Thinking about emotions
Do you remember your last vacation? You felt so happy, it was great… but do you also remember the rain that made you stay in the hotel for a full day, your stomach that did not react properly on the local food or the noisy neighbours that kept you awake at night? This story clearly illustrates one of the difficulties we are facing today when measuring emotions in research. Consumers post-rationalize their own behavior. But even worse, they think less than they think.
Recent neuropsychological work teaches us that our brain has two parts: a reflective and rational route, which is involved when we are really ‘thinking’ and an automatic route, which makes very quick effortless decisions based on the emotional evaluation of past actions (LeDoux 1996). Reality is, that most of our behavior is determined by this automatic emotional brain. Hence it is crucial to measure emotions in a good way.
Traditionally, qualitative research had been thé method to grasp the emotional consumers
Over the last years we have been bombarded with alternative and more quantitative approaches to measure emotions. Historically, we measured emotions in quantitative research by direct questioning (‘which of the following emotions do you experience’). It takes little explanation to understand that this is a very rational way of detecting feelings. Also, one can wonder to what extent consumers are aware of all their emotions and if they are even able to answer this question directly. Therefore, we recently experimented with three more implicit quantitative ways (through ‘dual tasks’, ‘indication under time pressure’ and a ‘picture collage’) of getting to emotional depth. The results were tested in an idea screener with quite surprising results! First of all, assessing emotions in quantitative research really proved to have added value. Whereas the concepts scored very similar on traditional screener KPIs, the emotional measures were very different. The difference was not only between the different concepts, but more importantly also between different target groups. Our implicit measures were also able to detect social desirable negative emotions that were not revealed through the direct quantitative measure or even in qualitative research.
There is more than measurement alone
One of the key questions that is often forgotten in this debate is on what emotions you need to measure. Starting with the psychologists Carroll Izard and Paul Ekman, many theorists have posited that there are a fixed number of primary, or basic, emotions (e.g. Ekman, 2003). In this perspective, 6 basic emotions (happiness, surprise, fear, anger, sadness, disgust) serve as irreducible emotional states, each with its own distinct set of feelings.
These six basic emotions are seen as the fundamental building blocks of emotional experience, in that any type of experienced emotion can be reduced to one or a mix of these basic emotional components. In support of the universality of these basic emotions, studies conducted in countries all over the world have shown that these basic emotions are recognized universally.
So measuring those six basis emotions is a good start. Or not?
In my opinion, one very crucial aspect is missing in this discussion namely the action ability of the results. What does it mean if your new product leads to 20% anger, 10% sadness and 50% happiness? Detecting negative emotions is only useful if one can actually do something about this. For example, early in innovation process, one might be able to correct for negative feelings but once the product is launched, this might be a lot harder. Happiness is also a very broad category. Excited consumers might react quite different than relaxed consumers… Experiencing an ‘active’ positive emotion might stimulate consumers for example to actually buy a product. We need to look at the correlation between emotions and key performance indicators in marketing.
And this is exactly what we have been doing at our ForwaR&D Lab in the last months. A recent meta-analysis across studies where we have measured emotions, helped us to identify the emotions with the biggest impact on marketing KPIs. For example, we found out that there is a big correlation between experiencing positive emotions and the net promoter score. Striking enough, the impact of negative emotions on word of mouth is much smaller. Learning about the correlations between emotions and marketing KPIs provides us with the necessary context needed to do something with the results. After all a piece of research is only as valuable as its usefulness afterwards.
To end take a look at this recently published research we did with AirFrance KLM, the results show that measurement of emotions is definitely not only restricted to qualitative research. Including emotional and implicit measures in quantitative testing can help us detect emotional differences between groups and help us complete the picture that we obtain qualitatively. It also helps us reveal emotions that consumers may not be aware of or that consumers find hard to admit.