Rich-text Reply

Track user through journey

nhatter 10-22-18

Track user through journey

[ Edited ]

A test I'm running with 2 variations is giving some very unusual results. 


The metrics I set up are:

  • (Primary) Clicks - graphic variations which link to another page
  • Page Views - the page which the graphic links to
  • Page Views - confirmation after conversion


The results show much lower clicks than page views it links to. The only thing I can think is happening here is Optimizely is not tracking the user journey from the primary metric through to conversion of the secondary metrics. It is including all view of the page regardless of how the user arrives to it or not. 


If this is the case, how do I set up the metric to only record page views IF the user arrived there by clicking the element in our primary metric?

Level 1

JasonDahlin 10-23-18

Re: Track user through journey

Hi @nhatter,

What you describe sounds correct... if you have more than one way of reaching the page, some users will reach the page via the content you added, others will reach it by some other link, or perhaps land on it directly, and all of these ways of viewing the page count as a conversion.


To count how many saw the page ONLY via the link, you'll want to put a URL parameter on those links and make your Page Views goal look for that parameter. This number should be nearly identical to the number of clicks... so what you have in place tells you how many are seeing the page and how many of them saw the page by clicking on the new content.



link to "" becomes ""

and modify your Page View goal appropriately.


However, I suggest that perhaps what you have in place is already correct. You've added a piece of content and are measuring how that content influences the customer journey. Some users will click on the link, others may be influenced by the content and end up on the target page but not through the link in the graphic.


If you are wanting to compare conversion rates of the target page and the thank-you page for all visitors vs those that click through the new content, you'll need to measure Page Views as you already are (every user who saw the target page and all of the conversions, for the A and B group) and then work out how to measure conversions for those who clicked through the link rather than accessing it some other way.


There are a few ways to do this, but my first thought is that if you have added a parameter to the content's link, you can set up a click event for the Submit button on the target page that only records when that parameter is present to see how many users submit that page (or, modify the URL of the thankyou page so that it contains a QS parameter that you record separately from the "all conversions" Page Views that you already record.

--Jason Dahlin
Analytics and Testing Guru Smiley Happy

Experimentation Hero