Fall out report - use page as dimension or the click to action engagement on the page as dimension

[AnqiZh]

Hi, our team are trying to find out the containment rate of our eform by fallout table. My colleague and i are using different dimensions and getting different answer. E.g. if we use page as demensions, the containment rate will be lower than the result if we use C55 - Page (click to action engagement on the page). Please can anyone help us to figure out how to get the best answer? 

[MandyGeorge]

Fallouts can be a bit tricky, as each step is mandatory to move on to the next step. Check to make sure that all of your conditions match. As well, there's 'next hit' and 'eventual' options for between the rows - that is another thing that could cause differences. 

It is possible it's related to the dimensions you're using. The default page name dimension doesn't fire on custom links, but you can set up evar/prop versions of page name to fire on custom links. So, if your action could happen on a page view or a custom link instance, using the default page name will only count actual page views, whereas a prop version of page name would capture both page views and custom link clicks (if it's set up to capture page name on every hit).

[Jennifer_Dungan]

This might be hard to diagnose without seeing your data or understanding how things are tagged.... the "rates" aside, what do the raw numbers look like in your funnel? Are there ways for people to bypass clicks during the flow (which would result in more Page data, than Click data.. and this could significantly change the calculations). Is your funnel based on Visit or Visitor? Is it looking at Eventual Path or Next Hit?

As with anything, my first recommendation is to test the flow(s)... try acting like a user... don't just follow the path you expect the users to take, try to take unexpected paths... try using your browsers back and forward buttons (this will then track additional pages but not your custom clicks), try going straight into pages bypassing some of the steps, etc. Can a user leave the site open for a hour, then come back and resume a journey?

Watch the data that is sent to Adobe and really try to understand the nuances between different behaviours, and what that does to your data collection.