Narrowcasting and the behavioural target

Behavioural targeting (BT) anonymously targets Internet users based on observed previous behaviour. It works, some experts say, under a rule based model whereby if a user does X, he is presented with Y, based on application of recency and frequency as key indicators.

Recency informs of how long since the user performed a certain action; Frequency establishes how often

The 3 basic questions BT asks are: where is the user now? where was the user before? what did he do there?

However, what I see as a clear challenge for the model, is that it will need to predict between intent based actions and routine based ones. How to predict intent is something I cannot answer ;-) I like the fact that it is a narrowcasting model versus a broadcasting one, it should no doubt deliver higher relevancy if the rules are efficiently designed and implemented. On the other hand, I still would like to see how to achieve and fine tune this efficiency: recency matters, but what about behavioral profiling; editorial adjacency; media utilisation models; depth of content versus vertical industry consumption…?

Narrowcasting illustration courtesy of Qeam. More information on this topic can be found in OMMA Behavioural

Tags: behaviour indicators, behavioural targeting, consumer habits, consumption habits, frequency, narrowcasting, online marketing, recency