All Collections
Nexternal: A/B Test Store Design Variations (without Google Optimize)
Nexternal: A/B Test Store Design Variations (without Google Optimize)
Split test or A/B test store variations to determine which design is more effective.
Pamela Topper avatar
Written by Pamela Topper
Updated over a week ago

Product: Nexternal

Opening Note: As you can see below, there are multiple steps to achieve this goal. We have not verified these steps for several years now and cannot consult if you are unsuccessful. We recommend using Google Optimize instead, which is built for this purpose. Click Here for instructions.

If you wish to proceed without Google Optimize, follow these steps:

1. Log in to your Order Management System (OMS). Go to Settings > Policies, Addenda, and Disclaimers. Enter your Google Analytics Tracking code if it's not already there, and your Google Analytics Domain, and click Finish to save this information.

2. Click Layout and create a staging layout from live layout values. (You must archive and then delete the current staging layout if one exists, and we recommend archiving the live layout first as well).

3. Edit the staging layout with your experiment - this will be the experimental layout.

4. Test your staging layout in several browsers. When you are happy with it, archive it and name it.

5. Log in to Google Analytics, select Behavior > Experiments, and click Start Experimenting.

6. Name your Experiment "Experiment 1", set a goal (Perhaps page views, revenue or goal conversions), a percentage of traffic, and email notifications (we recommend On), and click Next Step.

7. Enter the URL of any page of your live store, name the page Live Layout, and check "Consolidate experiment for other content reports."

8. For Variation 1, change the select box from http:// to relative, then paste in the query string of your experimental layout. Do not enter the full URL. Name this variation "version2", then click Next Step.

9. Copy your Experiment Code in Step 3 to your clipboard. Paste it into your HTML editor, then paste the following directly below the first comment:

<script> _udn = "";</script></p>

10. Replace "" with your Google Analytics Domain set in the OMS.

11. In the OMS, toggle to your live layout, Edit the Advanced Layout, add your revised Experiment Code to the top of the Head Addendum, and click Finish.

12. Click Next Step in Google Analytics. Confirm that it validates and click Start Experiment.

13. Visit your store in multiple browsers using the forced reset URL ( ACCOUNT NAME/?Reset=True&Staging=False). If you use at least three browsers, then you will probably see at least one instance of each of your two store variations. If not, clear your cookies in one browser and revisit the site, clearing your cookies each time, until you see both versions. If you don't see both versions after clearing your cookies several times, then there is probably an issue.

14. Check on your experiment in Google Analytics daily and use the results to refine your store.

FYI, it takes time for the experiment results to populate. So set it up, and then check on it the next day.

rev: 10/28/2013

Did this answer your question?