Millward Brown Link Alternative

How do you measure whether your carefully crafted ad campaign ‘works’?

First, you need to define what ‘works’ means.

Traditionally, advertising research tended to focus on the message. Instead of measuring whether a campaign increases sales – which is hard – research shifted to intermediate effects such as brand awareness, message comprehension and reported feelings. Millward Brown’s Link is a well-known copy-testing framework that aims to capture many of these facets in one unified methodology.

In this blog, you won’t read a critique on whether or not Millward Brown’s Link methodology succeeds in measuring an advertising message is delivered, understood and engaging. It could very well be supremely reliable and valid in that regard.

However, science’s current understanding of advertising effects do cause me to raise doubts whether the many underlying intermediate metrics, such as message delivery, understanding and creative engagement, were ever relevant to begin with.

If we look at science’s current understanding of the relationship between advertising and business success – that is, sales – we see strong evidence that we were looking at the wrong numbers all along.

What Advertisers Don’t Want to Hear: Communication Success is not Business Success

An ad can do a great job of sticking in the mind, raising brand awareness, evoking a positive brand image and all those good things. Yet, it could very well do nothing for a business’ bottom line.

This is the somewhat saddening conclusion of Binet and Field’s groundbreaking 2008 IPA paper, Marketing in the Era of Accountability:

“Intermediate effects of advertising such as brand awareness, image, and other measures of consumer brand health, do not correlate reliably with business performance.”

In other words: sure, you can measure brand awareness and message comprehension, but that tells you very little about whether your campaign truly works. It’s like trying to predict tomorrows weather by counting the number of umbrella’s today. There’s certainly a correlation, but it’s low and unreliable.

This poses problems with traditional explicit research methods, which are based on the assumption that brand awareness and ‘getting the message’ is what a campaign needs in order to work. As it turns out, it doesn’t.

Why do classic metrics of brand and message say so little of actual brand growth?

In order to better understand the problems with classic metrics such as brand awareness, image, and message comprehension, let’s turn to the current state of the art science on how advertising works. What’s the journey from viewing an ad to the purchase at the cash register?

In ‘How Brands Grow’, Dr. Byron Sharp challenges a great number of myths on how marketing works using rigorous empirical data.

The old view is predominated by linear models, such as the AIDA model, assuming a step-by-step order of effects before the consumer arrives at a purchase. It paints a picture of a rational consumer that needs to become aware of a brand, learn about the unique benefits, consider it, have a purchase intention and ultimately arrives at the check-out.

More recently, the so-called dual-process theories from psychology started to prevail. Made Famous by Daniel Kahneman in ‘Thinking Fast and Slow’, he divides the consumer mind in two vastly different systems. There’s the quick and intuitive system 1, which is driven by saliency, association and emotion. Its rational counterpart is called system 2, which requires a lot more resources to operate. People often walk through live on autopilot, entrusting system one what the majority of decision making.

Here’s where the problem with many intermediate campaign metrics such as awareness and message comprehension come in. They tap right into system 2, whereas most choices are made by system one. Hence why they correlate so weakly with actual campaign effects on brand growth.

Neuromarketing methods such as EEG and fMRI do a better job at capturing the campaign effects that truly drive brand growth. Expectedly, their predictive power of purchase behavior is superior in many product categories, especially those that are driven by system 1. For instance, in a joint study of CBS and Nielson, they found that EEG explained 62% of in-store sales, versus 24% for survey (which is still impressive!). Adding additional neuro technologies such as GSR increased the predictive power of EEG to 77%.

Copy testing with EEG

An additional benefit of EEG is that it records brain data at an extremely fast rate. This allows you to measure the levels of emotion, attention and memory at a second to second basis, giving you clear insights on what works and what doesn’t.

Neuromarketing isn’t merely a means to predict advertising performance, but actually increasing it. By testing an ad with neuromarketing, you get practical feedback on how to optimize your ad to increase its selling potential.

How much neuro is there in Millward Brown Link? A view on facial coding

Now, let’s turn back to Millward Brown Link. It would be very premature for to conclude it would only tap into the rational system 2.

As the methodology is quite broad, it covers many aspects of the brand and message. While many of its underlying methods seem to be explicit, it utilizes a neuro-like technology in the form of Facial Coding. Unfortunately, the predictive power of facial coding with regards to advertising is low. With only 9%, it’s actually below survey. For this reason, I would only recommend facial coding for qualitative research.

Neuromarketing as an alternative to Millward Brown Link

Personally, I like to keep market research lean. I regard it a true waste of time and money to measure something that never needed to be measured.

As we’ve seen in this blog, many advertising performance metrics don’t say much about anything.

Granted, neuro won’t tell you whether people ‘get’ a message. Nor will it tell you whether a brand has passed into the awareness stage. But the entire point is, it doesn’t need to. It will simply tell you whether an ad will help your business grow.

Isn’t that what market research should be all about?