Consumer demographics, product attributes, and purchase history has long been at the heart of marketing communications. That said, with the rise of big data and real-time consumer information, more attention has been paid recently to context-based marketing, in which consumers are shown different types of advertisements depending on environmental factors and their activities at a given moment. The success of this type of marketing therefore depends on how consumers respond in various situations. Hence, it is crucial to understand the effects of contextual factors when it comes to designing context-based marketing.

To that end, a recent study by Kohei Kawaguchi, Kosuke Uetake, Yasutora Wanatabe looked at how behavioural context factors moderate or facilitate the effectiveness of marketing communications, especially in terms of product recommendations. In particular, they examined the effect of time and crowd pressures on making purchasing decisions – two prominent examples of contextual factors in the consumer behaviour literature.

The authors faced several obstacles regarding measuring the effectiveness of product recommendations in context, since the joint distribution of consumer choice, time, and crowd pressures is rarely found in a real-world setting. These were overcome, however, by using data from a previous company’s experiment which focused on the purchasing of beverages from vending machines situated in train stations across the Tokyo metropolitan area. Equipped with facial recognition technology to help make product recommendations, the machines’ recommendations were then changed exogenously. This provided Kawaguchi, Uetake, Wanatabe with well-measured variables of the time and crowd pressures impacting the effectiveness of recommendations.

They discovered that recommendations increased sales of not only recommended products but also spilled over into non-recommended products; implying that marketing managers should take into account both the choice effect of recommendation as well as the spill-over effect when designing product recommendations. More importantly, the study found that time and crowd pressures impact the effectiveness of recommendations. First, the results revealed that time pressure lessens the choice and spill-over effects. In other words, consumers facing time pressure are less likely to buy either recommended or non-recommended products. Second, they showed that crowd pressure weakens the choice effect of recommendation, but the spill-over effect strengthens the effects on non-recommended products. As the authors explain, what this might mean is that “customers are more likely to make a purchase under crowd pressure, but the recommended product may not necessarily be what they want; hence, the spill-over effect arises to the non-recommended products”. With this in mind, marketing managers should pay close attention to tailoring recommendations when the presence of other customers affects the consumer’s decision.

Of course, as Kawaguchi, Uetake, Wanatabe are quick to point out, there are a couple of limitations to their study which offer promising avenues for future research. First, the study does not allow for the distinguishing of different psychological theories behind time and crowd pressures. “Identifying the exact psychological mechanisms underlying the results would provide useful suggestions for the design of product recommendations”, the authors explain. Second, they go on to say that the results do not necessarily provide evidence that time and crowd pressure effects exist in other situations. They agree that, given the dramatic increase in context-based marketing, investigating other scenarios and boundary conditions would be both and interesting and important research question.