Audience Segmentation: Friend or Foe?

"Given about five parameters – female, over 55, high income, city dwellers, upscale neighborhood – and you can come to some understanding of the group. But the more attributes you use, the more difficult the segments are to understand."

May Help Branding, Hurt Customer Experience

Audience segmentation, the technique of dividing a market into groups with similar characteristics, is designed to describe a group of customers in a way that suggests each person in that segment will behave in a certain desirable way. In the context of marketing, “desirable behavior’ typically means buying your product or more of it.

For example, you may find that women over 55, within the top income percentile, with an upscale city zipcode, may be interested in your expensive widgets. In that case, audience segmentation is your friend. If you want to target the marketing for your widget, you find the marketing vehicle (such as magazine, outdoor board, TV spot, newspaper) most like to reach the largest number of readers or viewers who best match your segment.

Moreover, because the segment itself helps you understand who you are selling to, you can use it to tailor the content of the ad to appeal to women, for example. As a bonus, you can also use the segment for some strategic decision making: where to open new stores, which companies to acquire, what type of staff to hire.

Segmentation is used to better understand customers in a particular group and works best when there are a small number of variable characteristics. Given about five parameters – female, over 55, high income, city dwellers, upscale neighborhood – and you can come to some understanding of the group. But the more attributes you use, the more difficult the segments are to understand.

Limits of Segmentation

Even with just a few dimensions, the human mind has difficulty processing dimensions that interact in a non-linear way. For instance, if older women with higher incomes tend to buy your widget, that’s one thing. But if they bought it from ages 55-65, but stop after 65, unless they live in the city on a smaller income – what do you make of that? See where this is going? (And it’s going there fast.)

Here’s the fundamental challenge: The easier your segment is to understand, the more likely it is to be flawed. Humans (and marketers) have a need to simplify life so they can model it in their minds and decide on an action. That’s not a bad thing; in fact, finding the model is a scientific process, and it’s used to understand things and make decisions and predictions.

Here’s the flaw. You need to make the model as simple as possible, but no simpler. And segments you can understand are by definition very simple, which means not terribly useful.

Segmentation Can Be Your Foe

Given that, segmentation should be restricted to marketing areas where there’s no alternative, such as advertising and branding: If Jennifer is in the same segment as Jane, they’ll be treated the same. Direct marketing is another story.

Remember, the segments are simple. So if we have 1,000 customer attributes to describe the two women, only a small fraction of those are used to describe the segment that includes them both. If we looked at more and more attributes, Jennifer and Jane might not even end up in the same segment – and might end up being treated very differently.

Back in the 80’s, there was no choice. Audience segmentation, still used for advertising and branding to mass markets, was the only option.

In the years since then, technology has evolved to support personalization on the individual level. The use of segments, beyond very broad-base segments like geography, is really only appropriate if the audience can’t be personalized at a 1:1 level, for example, with print ads, radio, or television spots, or when there's truly no individual data available.

One To One Personalization: A Better Friend

Outbound, company-initiated communication can be (and should be) as 1:1 personalized as inbound, customer-initiated interactions. There's no reason, no excuse, for not reaching out to a customer in a personalized, customized way at the right time, with the right offer.

Scheduled, segmentation-based campaigns are a thing of the past. Although targeting has improved with better analytics, there’s much more room for improvement

Even though digital marketing can make a non-personalized marketing practice bigger and faster, it doesn’t make it better: The sheer volume of ads served is what saves the day. But mass marketing decades ago worked in absolute numbers because some tiny fraction of mail recipients opened the envelope. No longer. It’s time to shift to a better option – and better friend.