The opening discussions of deploying the Tealium AudienceStream™ real-time action engine always revolve around use cases and data. Often times the discussion quickly jumps from the use case right to the data, i.e., which attributes to create, but this skips over many important details and is not our best practice.
Let’s think about this using some (hypothetical) weekend plans.
Let’s say I received an invitation for a party this Saturday night at an unfamiliar venue. It would be unwise of me to jump in my car and just start driving without knowing where to go, how to get there, or when I’ll show up. So the two key data points of this party that I need are 1) when and 2) where. By knowing where the venue is located I am able to figure out how long it will take to drive, which will tell me when I should leave. Now I can jump in the car, buckle my seat belt, whip out my Rand McNally Road Atlas and wonder, “Hey, why do they call it Arby’s?” I mean … enter the address into the navigation system.
Building off of this in the AudienceStream, we can think of the invitation as 1) which vendors you have, and 2) which visitors do you want to target. Then a series of questions allow you to work backwards to the data. Here is a sample questionnaire to use:
For example:
This is one approach that will help ensure you have all the necessary information to target the right visitor at the right place at the right time as part of an effective cart abandonment campaign. This type of approach can be applied to many other important use cases.
Fun fact, in case you were wondering: “Arby’s” is the phonetic pronunciation of “R.B.’s”, which stands for … “Roast Beef.”