I know; tough audience to pose this question to. But think about this: in his 2008 book Buyology: Truth and Lies about Why We Buy, Martin Lindstrom concluded that 85% of our buying behavior is dictated by emotion, not rational thought. If Lindstrom is even close to being right, can we really use analytics to explain and predict behavior that is primarily emotional?
But wait; isn’t the concept of order allocation key to understanding customer behavior for direct marketers? The historical answer is “Of course.” When I started in this business in 1986, we were operating in a closed system. We mailed stuff, and consumers either responded or they didn’t. We religiously captured key codes, since we understood that every incoming order was caused by a piece of mail we had delivered (well, except for that 10-12% of pesky “white mail”). Our media decisions for the next campaign were properly guided by our transaction history.
A little historical perspective here: the ecosystem I just described existed in a time when my vacation contact information was the phone number of the real estate agent who had rented me the beach house. Think about that – a totally different era, with completely different expectations and behaviors (i.e. before the web, before hand held devices, etc.).
What does this mean for understanding consumer behavior? It means the fundamental question that we are asking ourselves has changed dramatically:
- 1986: What mail piece and corresponding house file segment or outside list caused this order?
- 2015: What exact combination of marketing contacts contributed to this consumer’s order?
The fundamental issue I see here is that the latter question is unanswerable (also the wrong question, but I’ll get to that in a minute). As a consultant, I get to touch lots of businesses and observe lots of business practices. Regarding order attribution systems, I have seen both elegantly simple and fiendishly clever (and complex) approaches. But the three things they have in common are:
- They represent the very best interpretation of the world as experienced by the company’s marketing team,
- They are inherently biased by the preconceptions of the owner, CEO, CMO, CFO, etc. (the bias often centering on whether they believe catalogs are “good” or “evil”), and
- They have little or no impact on whether the company is successful or not.
Forward thinking marketing practitioners are generally asking a slightly different question today: what combination of marketing contacts will drive the greatest amount of profit for a given customer segment over a period of time? This is a better question, primarily because it can be answered by a combination of media holdout testing and/or statistical modeling. But, it is not the question we will be asking ourselves in 2-3 years.
Think about this: why do abandon site triggered emails perform so much better than batch and blast bulk emails (if yours don’t, contact us now)? First, we can provide content that relates directly to the consumer’s recent visit to our website. Second, the consumer is raising their hand and telling us, “I am here! Please show me something interesting!” Abandon site triggered emails are successful because we are delivering the right content to the right consumer at the right time – sounds like marketing, yes? Like it or not, this is the future. So what can you do to prepare for it?
- Stop spending money on order attribution systems. Whatever you have is perfectly fine, and if you leave it alone for a while, you will get a better sense of your business trends.
- Start evaluating your house file marketing strategy by measuring the effectiveness of various contact strategies over periods of time. You will never optimize buyer contribution margin if you look at things on an event by event basis.
- Take the money and time you were going to spend on a new and improved attribution system, and spend it on improving your content and messaging; particularly on personalizing content based on what folks are looking at on your website.
- Most importantly, use your website visitor browsing data to improve marketing campaigns, both online and offline. They are visiting for a reason and telling you lots about what they intend to do. This is just as true for prospects as it is for customers, and the opportunities are no longer limited by data processing capabilities and cost, only by what you are able to imagine!
Allen Abbott is SVP Consulting & Client Services for Cohere One