Validating insights that fall into social desirability bias
Consumer insights often present topics that deal with socially undesirable situations. These statements can tap into areas such as social preconceptions, the need to fit into a group or social image desires. In these cases, such insights may fail to validate, not necessarily because they don’t resonate with consumers or are not true, but rather because social desirability bias leads to low identification with the insight. In other words, low scores can be caused by the reluctance to admit embarrassing or socially unaccepted situations.
As this was the case in specific innovation platforms (e.g. consumption of alcohol for its ‘buzz’), we decided to collaborate on a taboo-focused research, to better understand how to counter this bias.
Together with Heineken, we set up an insight validation study aiming at testing different formulations of insights previously identified as ‘taboo insights’ which had failed validation. Three test cells were set-up:
- Original taboo insight (control cell)
Example: ‘When in a disco, I want to fit in as one of the crowd…‘
- Positioning the insight in a positive emotional context
Example: ‘When going out I enjoy that feeling of being part of a group…‘
- Presenting the insight in the third person, with even more context, through storytelling
Example: ‘When going out we’re usually just sharing good times and catching up….‘
The results were quite clear. Taboo effect can be minimised: whilst performing equally on all other KPIs, the Consumer Relevance KPI increases when positive (test cell 2) or storytelling (test cell 3) formulations are used.
A second confirmed hypothesis was that asking the relevance question in the third-person perspective (‘Do you think others would personally identify with this insight?’) would lead to higher Relevance scores, specifically in these insights. Since then, this question has been integrated in our studies, allowing the identification of potential taboo insights, in the case where considerable differences are observed versus the first-person identification question results.
Taking everything into account, taboo insights can and should be tested. Having this in mind, it is important to minimise the social desirability bias impact, by re-formulating the insights in a positive manner or by providing a storytelling context – naturally remaining faithful to the core insight idea.
Want to find out more about how insight validation can help you unlock the insights with the greatest innovation, activation or branding potential?
Download our free paper on Digging for Gold: How to select those consumer insights that will change your business.
Or our From Validating to Understanding: Why measuring insight strength is not sufficient.