• by Phil Leggiere , August 4, 2006

The promise of e-mail as an advertising medium has been to bring brands and consumers together more directly than ever before. Reality, unfortunately, has proven more complicated. For consumers, finding the needle (the offer they really want) in the haystack of info-glut has become more and more frustrating, For advertisers, it's meant their "signal" has been increasingly lost amidst the noise. One way of moving beyond these limitations is for direct marketers to adopt and adapt behavioral methodologies still mostly associated with their brethren in the online advertising world, as Elaine O'Gorman, vice president of strategy of Atlanta-based interactive marketing firm Silverpop, explains below.

Behavioral Insider: Why has e-mail advertising remained largely siloed in relation to other Web advertising?

O'Gorman: The advantage of e-mail is that you have a more precise idea of who the customers are on your list than you would of just who's browsing your Web site. But it's a different mindset to do e-mail marketing from the one advertisers bring. E-mail marketing has been the province of direct marketers. And, frankly, most direct marketers don't do a good job of executing e-mail campaigns. They haven't changed the way they work. Most direct marketers are stuck in an old-school mentality where you do mass mailings based on very broad and rigid targeting criteria, the broader the better. Messaging and creative were set far in advance and once set were not seen as variable.

BI: What sorts of targeting data has "traditional" e-mail advertising relied on?

O'Gorman: The pinnacle, the holy grail of what's possible, is targeting that ties together multiple streams of data to paint a far richer composite picture of the customer. This includes transactional data on e-mail click-throughs, Web site analytics on search, browsing and shopping cart behavior, and then demographic data atop that.

E-mail is incredibly measurable, That's its great edge--but also a big challenge. The reality of Web analytics is that although there's a ton of information, there are only a small number of advertisers at this point who can really deploy it well. The ones that gave successfully integrated analytics and transactional data are retailers or other kinds of e-commerce firms which have deep insight into the buying cycle of their customers.

It helps if the buying cycle is relatively simple. If you are a b-to-b firm selling airplane engines, for instance, the length and complexity of the buying process will be far different from and tougher to model than yellow sweaters. If you have a wealth of transactional data about sweater-buying habits, you can use analytics in a relatively straightforward fashion to time the right moments to push e-mail promotions to close the sales loop.

BI: What has been the extent of behavioral targeting deployment by e-mail marketers so far? Who have the early adopters of next-generation e-mail targeting been?

O'Gorman: Using transactional data about previous purchase and browsing behavior to push information about new products to high probability targets is the most developed form of targeting. Another increasingly common form of e-mail targeting is based on tying together current transactional data with purchase history data to know when to respond to changes in behavior. If you find a steady customer has suddenly dropped off in shopping frequency, or conversely, someone who's been a relatively inactive shopper on your house list suddenly steps up their browsing or shopping activity, you've got an opportunity to seize.

BI: Can you cite examples or case studies of how BT is currently being successfully deployed in the industry?

O'Gorman: An online banking firm we work with has used customized purchase life cycle data to get real-time visibility into where customers are in their purchase process. In the firm's business model, customers have to go through several steps, often taking some time between when they sign up and when the bank gets paid. They have to add merchants to their account and then actually make purchases with those vendors before the banks get paid.

So the bank employs an automated system to push particular messages at specific times. Knowing, for instance, that customers who sign up but don't choose merchants after X number of days are likely to drop out, the bank triggers follow-up push ads optimally timed to prompt action.

So the "campaign" is activated when consumers sign up, and is timed to coincide with phases of their purchase cycle based on what we know about their buying behavior and analytics about [broader] customer purchase patterns.

BI: What are the main challenges and opportunities for BT in this space going forward?

O'Gorman: E-mail is still a very immature medium. There are still all sorts of formatting and deliverability issues that preoccupy marketers. The time e-mail marketers should be spending on learning much more about their customers is instead being spent mostly on putting out daily fires, like "why is gmail blocked today." But marketers who are making the upfront investments are finding big ROI advantages to doing more enhanced targeting. An emerging body of research is documenting this trend. For instance, JupiterResearch recently found that though enhanced targeting methods increase amount of investment by about two-and-a half-times versus traditional batch and blast, they boosts ROI by over nine times, with big improvements in unique opens, click-through and conversion rates. It's the kind of calculation that's becoming more compelling for serious e-mail marketers every day.