±±¾©Èü³µpk10

Emotions tell us which bits of information to engage with and which to ignore, thereby directing our attention, enhancing our memory, and influencing our behaviour.  Very few remain to be convinced of the need to generate an emotional response – the challenge is to understand how you generate it, and whether it results in positive business impact for the brand.  
 
New techniques in research allow us to understand emotional response, helping us to assess both the overall emotive power of an ad, and the way this varies and builds as an ad progresses. Such techniques have been proven to discriminate very well in terms of predicting advertising effectiveness, and to identify opportunities for optimisation. Conventional research, properly applied – i.e. asking appropriate questions, to appropriate people, and reviewing results in a way which is appropriate to an ad’s objectives – can and does help advertising work harder for brands. However, neuroscience gives us the opportunity to get an additional layer of understanding and insight.
 
Critically, neuroscience is the study of the nervous system – not just the brain, but the network of neurons all over the body. When an emotional response occurs within the brain, it generates an automatic reaction in our nervous system that leads to changes in certain body signals.  These physical responses include changes in skin conductivity, heart rate and breathing.  Biometrics can measure these signals, and thus is the best way for us to measure emotional response to advertising, since it allows us to assess directly the activity in the emotional centres of the brain. By equipping respondents with a chest belt and finger sensor, we can collect information on these responses in a relatively unobtrusive way – respondents can forget they are wearing them much more readily than they can forget electrodes on their head.
 
Sounds good in theory – but what do you get from it, and what difference does it make to brands? We have been running extensive pilots, which have helped us develop a strong understanding of the insight biometrics can bring, when combined with more traditional research.
 
The key output from biometric research is a trace, indicating changes in emotional engagement during the course of an ad, and shown in the context of what we know to be a minimum threshold for active emotional engagement. This trace looks at first sight like the kind of traces you get from quantitative ad tests, derived from respondents moving an indicator on-screen to reflect positive or negative thoughts or feelings. However, relative to these "conscious" traces, biometric traces tend to show much more pronounced, immediate reactions. Importantly, these signals tell us a lot about how the person is reacting.  It is indicative of whether they are attending to the ad or ignoring it, and the intensity of their response. Relating these reactions to a playback of the ad helps us to understand how the ad is working – for example, is there engagement at the point where the brand is visible, or the key insight is delivered?
 
Pilot research with food clients has shown – unsurprisingly – that the level of engagement with shots of the product is a key predictor of effectiveness. An ad for a new food product, known to have underperformed in-market, was seen to have peaks in engagement, but not when the food was shown. An ad for a new alcoholic beverage saw engagement peak when the product was poured into a distinctively branded glass. That brand has been highly successful in stealing share from the established market leader.
 
Indeed, the extent to which engagement peaks is a key part of the output -  we call it the Max Value, and it is highly correlated with an ad’s capacity to cut through and be noticed – a pattern observed through comparisons to conventional pre-testing, and to second-by-second set-top box readings. Ads which generate that big emotional pay-off grab the attention, and stop the viewer from flicking away.
 
Similarly, Biometric Build – essentially, how the ad sustains emotional engagement – is highly correlated with audience behaviour. Specifically, it predicts buzz potential or viralability. The traces for two famously viral recent ads are shown here – emotional engagement both peaking and sustaining.
 
What biometrics won’t tell you is whether that noticeability or viralability will do anything for the brand, or why. This is the key reason why we believe that biometrics are most compelling in combination with conventional research. Biometrics will tell you:
 
·        Whether key parts of an ad are working optimally – e.g. where brand benefits are presented or implied.

·        The key parts of an ad to keep – in creative development or when seeking to create cut-downs


There are instances of ads not generating particularly strong emotional response, but nevertheless being known to be successful in-market. Typically, such ads tend to function by reinforcing what is already known or felt about an established brand – such ads perhaps don’t need to work  quite as hard as other ads to generate that emotional pay-off. However, even established brands should not rely too much on work like this – it will do a holding job, but you will need something more powerful in the long run.
 
The number one key finding from biometrics in ad research to date has been, reassuringly, to back up what we already knew – that the power of the creative is the most important driver of advertising success – a power that can be felt.

Keith Glasspoole is the Deputy COO of Ipsos ASI