Discrepancy in total views in ab test | Community
Skip to main content
July 17, 2013
Question

Discrepancy in total views in ab test

  • July 17, 2013
  • 4 replies
  • 1255 views
We have two landing pages (LP-A and LP-B) set up in two smart campaigns, and also created a test group to add these two pages (Test-A and Test-B) into ab test. The traffic to this test should be splited roughly 50/50. Then I created a landing page report in analytics to include these 4 items, however, the report shows LP-A has a much lower 'Total Views' than Test-A. That shouldn't had happened because they are actually the same page. What could be the reason that the discrepancy occurred in this test?
This post is no longer active and is closed to new replies. Need help? Start a new post to ask your question.

4 replies

July 17, 2013
Please provide some screen shots of what you are observing and a bit more detail about the Test Group you created.
July 17, 2013
screenshot included:



Row 3 and 4 represent LP-A and LP-B; row 1 and 2 represent Test-A and Test-B.

Thanks.
July 17, 2013
Hey Aaron, two immediate casues to consider:
  • Test viewers of the pages will look at a specific page, possibly more than the other
  • Prospects viewing the page might reload, in which case you'd expect them to see the same one multiple times.


Hope this helps.

July 18, 2013
I don't understand why you say that the LP-A in row 3 and Test-A in row 1 are the same landing page.  Please confirm the following.
  • They may appear to be identical, but are physically separate landing pages? 
  • Different leads are going to those landing pages? 
  • Whatever you are using to drive traffic to those two landing pages has a different link for the two landing pages?
  • Is the content the same or different on whatever you are using to drive the traffic to those two landing pages?
  • The landing page in row 3 has 1 less day of activity than the landing page in row 1?
If you are driving different people to these landing pages using the same CTA for the same period of time, your sample size is so small that the difference is probably not statistically significant.