The problem with the messaging matrix

“Assumption is the mother of all fuck-ups.” These words of wisdom were drilled into me by Derek Lucas, a retired Royal Marine Commando whom I worked for as a teenager. Thirty years later, they still ring in my ears.

Unpack any fuck up, and you’ll find an assumption. Fukushima: we’ll build the reactor to withstand the largest earthquake experienced so far. The GFC: repackaging the downside risk and selling it on eliminates it. Brexit: what people tell a pollster reflects how they actually vote.

Whenever you find an assumption, you’ll usually also find a fuck up. Find two assumptions, and the fuck up is likely to be bigger.

Which brings me to that beloved and reviled marketing tool, the ‘messaging matrix’. Why? Because it’s based on two assumptions. Two big ones.

Assumption one: sales funnels are an accurate representation of the phases people go through when purchasing.

Assumption two: segmentation studies accurately map the different types of shoppers in a category, including their needs and purchase drivers.

But do these assumptions stand up to scrutiny? Not really.

The sales funnel is about 100 years old and was concocted mostly out of thin air by salespeople and economists. Salespeople want to close deals, and rightly so – it’s their job. But they’re not primarily interested in modelling ‘reality’. Indeed, those who developed the sales funnel were selling the funnel itself to businesses – i.e. were concocting a methodology to sell.

Economists are not interested in ‘reality’ either. They are primarily interested in theoretical models, which is why they are routinely blindsided when the real world doesn’t align with those models. (For more on the history of the sales funnel, Paul Feldwick’s Anatomy of Humbug is a great read.)

In the real world, humans aren’t very rational at all, certainly not in the way the sales funnel depicts. We very rarely step methodically along a linear process from Awareness through Familiarity, pausing at Consideration before arriving at Purchase, then repeating it all out of Loyalty. Indeed, McKinsey shot holes through the sales funnel back in 2009, yet a decade later it lives on.

What about segmentation studies? Don’t they capture ‘reality’?

In a word: no. Anyone who has worked on creating customer segments from the vast pool of data delivered by the foundation survey – and I have done this several times myself – will tell you there is nothing more creative than a segmentation study. Nothing.

The patterns and bits of data selected to create the segments are chosen at the cost of ignoring others, consciously or not. Even the way you name the segments has been shown to influence how clients use them. In a nutshell, segments are proxies for reality – sometimes fairly good proxies, but usually crap ones. (This is why sometimes when you try to recruit segments for qualitative research you can’t find them: they only ever existed in the study, not in the real world.)

So here’s where it gets problematic. Messaging matrices are developed by plotting the sales funnel down one axis, and your segments along the other. Put differently, you’re plotting the theoretically dubious against creative proxies. Then you develop messages for each cell, and your messaging matrix is complete, after which it becomes sacrosanct – an almost holy document that guides your creative and media agencies.

Why does this matter? Why does it lead to fuck-ups?

It often leads to fuck-ups because building big, successful businesses involves marketing your way into owning a single-minded brand positioning in the minds of your customers, prospects, and employees. Doing this efficiently – being single-minded – requires simplicity and focus. Yet all too often the fragmented messaging that comes from brands who dogmatically adhere to matrices is the very opposite of single-minded. It dilutes the clarity and focus of your brand in the mind of the market, undermining your ability to efficiently arrive at a distinctive positioning (which itself is needed in order to grow sustainably).

Additionally, the complexity of messaging matrices drives costs across the marketing production process.

And finally, the entire approach assumes that media targeting is so sophisticated it can identify members of a priority segment, can tell where they are on the sales funnel and then can accurately target them with the right message. Anyone who’s been retargeted for days by a product you’re either already purchased or rejected knows how dubious this assumption is.

So don’t assume a messaging matrix is what you need. Don’t assume they’re scientific or based on anything particularly solid. Don’t assume they’ll drive the business result you’re looking for. Why? Because assumptions are the mother of all fuck ups. A matrix might work… but likely it won’t.

To round out the point, I’ll leave you with the messaging matrix for Canadian Club which was behind the campaign that won a Grand Effie in 2015.

[Click to enlarge]