Find Optimal Store Locations (Mean Store)

Overview

The Find Optimal Store Locations analysis creates a centroid in the mean geographic center of your customer points. Additionally, you can use clustering if more than one mean location is desired.

This centroid can be calculated by either of the following:

• Number of customers (taking only geographic location into consideration)
• Weighted value (such as sales or visits)

Clustering using the K-means algorithm

Creating many Find Optimal Store Locations uses a clustering algorithm called K-means. The K-means approach finds geographic concentrations in a point database and determines their center points. After identifying a cluster partition, the process continues iteratively until all points are associated with the closest mean center.

Input Prerequisites

You must have a point layer that can contain volumetric data.

Output Example

Example of a single Find Optimal Store Locations using locations

This example shows a single Find Optimal Store Locations. The location is based on the geographic locations of the surrounding customer points.

More about calculating the centroid by number of customers

When the centroid is calculated by the number of customers, each customer point has an equal value. Since the centroid represents a balance point between all customers, it will be located roughly in the center of the customers. If customers are more densely populated on one side, the centroid will be pulled in that direction.

Real-world example: Locating a new store using existing customer residences

Suppose you want to expand your chain of sporting equipment stores into a new market area. Your existing customer profile shows that you sell to a limited demographic segment—high-income, well-educated people who play golf.

To begin, you might purchase a mailing list of households with similar demographics in the expansion market, geocode them using the Customer Setup wizard, then calculate the centroid by the number of prospects. The resulting centroid would be a good place to start looking for a new location.

Example of a clustered Find Optimal Store Locations using locations

This example shows Find Optimal Store Locations clustered into four different points. The Thiessen polygons in the second picture below give a visual representation of equally distributed areas from each clustered point.

Example of a clustered Find Optimal Store Locations by weighted values

This example below shows the difference between Find Optimal Store Locations using simple clustered geographic locations and using the locations with a weighted factor. In this instance, the weighted factor is a dollar amount identified with each customer point. The black arrows demonstrate how the weighted factors move the clustered mean centers of the customer points by "pulling" them toward the larger features. The larger features (green circles) are simply a graphic representation of their respective volumetric attribute (sales volume). The larger the sales figure, the larger the circle. This is not part of the Find Optimal Store Locations tool but is a function within the ArcGIS layer symbology used to better demonstrate the scenario.

More about calculating the centroid by weighted value

A centroid calculated by a weighted value considers each customer to have an individual value. The centroid is not created in the center of all customers but in the center of the customers who most satisfy the value you have weighted.

Suppose you want to calculate the centroid by customer sales. The location of a customer who has spent $100 at your store will be counted 100 times more than a customer spending only one dollar. When the centroid is calculated, this weighting pulls the centroid toward the more important points.

Notice the location of the centroid in the graphic when calculated by a weighted value, in this case, sales. No longer is the centroid in the center of the customer points, but it has shifted toward the customers who spend more money.

Real-world example: Locating a new store by weighted value

Suppose the building leases for two of your bank branches expire at the end of the year. You want to know if the leases are worth renewing. Using each branch's customer set, you calculate a centroid weighted by the number of visits or total deposits. You can compare the resulting centroids with where the actual branches are. If a branch is fairly far from a centroid, you might consider looking at other properties instead of renewing the leases.

The following are other examples of how businesses use centroids:

• A high-end men's clothing store loses its lease at a longtime location. They use the customer database weighted by total sales per year as the base in its search for a new location.
• A quick auto oil change franchise uses the business addresses of existing customers to find an optimum location for a new operation to serve customers near their workplaces.
• A bank derives a weighted centroid for each product type (home equity loans, auto loans, CDs, investments, and so on) and assigns the branch closest to each centroid to specialize in that product.