I don't suggest avoiding segmentation -- that would be blasphemy! -- but devising a meaningful segmentation strategy requires clear goals, adequate data, and a long-term view, not an unquestioning devotion to the latest marketing buzzword.
Defining the goals is easy, right? Typically, we want to tailor content to consumer needs and preferences to improve response and build customer relationships.
Tailoring content effectively, however, requires clear indicators of consumer needs and preferences. Sometimes it's easy: You're a lawn care company and you know that consumers in different climates have different lawn-care issues. Or you're Amazon.com and have a zillion data points on what I've purchased and what other people who like what I like have purchased.
For the majority of e-mailers, the indicators are not so clear. Just because I bought the dog collar, does it mean I'm not interested in cat toys? If I haven't found something of interest in the last several e-mails you've sent, do I want to sever my relationship with your company? If I don't fit the profile of the advertising department's target customer, should I be overlooked? Not necessarily.
advertisement
advertisement
In the examples above, e-mail response will improve if non-target consumers are eliminated from the mailing, but you sacrifice reach for better response. Is it worth it to mail an extra 100,000 e-mails in order to get 1,000 responses? Maybe. Because now you've moved those 1,000 consumers into a desirable segment. (Look at the Disaffection Index to help determine whether the additional reach from mailing to non-target customers was "worth it.")
Remember: an active e-mail list is a responsive e-mail list, so if you take the "wrong" customers out of your regular mail plan, they will quickly become former customers. Over time, an e-mail universe will shrink dramatically if there is no plan to keep in touch with non-target segments.
How do you collect those clear indicators of consumer preferences? Ask. Expressed preferences, entered by the consumer via a profile page, are a million times better than inferred preferences, as the examples above show. General desires and dislikes are easily gathered through surveys. Ask a few questions on your unsubscribe page to get deeper insight into why a consumer is opting out and ask your opt-ins what they'd like from your e-mail program.
If you define goals, collect consumer preference data, and make a plan for the "segments left behind," you'll be on your way to a useful segmentation strategy.