the behaviour on the page. We thought that we can achieve this with optimizely, but reseantly i can not find any solution for this.
Our problem is, that we can not take a real split-url-test and have only the code in the usersession, which assigns the new design.
Do you have any idea to solve that problem?
thx very much for your help!
If you are wanting to test a full site redesign with Optimizley, the standard way to achieve this would be via our URL redirect. You could run this experiment to target your domain by 'subtring match' so it would run on every page. Your original would be on your site:
And your redirect would be to:
Then, once on this domain, they would be browsing your site on your redesign so they would not enter the original again (and if they did browse back to the original, they would be redirected again).
So that's a typical approach - details of how to set this up are detailed in our article, here.
If I've misunderstood anything, please let me know! Would you be able to provide any more details of why this would not be achievable, or perhaps also more details on the problem you're facing settings this up?
I hope that helps!
First time optimizely tester here... this was a great response! We're in the process of launching a full site redesign test. The only difference is that we created a subdomain (www2) and that's where we're redirecting the variation traffic to.
I'm having issues because my homepage is the only page that's redirecting to the variation. The URL targeting is set to substring match. My thought is that www.xyz.com/locations could redirect to www2.xyz.com/locations if I qualified for the testing experiment.
Any ideas what this newbie may be missing?