I encourage you to read my earlier posts on Variables in Adobe Analytics.
So, you can now slice and dice the variables as per the different lenses :
Below are the lenses available to you :
- Scenario A) Depending on the values: whether they are Dynamic OR Static
- Scenario B) Depending on the type of value it holds: whether it is a normal variable OR an array
- Scenario C) Depending on the type of data it stores & How it is reported: Dimension OR Metrics
- Scenario D) Depending on the need for definition: whether they are predefined (built in) OR not predefined (custom)
- Scenario E) Depending on the type of capture: Whether their values are captured automatically OR need definition
- Scenario F) Depending on the type of nature: Whether they store Configuration Values or non configuration values.
However, the most important and widely used (and helpful in implementation as well)lens are :
Lens E : Whether a variable is Automatically captured or not
Lens F : Whether a variable is Configuration or not
We already discussed these in post 1. In our previous post we made an attempt to understand the technical side of Traffic Variables and Conversion Variables. However, it is very important to understand the functional side or the use case side of these variables. I have seen lot of people struggle with the below :
- What is the difference between eVars and a Props
- Should i use an evar or a prop
- How the data will look like
Point 3 is already addressed because we know the technical side of eVars and Props. Let’s dive in .
Props :
Prop is a custom traffic variable. The main use case for a prop is “pathing”. Pathing means flow. You want to see a sequence of values captured. The most common scenario when we want to see a sequence of values is for the page name variable. This will help us understand the path users are taking. We can draw inference from the starting values and the final values. Starting values in this case of a page name variable would become landing pages or entry points. The final values in this case become the exit pages. Ofcourse, we can understand what happens next. So, whenever there is a need to understand a sequence or navigation or pathing we use props.
Props also enable you to do what Adobe Analytics calls “Correlations”. These are not statistical correlations. Here, it means break down the values which are captured in the same image request. Read the word “same image request” again.
So, you can do things like :
Identify people on particular page are using what type of browser, coming from which cities, are using which devices etc.
What you cannot do is:
Identify the purchase on a site happened because of which event ? This is because the event has happened and was sent in a previous image request while the purchase event is sent on a different request.
eVars :
The above limitation leads us to eVars.
One of the main utilities that Marketers look from Analytics is to be able to understand “Attribution”. They would like to know – what was the driver behind a conversion on the site or who should be attributed the conversion on site. This becomes very important for eCommerce companies.
eVars remember values across different image requests. For example a user came through a site from a search campaign, made an internal search and finally made a purchase. If you use an eVar you can pinpoint and say that – purchase happened because of search event or abc campaign. While defining the eVar , you will have to define :
- Allocation Method – Most recent, linear or last
- Expiry – Visit etc. You have to set this as per the purchase cycle of your business. Some businesses like furniture would have a longer purchase cycle .
So, there you go – this is the basic use case of an eVar. Also, now you can breakdown entities across different image requests (Adobe calls this as sub relations). There are other use cases for an eVar as well like – merchandizing.
Coming back to our initial question : When to use eVar and When to use Prop
If you just need pathing insights – use a prop. If along with pathing you also need attribution insights – use an eVar.