Much of the world of business analytics is focused around the management, understanding and leveraging of customer relationships. And for good reason, acquiring a new customer is at least five times the cost, on average, of keeping an existing one. It also means that marketing will be primarily to existing customers, not only as a customer retention strategy, but also to up-sell and cross-sell. This has, by in large, driven the proliferation of database marketing and customer relationship management software. Coupling this with declining data storage costs over the last couple of decades have helped drive the exponential growth in the capture, storage and analysis of "big data". In summary, the research and development and subsequent innovation that continues to develop in the field of business analytics over the last couple of decades re-enforces customer level analytics as the primary focus.
In contrast, market analytics looks at the characteristics of the aggregated market and provides the supply (competitor strength) and demand (consumer spend) foundation for capital expenditure decisions including site evaluation and selection. Since market analytics relies on an accurate S&D measurement, it requires an accurate geographic trade area definition. Something not required by customer analytics. Even marketing campaigns only require a general estimate of consumer geography.
Not surprisingly, technological advancements relating to market analysis haven't kept pace with customer level analytics. Most advancements over the last five decades have been to operationalize various trade area types within GIS software. It's put market analysis into the hands of the qualified and unqualified alike. And in turn, its turned market analysis into a wild west culture, where trade area simplicity and development speed generally trumps the accuracy of more complex scientific methods.
The first problem is that simple trade area definitions can't explain and quantify the complex nature of consumer purchase behavior where mobility means that consumer purchases are typically made across a wide variety of trade areas. Technically this means that demand rarely equals supply within a trade area, and the subsequent disequilibrium (S≠D) creates a major accuracy problem. Figure 1 shows the overlapping radii trade areas that are necessary to account for consumer purchase behavior. This creates analytical chaos making it virtually impossible to accurately calculate useful and comparable trade area metrics using conventional trade area definitions. Exceed uses a system of multiple trade area layers to solve the mobile consumer problem and calculate accurate S&D metrics. This allows us to compare and rank the market potential of individual trade areas.
The second problem is that the actual trade area definition largely defines the analytical solution. Figure 2 shows that the consumer spend by trade area (GTA region) estimated with the Exceed trade area model (green points) creates vastly different consumer spend estimates than a radii solution sized by the ICSC (blue points). It reveals how important trade area definition is to market size estimates and why the Exceed trade area approach is entirely driven by data and not by user defined assumptions.
The third problem is that the size of the trade also has a significant impact on the overall analytical solution. Figure 3 compares the consumer spend by trade area (GTA region) estimated with the Exceed trade area model. The green points represent the actual model results, and the blue points represent modeled results plus 1 city block. It shows how sensitive trade area size assumptions are to market size estimates and why the Exceed trade area approach is entirely driven by data.
The Exceed solution is a proprietary data driven spatial break-point model to estimate the size and shape of each trade area. It uses a layered algorithm to account for mobile consumers that purchase goods in multiple trade areas.