Through a glass, darkly: Branded glassware
Branded pint glasses are now ubiquitous in pubs and bars across the country. Brands have spent millions ensuring consumers drink from the correct glass whether that’s a simple branded glass or the increasing popular chalice. But what impact does all this investment have? Richard Shotton, Head of Insight at ZenithOptimedia, finds out.
Consumers’ claim that branded glassware has a minimal impact on lager’s taste. We surveyed 119 drinkers and only 8% said that drinking out of a branded glass significantly improved the taste. So is the investment wasted? Insights from behavioural economics suggest that there might be good reasons to invest in glassware.
The theory of behavioural economics provides robust insights into how consumers actually behave rather than how they claim to behave – reminiscent of David Ogilvy’s belief that “consumers don’t think what they feel, don’t say what they think and don’t do what they say”.
Since many decisions are made sub-consciously consumers don’t always have full introspective insight into their motivations. Whilst consumers insist that their preferences are mainly driven by the intrinsic attributes of a food or drink, an increasing body of evidence suggests that this isn’t always the case.
Wansink, Professor of consumer behaviour at Cornell University, ran an experiment amongst 175 students at Urbana, Illinois which demonstrated the power of the serve in determining taste perceptions. He served students brownies in three separate situations: on a napkin, a paper plate and a china plate. Despite the brownies being from the same batch there was a huge difference in both taste ratings and willingness to pay. Consumers were willing to pay $1.27 for the brownie on a china plate but only 53c when they were served it on a napkin. Wansink’s explanation is that our taste perceptions are biased by our expectations: if you think the food or drink is going to taste good or bad, it probably will. This tendency has since been termed ‘expectation assimilation’.
Wansink’s work suggests that serving lager out of a high quality branded glass should boost taste perceptions. We have run taste tests amongst hundreds of consumers to see if this is the case. As part of our research we served separate groups of consumers’ lager in plastic glasses, unbranded glasses and branded glasses and then compared the results. Overall, the results have shown that expectation assimilation occurs in lager just as it does in food. However, the results have shown considerable nuances. The impact on taste perceptions is much more marked amongst smaller brands than larger ones. Our hypothesis is that if consumers have had thousands of experiences of a brand, whether that is through exposure to ads, sponsorships or tastings, then the impact of a good serve makes up a smaller proportion of the overall knowledge and therefore the impact is reduced. The second startling result is that in certain cases drinking out of a branded glass can actually damage taste perceptions. Our hypothesis is that if consumers have a negative perception of a brand then serving the lager in branded glass makes those negative connotations more salient, thereby impacting the taste.
Our research suggests that expectation assimilation might not always justify an investment in branded glassware. However, there are other biases identified by behavioural economics which may justify the investment. One such bias is social proof. This is the idea that consumers choose brands not just on the inherent characteristics but that they are significantly influenced, consciously or subconsciously, by the behaviour of others. Branded glassware can tap into this by demonstrating the popularity of a brand.
A combination of social proof and expectation assimilation suggests that branded glassware has a significant impact on consumer behaviour. However, the scale of the impact varies from brand to brand. Companies may want to take a nuanced approach to branded glassware and only make the significant investment on brands which will benefit most.
