It’s the End of Segmentation As We Know It (And I Feel Fine)

OK, it’s a sensationalist headline. I don’t really think segmentation should be ditched. Far from it.

But segmentation based on high-level, limited demographics and not behavior is dead in brand marketing because it ignores the real promise of social media.

Segmentation as practiced in marketing and advertising today seeks to group people into narrow categories based on demographics. Typical market segments might include the stereotypical soccer mom, limousine liberal, weekend warrior or other combinations of the standard variables age, life-cycle stage, gender, income, social class, and lifestyle.

These variables don’t even begin to describe the whole person, but they are close enough for current practice. However, relying on segment stereotypes can fail to reveal important behaviors.

Back in the late ‘90s, I met with the CTO of the Nielsen Company (my boss’ boss) and offered a heretical thesis: Demographics are crap.

She let me explain this shocking assertion, and then thought of an example from her own life. “We’re a double-executive-income family living in a pricey Zip code, and you would never know from that that we consume about a case of Kraft Macaroni and Cheese a month,” she said. Right. There’s no accounting for taste . . .

But flippancy aside, she was right. No marketer would target her family as a brand-loyal mac and cheese consumer. They’d be more likely to try to sell them truffles and expensive French cheese.

It makes you wonder what other outlier behaviors families like hers (or yours) exhibit. Are you a wine-lover living in a beer and pickup truck neighborhood? Or a Republican living in a Democratic district?

That’s the problem with the current practice of segmentation: It’s too imprecise. Yet it does work, however imperfectly.

A great example of data-driven segmentation working very successfully was the Obama campaign during the 2012 presidential election. The campaign was not satisfied with addressing broadly general segments. Instead, by combining Big Data techniques and social media, the Obama campaign used micro-segmentation and targeting to pull off a striking victory. By understanding the electorate at a deeper, more personal level, the Obama campaign was able to better-target their fundraising and get-out-the-vote efforts.

According to KPDi[1], this degree of “consumer segmentation” had strategic importance in two respects:

• The Obama campaign was able to show a genuine appreciation of each person’s role in the re-election effort. People were made to feel important.
• It reflected a “customer-service” orientation that was projected outward from all campaign touch points. The get-out-the-vote superiority was just one of the benefits. More esoterically, the voter-centric model got the campaign out of its own head, and focused on the task at hand: winning vs. ideology.

So you might think that this example shows that the old model works. Well maybe, but it needs to be adapted to the fact that there are more people using social media today than were using the entire Internet in 2008.

An article in by Bob Garfield in MediaPost[2] reveals the lesson of the election: “It should now be blindingly obvious to every marketer, and to more evolved bipeds, that nothing that comes out of the mouth of a brand or any other institution has remotely the influence of what comes from the mouths of 7 billion bystanders freely trading opinions online.”

Time magazine[3] states that social media was extremely important to the get-out-the-vote efforts of the Obama campaign:

Online, the get-out-the-vote effort continued with a first-ever attempt at using Facebook on a mass scale to replicate the door-knocking efforts of field organizers. In the final weeks of the campaign, people who had downloaded an app were sent messages with pictures of their friends in swing states. They were told to click a button to automatically urge those targeted voters to take certain actions, such as registering to vote, voting early or getting to the polls. The campaign found that roughly 1 in 5 people contacted by a Facebook pal acted on the request, in large part because the message came from someone they knew.

The numbers also led the campaign to escort their man down roads not usually taken in the late stages of a presidential campaign. In August, Obama decided to answer questions on the social news website Reddit, which many of the President’s senior aides did not know about. “Why did we put Barack Obama on Reddit?” an official asked rhetorically. “Because a whole bunch of our turnout targets were on Reddit.”

According to Ad Age,[4] for the period between Sept. 1 and Oct. 14, the Obama camp had 497 creatives deployed across the Web compared with the Romney camp’s 90.

So advanced segmentation works, both on and off line. What does that mean for the future of brand marketing? That’s an answer for the next post.

Next up: Infinite Touches – Segmentation on Steroids