How To Find Your Most Valuable Service Outlets: Measuring Influence Using Network Analysis
Posted: 25 Apr 2019 Last revised: 26 Mar 2020
Date Written: April 4, 2019
Consider a network of stores operating under the same brand, for example, a chain of coffee shops (e.g., Starbucks) or banks (e.g., Wells Fargo). Improving quality and increasing sales at one store may have different impact on the other stores, from negative impact as a result of potential cannibalization to positive impact from improved brand reputation and knowledge spillover. In this paper, we adapt an approach from spatial econometrics utilizing a weight matrix and extend it to allow for identification of both positive and negative peer effects between stores. This semi-parametric methodology allows us to handle a large-scale network and provides causal estimates for the network effects between all store pairs. We use three distinct sets of instrumental variables that are based on location-specific (weather and nearby sports events) and store-specific (deviations in daily staffing) attributes. By applying our method to a large dataset from a major national restaurant chain in the US, we show that when considering only the direct influence, the total influence of each store on the network sales is substantial: a $1 sales increase in one store can generate a total of up to $3.87 in additional sales across all other peer stores in the network, while it can also reduce the sales in the network by a total of $3.12, with the average influence of $0.55. Our influence estimates provide valuable insights for the brand; for example, it can prioritize stores with the highest influences for ownership or improvement to optimize return on investment.
Keywords: Network Effect; Service Network; Causal Inference; Restaurant Operations
Suggested Citation: Suggested Citation