Social network analysis has been a topic of regular interest in the marketing discipline. Previous studies have largely focused on similarities in product/brand choice decisions within the same social network, often in the context of product innovation adoption. Not much is known, however, about the importance of social network effects once customers have been acquired. using the customer base of a telecommunications company, our study analyses network autocorrelation in the distribution of customer-level revenue within a social network. Our results indicate a significant and substantial degree of positive network autocorrelation in customer-level revenue. High (low) revenue customers therefore tend to be primarily related to other high (low) revenue clients. Furthermore, we show that approximating communicative proximity by spatial proximity leads to a substantial underestimation of these effects.
HAENLEIN M., (2011), "A social network analysis of customer-level revenue distribution", MARKETING LETTERS, Vol.22, Issue 1, pp 15-29, 15 p.
Among sociologists it has been known for a long time that people generally prefer to interact with others who are similar to themselves. Looking at your own social network, you will for example realize that your close friends have, on average, the same type of education, the same type of job and the same age as yourself. This phenomenon is usually referred to under the term "homophily". Homophily occurs because people with the same characteristics tend to be "on the same wavelength", which increases their chances to become friends. Also, people usually make friends within their immediate environment and certain types of environments tend to attract the same type of people. For example, some of your close friends may stem from your time at ESCP Europe, which by definition makes them similar to yourself in terms of educational background and age.
Interestingly, the very same characteristics for which sociologists have observed homophily have been shown by marketers to influence consumption, purchase and ultimately customer value. For example the idea that customers with a certain educational or professional background might be particularly attractive for certain types of industries is the underlying principle for many types of socio-demographic segmentation. Also, it has been widely accepted by academics and practitioners alike that age influences factors like loyalty and repeat purchase – see for example the research of my colleague Raphaelle Lambert-Pandraud in this domain.
Combining these two observations makes it likely that the distribution of value within a social network should not be random. Instead, it is much more probable that "good" customers (in terms of either revenue potential, profitability or customer lifetime value) have friends that are "good" customers themselves and that "bad" customers are primarily related to other "bad" clients. In a previous research project, the issue was investigated using the customer database of a cellular service provider. Using call detail records (i.e., the list of numbers called by a certain customer within a certain period of time), it was possible to reconstruct the social network of several thousand customers. Combining this information with revenue data for each customer made it possible to generate a dataset that allows to look into customer value and social relationships simultaneously.
Indeed, it turns out that customers tend to cluster together in buckets of "good" and "bad" clients – a result that has substantial implications for the customer acquisition process. Value clustering implies that one attractive way to acquire attractive clients might be to pass through your existing high-value customers. Referral reward programs can, for example, be used to convince existing customers to bring their friends to the company and if these existing customers are valuable themselves, their friends will likely be similar. In business as in private life, being introduced to a stranger bye one of your best friends might be the most promising way to meet new good friends.