Anyone who has accidentally received their neighbor’s package knows the importance of pinpoint accuracy when it comes to customer address information. For business-to-business trade with large shipments worth millions of dollars, the importance of accuracy grows exponentially. While improvements in address information validation and navigation systems have reduced the challenge, many turn to geospatial data to improve location accuracy, visualize proximity, and provide an additional data point for insight.
In this post, we’ll discuss how geospatial data can be incorporated with customer data mastering to increase the accuracy of customer information, as well as how a multinational food-products corporation used geospatial data to improve sales operations.
What Is Geospatial Data?
Geospatial data is information that describes objects, events, or other features with a location on or near the surface of the earth typically as points, lines, polygons with corresponding latitude and longitude coordinates.
Aside from name, address is the most common attribute used to match and identify unique customer accounts. When comparing the written addresses of two records, textual information is used to assess whether the locations are the same, not how they are located in space with respect to one another. Because of this, addresses can be misleading: two sites with different street addresses, and even zip codes, may seem like they are not similar, but they may actually be located directly next to each other. Incorporating geospatial data is important for customer data mastering, as the distance and similarity can be considered, providing a more accurate and true way to compare site locations and the corresponding customer records.
Geospatial Data Adds Important Context to Data Mastering
Taking geospatial data into consideration during a data mastering initiative provides a deeper level of granularity and an additional data point of reference that improves matching. With geospatial data, objects in space can be represented as points, lines, and polygons at an exact location. Having this information can allow for increased matching accuracy, by adding both distance and similarity metrics, as well as visualizations.
There are multiple ways that distance and similarity can be calculated. Both shape and standard distance can be taken into account, providing a metric to compare how similar two geographic objects are, which allows for a more complete view of matching based on site location.
Having and knowing geospatial data is one thing, but being able to see this representation is another. Visualizing this geospatial information allows you to see the shapes and similarities, making mastering clearer and more obvious. You can see overlap between shapes, points, and lines, or a mix. Records that may not have seemed to be the same entity with just textual information can be determined to be a match by viewing their proximity and similarity to one another in a map visualization.
The image below actually shows what this type of visualization looks like in Tamr’s UI. This example shows an obvious overlap between a polygon and a point, which you can determine represent the same entity.
Case Study: How a Multinational Food-Products Corporation Uses Geospatial Data
A large, multinational food-products corporation was struggling to create a single view of its trade customers: the many convenience stores and supermarkets which sold its products. The company began by evaluating its customer base in Mexico with Tamr. The company’s household-name products were a staple at most small convenience stores throughout the country. With tens of thousands of business customers located in big cities and small towns dotted throughout the country, the multinational corporation required a robust distribution network, as well as data accuracy to streamline representative and delivery coverage.
Sales data silos, stale address information, and errors in customer contact information were challenging for the business:
- Multiple deliveries to the same locations were common due to the inability to cluster delivery locations
- The business was unable to carry out product portfolio sales analysis by region to identify which products were popular as well as opportunities for cross-sell.
- The business had trouble identifying which stores were active and which needed to be prioritized
By leveraging Tamr’s customer data mastering and incorporating geospatial data, the multinational corporation was able to create a single view of customer accounts and identify regional clusters of customers. With new visibility on product sales within an area, sales representatives can make data-driven decisions to recommend additional products to customers based on sales in the area and identify other businesses in the vicinity for expansion.
In addition to gaining a unified view of its customers and the products that they purchase, sales operations and distribution could also now be optimized. By seeing exact locations of customer delivery sites along with a complete view of other customer information, the company could make delivery routes more efficient by grouping shipments together and reallocating sales representatives based on customer concentrations.
Considerations When Putting a Customer Mastering Solution Involving Geospatial Data into Practice
While there are many benefits to incorporating geospatial data with customer mastering, we will review some practical aspects to consider when putting a solution into practice for your company.
Enrichment to append geospatial data: While some companies will already capture geospatial data in customer records, many don’t. Most company’s customer data solely includes site location in the form of address information. As part of the enrichment services offering, Tamr can enrich address information to append geospatial information, including longitude and latitude coordinates as well as building shape, providing a more accurate mastering.
Ability to handle scale: With geospatial information comes a large magnitude of data: one site location may be represented by a polygon that is made up of many points, each with their own set of longitude and latitude coordinates. Scale this to multiple datasets with millions of records each- it is clear that your chosen solution needs to be able to computationally handle this type of scale and do so in an automated way. Tamr can do this, and with a proprietary, patented method for geospatial binning, Tamr has proven to execute large-scale deduplication with geospatial data.
Incorporating geospatial data as part of customer data mastering can increase the robustness and pinpoint accuracy of customer records. Geospatial data can also add a visual element that allows analysts and business users to see overlap between shapes, points, and lines, adding another reference point to improve mastering. When integrated as part of data pipelines and data enrichment, geospatial data can be used to drive tangible business value such as streamlining sales operations, including delivery optimization and sales area coverage and analyzing product sales by area to drive upsell.