Multiple Email Addresses for Customers Can Hurt Response

As marketing databases mature, they inevitably accumulate more data, perform more slowly, and—if you’re not careful—can impact your bottom line with poor targeting and reduced response rates. One glaring culprit is over-soliciting consumers that have multiple email addresses.

As email usage evolves, more and more people have multiple addresses (I use four on a regular basis). Practically everyone in the workforce has both a work email address and a personal address, and – as people become increasingly savvy to the wiles of marketers – they create multiple personal addresses for different purposes: registering on a website, signing up for a newsletter, order confirmations, and of course, the one they actually use for exchanging emails. Further, it’s not uncommon for companies to set up generic email addresses, such as [email protected].

Over time, these addresses can creep into the marketing system and skew results. A common cause is from order confirmations, specifically consumers using a different email address from one order to another. Here’s a simplified example that shows what I mean:

NAME        ORDER DATE        CONFIRMATION EMAIL
Jon Doe        12/11/10                 [email protected]
Jon Doe        01/01/2011             [email protected]
Jon Doe        01/06/2012             [email protected]

As orders are loaded into the database, the attached email addresses are captured and stored for marketing purposes. In our example, all three email addresses are valid, but they all belong to the same person. This can cause Jon Doe to receive three different offers per campaign. There are a few downsides to this:

•    If Jon routes all three emails into a single inbox (such as a smartphone), he could become annoyed and delete everything, or move the messages into his junk mail folder. By inundating Jon with offers, we lose relevance.

•    If Jon responds to the campaign with a purchase, our campaign report will show a 33% response rate rather than 100%, making it appear that we’re performing worse than we really are.

•    As Jon opens and clicks between the three offers to choose which one gives him the better deal, email metrics (opens, clicks, etc.) become artificially inflated.

•    If the campaign tests different offers, it’s harder to know which is more effective. For example, suppose one offers free shipping, another a discount on large orders, and the third uses a standard non-discounted offer. Since Jon has all three offers, he can simply open each one and choose which is best for him. This makes the standard offer appear less effective and impacts our revenue, since he’s decided to purchase and is now simply choosing the lowest price.

•    If you’re paying an email service provider based on volume, sending consumers multiple offers per campaign rather than one increases your fees.

The solution has been around since ancient times – before email, social media and even (gasp) before the Internet: old fashioned address-merge processing. Even in today’s modern, always-on, always-connected world, if people buy the stuff we’re selling, we usually have to ship it to them and we definitely have to charge their credit card. Therefore, our database should be chock full of “snail mail” addresses—both bill to and ship to—linked to each order. By running this data through commercial address matching software, we can assign unique customer numbers and attach these numbers to our email addresses. This lets us apply business rules to choose which one should be used for e-marketing.

Here’s what the email table looks like after doing so:

Consumer #    Email Address                              Primary
123                  [email protected]              Y
123                  [email protected]        N
123                  [email protected]                       N

The Consumer Number shows that all three addresses belong to the same person, but only one of these is flagged as the “primary,” which is the one to be used for marketing. In this example we’ve simply chosen the most recent, non-bounced email address provided by the consumer. In practice, you’ll want to enhance this rule by allowing consumers to adjust their solicitation preferences online, and you might also add logic to promote a different email address if the one being used isn’t producing.

Executed right, address-merge processing is still a valuable solution that has an essential spot in the toolkit that smart marketers are carrying into the digital world.

Jeff Fowler is president and founder of Decision Software.