“We’ve got to use every piece of data and piece of information, and hopefully that will help us be accurate." -Billy Bean, Oakland A’s GM.
In Performance Marketing the average is the enemy. More data is always better, particularly if we can look beyond the average. Take any one of the metrics marketers manage:
By definition, these metrics are about a big group of people. We're summarizing what happened and when we talk in averages, we sometimes lose track of the fact that there are distinct groups of prospects or customers inside the average that make up that number.
For example, let's look at this group of customers: reactivated purchasers. By definition, these customers have at least two purchases in their history, and they gained repeat status by purchasing in the last, say, 12 months. If we split this group of customers into whether they are engaged with the email program, we get two distinct groups. Now I'm going to ask the type of question that we've heard smart marketers ask themselves: "If I cannot reach this group of customers by email, should I spend more money to reach them in my remaining channels (social, display, search, video, direct mail, etc.)?" That question is the foundation of looking for lift.
Marketer #1: “YES, I'd be willing to spend more to reach non-email engaged reactivated customers on Facebook because otherwise I'd be missing a chance to engage with them".
This is a valid hypothesis.
Marketer #2: "NO, these customers don't want to engage, and additional Facebook spend would not work anyway".
This, too, is a valid hypothesis.
An adept reader may also posit that as more channels are considered, both hypotheses could be true: perhaps you SHOULD spend more on Facebook but less on SEM or vice-versa.
LiftLab customers know that the right answers can be found by trusting and using the Scientific Method. They will kick off a series of tests on these audiences across all the remaining channels and get hard evidence as to what level of spend will maximize return on investment. Finding lift takes running experiments, and LiftLab sets you up to do it better than ever before.