Touchpoint attribution--the ability of marketers to attribute a sale to the correct marketing effort--has been called the holy grail of marketing. Does this seem a little extreme? Still not sure attribution should be considered the pinnacle of marketing? The claim is likely more appropriate than you might think. Conquering touchpoint attribution may lead you on an arduous journey, but there's a big payoff in the end. Here are the four perils marketers must overcome to fulfill this quest:
1. Data collection - Collecting data from each marketing event appears to be the biggest problem to overcome. In the online world, the solutions often touted for the best data collection involve using universal IDs and complex tagging. In the offline world, marketers face the complication of media touches that can't be tracked directly, but which may lead to "organic" responses. For all of these, the ever-increasing arsenal of tools is making it easier and easier to collect data to track the impact of these touches. We have never before had this much data to indicate the customer's thought patterns.
2. Modeling - As is so often the case, having data in hand is only half the battle. The data needs to be incorporated into a predictive model to assign weights to each type of touch (or each conceivable combination of touches) to discern how much of the revenue can be credited to each. In the offline world, marketing-mix modeling has been used historically for this exercise on aggregated data, but so much of the online data that we can now add is available at the individual customer level that multiple modeling approaches need to be integrated to construct the most effective model.
3. Optimization - After we've survived the ferocious battle to take in the valuable data and have won the prize of understanding what the data means, we face the challenge to optimize the value of that knowledge. Our new channel weights now become a set of assumptions that drive a what-if tool that demonstrates the implications of changing the budget allocation for each channel. The new scenarios are then tested by tweaking the channel-marketing spend, and observing whether the predicted effect is realized.
4. Customer engagement - The last obstacle is pulling it all together to know when, where, and how to best engage the customer. Certainly, the power of knowing how to allocate budget for the next campaign or for the entire next year of campaigns, plus knowing in advance your returns from that budget, is valuable information. But here's another useful by-product of touchpoint attribution: We can learn a lot about our customers' engagement preferences through the attribution process. The more we know about engagement and where the consumer is likely to start the product-purchase process (online versus offline, for instance), the more we may be able to influence it. So, we would be further strengthening our ability to not only measure impact--to know what each allocated marketing dollar is worth--but also to improve impact/value of that dollar.
What's awaiting you once you've demonstrated the strong character needed to clear these perils? An optimized process for deciding exactly where to put your next marketing dollar and a good idea of what that spend will yield. For marketers, there's no more worthwhile quest.
+ + + + + + +