As industries and their technologies continue to evolve, postal addresses have faded in their ability to deliver customer analytics and relevance; however, this supposed “fade” is also due to a plethora of enterprises not effectively leveraging addresses to enhance their understanding of the customer and maintain their loyalty.
Antoine Emond-Verrault, account manager at Korem, and Scott DiGiacinto, senior enterprise geospatial expert at Korem, joined DBTA’s webinar, “Enhancing Customer Analytics with Advanced Address Management,” to explore the way customer addresses in the U.S. and Canada still offer critical knowledge surrounding customer behaviors, paired with best practices and deployment options to adopt effective address management.
Emond-Verrault explained that address management boils down to a territory approach in terms of customer understanding. Addresses can reveal key information about land, the people on that land, certain risks, and behaviors, as well as the buildings that comprise the territory.
He further argued that mail addresses are more than just addresses, especially today; due to the rise of work-from-home environments around the world, addresses represent a large subsection of a customer’s life. It serves as a unique, tailorable ID that is used to reach customers, which is further attached to demographic and behavior markers.
As mentioned previously, despite an address’ informative nature, many organizations are incorrectly managing them—and it comes with a price tag. For one insurance company in Florida, ill-managed addresses led to both underpriced and overpriced customers, which bore a $250 million price tag. Another company had a significant case of Undeliverable-as-Addressed (UAA), or a percentage of addresses that were not properly set up in a system, which resulted in a $1.4 billion price tag.
Why does this happen? According to Emond-Verrault, it’s because addresses are dynamic; around 16-18% of U.S. residents move yearly, making up 80% of the UAA; and in Canada, 10-15% of postal codes are changed yearly. Furthermore, organizations take address management for granted, making significant room for human error. There is also a lack of distinction in postal and physical addresses, which poses a massive challenge in proper address management.
DiGiacinto continued the conversation by highlighting the most common problems seen in address validation, which include typos, full naming of address conventions, inexact unit descriptors, missing data, mis-fielded data, non-standard character insertion, and incorrect address input. These errors conjure a variety of scenarios that result in lost revenue, lost customers, and lost opportunities.
Geocoding plays a major role in proper address management, according to DiGiacinto. When geocoding is implemented within address management systems, the address input into the system can be corrected according to data sets that represent accurate address information. For example, an address input as “Mesilla Valley Mall 88011” can be corrected to “700 S Telshor Blvd, Las Cruces, NM 88011-4669, P0000G4I6A4F” with precise geocoding and its subsequent data sets, which combines postal and physical data.
Physical data can manifest as a broad selection of data types, explained DiGiacinto. Geocoding by zip code, for example, has been the default for locating property for many years; however, it represents a wide area of coverage, and for dense areas like New York City, it isn't specific enough.
Luckily, spatial location is a rapidly evolving technological space that is becoming more and more accurate. From humble beginnings in street segment interpolation and parcel boundaries that were somewhat accurate (and oftentimes, just lucky), technology has advanced as point-based geocoding, which can identify hyperlocal specifics of a building as an entrance or driveway.
A quantity mentioned in a previous example—P0000G4I6A4F—is representative of a trend in address management that ascertains a variety of property attributes and compiles this data as a consistent, persistent code that matches up differing data sets for a single address.
In a commercial use case, this data can manifest as SIC, when it was built, what and how many companies are in a building, the square footage of the building, its number of stories, etc. For residential use cases, this unique code can offer information such as how many bedrooms and bathrooms it has, who lives there, when it was built, and its value. With over 200-plus attributes per property, this persistent ID can be a radically informative tool in uniting a variety of data sets within a single, centralized point of data.
To learn more about proper address management, best practices, and use cases, you can view an archived version of the webinar here.