Please login to the form below

Not currently logged in

Death by Numbers

The iconic ads that could have been

Death by numbers

This article was inspired by something we have probably all experienced. Creative development research led to a bold creative concept being ditched in favour of a more conservative, rational approach. Traditional metrics were being used to evaluate the idea. But many of us were left feeling that the original idea, despite ticking fewer boxes, would have done a better job for the brand by getting noticed and building equity. We were sure there was a better way. And there is.

Advertising and emotion
It's hardly news that much of the most effective advertising works through emotion.

Analysis of The Institute of Practitioners in Advertising's (IPA) effectiveness awards shows clearly that campaigns built around emotion-led strategies are more likely to be proven effective.

It's tempting to think pharma is different - more rational because we're often targeting physicians to talk about an inherently 'serious' subject. Not so. Take the example of this campaign for Sativex, a drug used in the treatment of Multiple Sclerosis. The obvious solution to the brief was an ingredient story - Sativex is derived from cannabis. But the team set out to forge a more powerful emotional connection with physicians. The trigger here is independence. It's what MS steals. The campaign won a Gold Lion at Cannes.

It's tempting to think of creative awards as irrelevant to effectiveness. But IPA's analysis shows creatively awarded campaigns are disproportionately likely to be effective. They are more efficient at getting noticed and subsequently lodging in long-term memory. That's imperative in pharma, just like other categories. Call it impact, fame or creativity; these campaigns work.

Why does emotion influence our actions?
Creative people have long understood the importance of emotion. The rest of us are playing catch-up. Recently, we have been helped along by a plethora of books and articles drawing on evidence from psychology, behavioural science, neuroscience and even evolutionary biology.

The most influential of these is Nobel Prize winner Daniel Kahneman. In his seminal work 'Thinking Fast and Slow' he describes two alternative modes for processing information. System 1 is the fast, instinctive, emotional system we inherited from our cave man ancestors. System 2 is the slower, more arduous system characterised by intellectual arguments (and reading articles like this). Kahneman shows most everyday choices are made in System 1. It's quick and efficient. In a crisis it could save your life. In daily life it saves energy. Most brand choices happen through heuristics - System 1 'short-cuts' - designed to save mental energy. We post-rationalise them in System 2.
By way of an analogy, Kahneman says System 2 feels like the Oval Office, making key decisions, but in fact it's the press office, simply presenting those decisions to the outside world. Rory Sutherland's analogy is more colourful - think of a man riding an elephant. The System 2 man believes he's in control, but in reality the System 1 elephant goes where it wants.

Where does this leave research?
This has profound implications for how we use research to develop and evaluate advertising.

Historically, advertising evaluation has used tools based on the sequential models of the 1920s. Tracking studies and quantitative pre-tests measure shifts in awareness, attitudes and purchase intention to establish a causal link between marketing activity and sales.

The sequence is Think > Act > Feel. But given what we now know about psychology, this should be turned on its head. It's more like Feel > Act > Think.
Even our sensitive qualitative methods tend to ask what the idea communicates. A typical brand proposition contains the single-minded proposition with reasons to believe. These principles are embedded in our thinking.

We also need to measure the subconscious and the emotional. But how? Let's look at the state of play. To this end we have (a) looked at the technologies and approaches now available and (b) set up an experiment to see if an implicit approach to 'testing' advertising might yield different results from traditional explicit measures.

...much of the most effective advertising works through emotion

a) We have a range of tools:

1. Self reporting
We can ask people how they feel - either verbally or visually. Verbal self-reports are scaled questions or word-boards describing emotions - 'pick the word, which best describes how you feel'. Visual self-reports use visual signifiers, pictures or cartoons to measure and describe emotions in a similar way.

They're quick, simple and they fit easily into existing survey protocols. However, they rely on respondents being aware of their feelings and being happy to describe them. Are doctors always happy to discuss the softer associations when they're processing advertisements?

2. Implicit association tests
The implicit association test (IAT) taps into our implicit, subconscious (System 1) associations. It first came to prominence in psychological research to investigate racial bias, which people wouldn't admit to explicitly.

IAT asks people to swiftly pair two concepts together. You might be asked to pair 'women' with 'maths' and 'women' with 'liberal arts'. IAT applies the principle that closely associated ideas are paired more quickly. We measure the time you take to make pairings and draw conclusions.

As an illustration try this simple test. Look at the top line and read out the font colour (not the word) from left to right. Now do the same for the bottom line. Most people find it much quicker to read the second one because the visual and verbal associations match - they're more closely associated.

3. Biometric tests
We can measure how our physiology responds to stimulus, thus short-circuiting any tendency to post-rationalise, lie or delude ourselves.

We have tools like the EDA (electrodermal activity) watch, which measures skin conductance, heart rate and other biometric measures. Sympathetic activation (the fight or flight system) increases when we experience emotional responses like excitement or stress. Interestingly the skin is innervated purely by the sympathetic nervous system. Monitoring these signs, while applying a stimulus (say, watching an ad) shows how we are responding emotionally. The down side is that we can't necessarily tell whether that response is positive or negative. In EDA or heart rate terms, joy may look a lot like disgust.

Facial coding is another popular biometric tool. The human face is a window into our emotions. We use this instinctively. A raised eyebrow, a frown or a half-smile can tell us more than words. Psychologists and anthropologists tell us there are a finite number of universal, recognisable facial responses. These indicate our emotional response.

In the early days of FACS (facial action coding systems) trained researchers would painstakingly code video footage of respondents' facial responses. Now there is automated FACS coding using footage from webcams.

Eye tracking is another technique gaining in popularity. This doesn't measure emotion per se. But it can be used in conjunction with other methods to help explain how we're responding - for example what exactly are we paying attention to, when the emotional peaks occur?

4. Neuroscience
Brain scanning is increasingly accessible, affordable and scalable. Portable scanners measure electrical currents, associated with firing neurons, using sensors in a skullcap. This way, we can see which areas of the brain are stimulated in a moment-by-moment trace. New unobtrusive dry caps are available, which no longer have gel filled electrodes. However as the technology has become more portable it has also become less discriminating so there is a trade-off between accessibility and quality of information. It is important when using this type of approach to have a clear hypothesis of what you want to evaluate and unambiguous criteria for success. In this area, it is especially important to work with the experts.

A good example of this was Dutch mortgage broker De Hypotheker. The company used FMRI brain scanning to understand the different responses to two approaches - one based on heightening customers' financial anxiety and then providing a solution and the other highlighting key advantages over the competition. The 'financial anxiety' route provoked a better response and it catapulted the company to brand leadership.

It's increasingly clear that subconcious processing is fundamental to communications effectiveness

b) Explicit or Implicit: does it make a difference?
But will these new methods actually help us discover things we couldn't have found out using traditional research, done sensitively? In the past, we asked people about their feelings - but we did it explicitly. Our hypothesis is that we can get closer to the true sources of motivation if we look implicitly.

We constructed an experiment to see whether explicit and implicit techniques applied to the same stimulus would yield different results.

And the short answer is yes, it did!

We exposed two healthcare brands' award-winning press advertisements to samples of fifty GPs. The adverts were chosen to represent different approaches to the same category.

We asked the GPs to evaluate each advert in three ways:

  1. Standard rational measures (eg persuasion, impact and likeability)
  2. Verbal and visual emotional measures (an emotion questionnaire, word board and visual scaling using photos of universal emotions) 
  3. Measuring the strength of their implicit associations.

The results showed clear differences between implicit and explicit results, especially for the more emotionally challenging executions.

The most emotional treatment showed a baby crying. This had polarised opinion among the brand team so it was hoped the research would resolve the argument. Measuring the advert's performance explicitly indicated that it performed around benchmark but that it conveyed sadness, worry and unease. No positive emotions came through at all. This supported the ad's critics.

However, the picture painted when it was researched implicitly was quite different. Yes, the advert did create unease but it also conveyed joy, motivation and liberation. Here was ammunition for those backing the ad to succeed.

Using implicit and explicit measures would give us a fuller understanding of advertising performance.

Effective new methods
It's increasingly clear that subconscious processing is fundamental to communications effectiveness. We still mostly score our ideas rationally even though we know they need to connect emotionally. This isn't just an academic issue. We risk ending up with 'safe' ideas that don't challenge our audience. We're too often choosing winners in a beauty parade. How many great concepts have been wrongly consigned to the rubbish heap?

Given all we've learned, there is surely a better way, using implicit association and selecting from an emerging toolkit of new methods. If we use these judiciously and take the right advice from the experts, we can better understand what's most likely to be effective.

Vivienne Farr is a director at Narrative Health and Andrian Zambardino is director of planning at Langland

23rd February 2016

Vivienne Farr is a director at Narrative Health and Andrian Zambardino is director of planning at Langland

23rd February 2016

From: Marketing



Featured jobs

Subscribe to our email news alerts


Add my company
infill healthcare communication

infill is a dynamic and innovative, international healthcare agency with >25 years' experience in delivering healthcare communication, medical education and...