The most engaging content is that which speaks directly to a person’s needs.
In the future, with the help of AI and big data, marketers will be able to slice and dice their audiences like the most precise cheese grater the world has ever seen.They’ll be so selectively chopped up individuals will be served utterly unique messages, crafted specifically for them.
Evolving into a sophisticated breed, users’ threshold for content that is broad has virtually baselined. The argument for personalisation through segmentation – where audiences are divided into subgroups for effective targeting – is clear.
Here, we cover the four most common segmentation blunders so that you don’t fall afoul of them yourself.
1. Failing to segment full stop
The biggest problem? Not bothering to segment at all. If you’re a small start-up serving the ‘nichest’ of niche audiences, you’re excused – for now. But for large brands with established audiences and full data sets, the phrase ‘shooting yourself in the foot’ comes to mind.
Are marketers being led astray here, we wonder, by Byron Sharp’s ‘market to the masses’ appeal? Something of an industry iconoclast, this University of Adelaide professor’s unorthodox advice in his 2010 bestseller, How Brands Grow, was grabbed worldwide by marketers trying to get a handle on the new, fraught digital landscape. In his book, Sharp is derisive of surgical targeting. “Sales growth won’t come from relentlessly targeting a particular segment of a brand’s buyers,” he declares. “This fantasy is harming marketing effectiveness.”
Instead, Sharp argues that reach and frequency are more effective than segmentation. This way, the sales net can be widened to capture not just brand loyalists (who can, after all, make only so many product purchases), but potential buyers who would have otherwise been excluded because they didn’t fit the target group criteria. Since brand loyalty is a thing of the past, Sharp adds, once someone is aware of your product and service, they’ll suffer no compunctions switching to it away from a competitor if they’re suitably impressed.
There are bundles of research papers to back up Sharp’s claims. But we’d never recommend the ‘spray and pray’ approach to publishers – and certainly not for content marketers, whose aim is to develop a known audience through serving consistently useful, meaningful and relevant content.
Our rebuttal to Sharp’s logic is this: if you’re only geared towards achieving a broad reach, then you’ll only fulfil a ‘top of funnel’, awareness-raising objective. When delivering content to those who have interacted with your brand – shown interest in it, engaged with it, demonstrated a clear pattern of preferences – this is when targeting through audience segmentation is invaluable. By putting the right content in front of the right people, segmentation is one way that brands can nurture individuals and earn their loyalty and trust.
2. Getting audience understanding all wrong
The second common segmentation slip-up is failing to develop sophisticated audience personas or neglecting to evolve them over time. In other words, you can’t segment your audience because you don’t really know who they are.
A related misstep: confusing audiences with segments.
Let’s say you’re a major sports company committed to content marketing. You’re hoping to sell products and services to both gym owners and fitness-minded individuals – with offerings spanning gym classes, sporting events, sweatbands, top-range rowing machines and hundreds more.
Now, let’s be clear. These two groups I’ve just named are two different audiences. They are not segments.
A segment would be taking one of those audiences (let’s go with ‘fitness-minded individuals’) and dissecting it into sub-categories across gender, age, fitness level and location. This is an example of basic segmentation.
The more savvy marketers will go beyond these basic demographics however – which are increasingly regarded more suited to the pre-digital age of clunky and limited data sets. Liam Brennan, Global Digital Director & Director of Innovation Programmes at MediaCom, puts forward the case that subtler signals can be far more effective at leveraging certain groups. In a column for Ad Age, he advocates the ‘BEM approach’ – an audience segmentation model based on behaviours, emotions and moments. He outlines the categories as follows:
- “Behaviors: Have consumers demonstrated (or exhibited proxy behavior) that indicates interest in a specific or related product area? Have they actively sought out or mentioned a particular product or service?
- Emotions: Has a particular product or service suddenly become more relevant? Are consumers posting emotional responses that suggest they would be receptive to certain brand messages?
- Moments: Has a trigger like weather, transportation snarls, or other live events caused a product or service to become suddenly relevant? Has the consumer entered a specific location where helpful products are easily available?”
The BEM model gets the Mahlab nod because it recognises that external, real-world and personal factors are always going to influence how an audience interacts with your brand. And it acknowledges too that these factors are not fixed – they vary over time.
It also motivates organisations to think of audiences as not just bundles of superficial traits, but as real people, whose actions define them more than their birthrights or backgrounds. Compared to traditional demographic targeting, using the BEM model “brand preference and purchase intent lift can nearly double,” says Brennan.
3. Conflating segments with stereotypes
There’s another reason to be wary when using demographic data to segment: it can make you as a marketer answerable to unintended acts of outrageous bias.
“Only crap marketers mistake stereotypes for segments,” huffed Mark Ritson in a Marketing Week column posted in July. His main grizzle-point was, as ever, the category of millennials. Marketers are fascinated by this group. Endless campaigns, products and services are rolled out to pander just to them. To take two recent examples, a new Paris airline called ‘Joon’ has seats available only for the millennial tribe, and Vodafone has created a mobile network called Voxi designed exclusively for the under-25s.
The problem is that the idea of the ‘millennial’ as a coherent entity is pure bunk. The millennial doesn’t represent any stable set of characteristics, buyer preferences, interests, politics or habits. It represents a set of superficial assumptions and reflects mostly our own obsession with youth. On almost every point relevant to most brands, people born somewhere between 1981 and 2000 share more differences than they do similarities.
Segmentation based on gender is another leading offender. It’s how we still see content blaring out the same old Benny Hill era prejudices, and one reason why representation of women in advertising hasn’t improved in a decade.
If you segment your audience into men and women, the data may very well insist that you’ll get more engagement with an article called ‘How to get rid of those nasty shower stains’ by putting it in front of mainly women. And perhaps your research tells you that men are more interested in articles about career success – so you may be motivated to exclude women in your targeting. Not a way to move the world forward really, is it?
4. Not using the data on hand
Not using the data you have to better serve your audience is nothing short of a travesty. And it leaves you looking ham-fisted too.
I’m going to bring up an example I’ve used in a previous article, for no other reason that it’s a massive bugbear and positively scorches my under-collar region with annoyance.
The situation is this: my health insurance provider is a content marketer – good on them. I previously subscribed to their enewsletter – good on me. But after receiving a consecutive stream of irrelevant emails, I gave up on them. I would get sent stories on cars even though the only vehicle I own is a bike. I would receive information on how to fireproof my house, when I live in an apartment block. Just because I was a thirty-something man, they’d leapt to the same antiquated assumptions an estranged uncle would make at Christmas lunch.
And the worst thing about all this is that they knew these things. They’re an insurance company! I’d handed them over scores of data to them when I first signed-up – freely disclosing both where I lived and how I got from point A to point B. They had no excuse.
It’s true that some companies won’t be flush with such quality and quantity of data. And demanding information that doesn’t bring the customer any clear, perceivable benefit is not a great idea (a topic for a future article, guaranteed). But if you have the data before you, and you’re not making the most of it, you’re throwing away a blessing that most others would – and do – pay a premium for.
Kate Prendergast contributed to the writing of this article.
