When attribution becomes guesswork
At Loyalty Builders we are big believers in the power of direct marketing to generate revenue. We’re not alone. Catalogers, emailers, and multi-channel merchants send huge volumes of all types of communications every day. But the way, some of them measure the effectiveness of their campaigns has us scratching our heads.
All companies want to reduce their marketing costs without impacting revenue – especially catalogers facing the escalating costs of postage. Their main tool for measuring the effectiveness of their campaigns is attribution, the process of allocating sales revenue to the various pieces of marketing material sent to the customers who made purchases.
Sometimes attribution is easy: you send an email on Tuesday, and within the next day or so a recipient clicks on a link in the email and makes a purchase. It’s clear that the email gets all the revenue credit.
However, if the sale is made in a retail store, or if the customer received two catalogs, a flyer, and three emails in the preceding few weeks, attribution becomes extremely murky. But, the difficulty of the problem doesn’t stop companies from trying. The most common approach to attribution is ‘last touch,’ in which the most recent message received by the customer prior to the sale is given 100% of the credit for the purchase. It’s an easy answer, but a radical over-simplification.
More sophisticated solutions, called multi-channel attribution (MCA), can involve time decay algorithms, an immense amount of work, and are not even close to being precise. We have even seen a company credit every catalog in market at the time of a sale with 100% of the purchase price if the item purchased was in the catalog, often resulting in double and triple counting of the revenue. Sales due to those catalogs sure look a lot better to management with that scheme. Almost all of these solutions ignore what may be the most potent factor in retail sales, word of mouth, or advice from a friend.
We would like to dismiss MCA as simply a waste of time, but companies continue to use the results of attribution analyses to make important marketing decisions such as how many catalogs to mail, which customers should get a catalog, and even which marketers and vendors to hire and fire.
Our view is that when there are multiple communications in market, attempting to assign sales revenue to individual pieces is a futile exercise.
What’s the right thing to do? With multiple communications what should be measured is the effect of the total communications package – including print and online components. Treating and evaluating print and online campaigns as separate entities when the customer is exposed to the entire communications stream is, in our opinion, just plain crazy. Even in the simplest case, the common sense goal should be profitability of the entire package, not maximum revenue from an individual catalog or email.
Not only is the total approach simpler, our experience shows that when you build the campaigns using individualized marketing, where each prospect receives messages designed for his/her preferences, profitability is substantially higher. Read our white paper Abandon Attribution to Build an Effective Mail Plan to see how to do it.