What is master data management?

Master data management (MDM) is a data governance principle centered around creating a single source of truth for multiple departments and organizations to leverage in decision-making. To achieve true MDM, a single master record for each specific entity must be established, linked to other relevant data points, and shared with all other stakeholders. This ultimately means all individuals are working from the same information when performing analysis and making decisions, reducing errors and redundancies. 

Geospatial master data management

MDM principles apply to all types of data, including geographic information. The major focus of geospatial MDM is identifying an authoritative high-precision location for individual geographic entities, and then linking detailed attributes to those locations to inform standardized spatial analytics and decision-making. 

MDM is crucial in geospatial analytics because a database of places can include geographic coordinates, place names, place types, addresses, contact information, and more. All of these attributes can change, especially given the dynamic nature of the real world. Additionally, these attributes are not unique to a place - even latitude/longitude coordinates or addresses. Further adding to the complexity of this geospatial information is the potential for individual end users to rely on slightly different datasets that vary in freshness, accuracy, or completeness. 

Geospatial MDM helps resolve these inconsistencies by establishing a unified system of record for all of these places, where all end users are working from the same definition of location as well as quality and freshness of attributes. The best way to think about this is to picture a table; each row represents an authoritative location of which there are no duplicates, while each column provides additional detail about that location. 

Any industry working with geographic data must practice MDM in order to maintain a source of truth for the physical world that powers mission-critical analytics.

Geospatial master data management for P&C insurance

Property and casualty (P&C) insurance is one industry that focuses on geospatial MDM to streamline analytics and eliminate costly data redundancies. MDM in P&C is centered on a single source of truth for individual properties so that different departments can accurately assess risk and inform their decision-making. While the foundation of property analysis is an address, there are many more geospatial data components involved in P&C MDM that must be correctly accounted for.

For instance, P&C carriers need to accurately locate an address on Earth’s surface in order to fully understand its risk profile. From there, the property risk must be calculated based on any structures, their characteristics, and how susceptible they are to perils. This is often done using detailed property attributes, building footprints, and environmental data like elevation or flood zones. It’s important to also note that carriers must factor in the property parcel’s overall risk, and not just those associated with its structures. 

Property risk assessment is an integral part of operations across departments at P&C insurance carriers. From initial policy pricing during the underwriting process to reinsurance strategizing to claims management and everything in between, carriers need to have a single system of record for the properties they insure. Geospatial MDM helps insurers effectively leverage all of the previously mentioned property data across departments to prevent discrepancies in analytics that can undermine the bottom line, as well as reduce redundant and disparate datasets floating around the firm.

5 steps for better geospatial MDM in P&C insurance

Anyone who has worked with geospatial data knows that it is extremely complex and can be difficult to manage. Luckily, there are many different tools and strategies available to implement an MDM strategy to help you work with dynamically changing geographic information. This checklist outlines the five steps organizations should follow to develop a single source of truth for geospatial data in P&C insurance and any other industry looking to enhance its use of location intelligence.

1. Establish an authoritative location

The foundation of effective geospatial MDM is a high-precision location for every entity in your database. In P&C insurance, this is typically the address of a property being insured. While location may seem pretty straightforward, it’s actually where most departments and organizations experience the most discrepancies between datasets. This is because there are a few different methodologies for geocoding (deriving latitude and longitude coordinates from an address), which produce conflicting results. 

If one department at an insurer is using street-based geocoding to locate properties, they will be assessing risk based on a point location along a street segment. Another department could be relying on parcel centroid-based geocoding, which would have them measuring the risk at the center of the property’s land. Not only does this lead to discrepancies in risk assessment, but neither of these methods would locate the actual structures being insured. 

Instead, departments should agree on one geocoding methodology for locating properties on Earth’s surface. Building-Based Geocoding, also known as rooftop-level geocoding, is the optimal choice for geospatial MDM, as it provides a foundation rooted in the location of structures and enables end users to derive deeper property relationships.

2. Leverage a system of unique and persistent identifiers

Understanding property relationships is one of the more complex aspects of geospatial MDM. Properties are made up of many different elements that all contribute to a risk profile, including the parcel of land, main structure, and outbuildings. Addresses are also important elements to factor into property relationships, as they provide insight into how structures relate to each other and the larger parcel of land. These complexities and relationships can be difficult to reflect in a database but are critically important for P&C risk assessment and analytics.

True geospatial MDM involves using a system of unique and persistent identifiers for all of these elements in order to understand their relationships and organize data accordingly. It’s essential to note, however, that these IDs must be both unique and persistent in order to be of value. Real-world property elements like addresses and geographic coordinates can change, vary in precision, and be shared by multiple properties, rendering them less useful as a single source of truth in a database. 

Unique and persistent IDs are more resilient to real-world conditions, as they do not change even as other data attributes are updated, and are always unique to the row in question. This helps ensure everyone across an organization is working from the same accurate information, derived from the high-precision location established in step one of this checklist. You can think of it like a social security number (SSN) for people in the US, or an employee identification number (EIN) for individuals within organizations. While relevant attributes may change, the ID stays constant for streamlined data analytics and management.

3. Append relevant detail

Once you’ve located a property and all of its elements, and accurately reflected those geographic relationships in your database, you can layer in more attributes to add further detail for analysis. Appending additional columns to rows helps provide critical context for geospatial analysis, especially in risk assessment. 

For example, P&C carriers often leverage MDM to better associate risk scores and property characteristics with a record in their database. Property attributes such as roof type, number of rooms, square footage, and construction details are typically appended to an address record, indicating how resilient a structure is to damage and what its replacement cost will be. Columns denoting environmental data can also be associated to each row, providing information about how susceptible a property is to flooding from a nearby river or other natural hazard risks. While climate conditions or property details may change, the record itself stays persistent for stronger data management and analysis.

4. Develop a data sharing ecosystem

The goal of any MDM strategy is to develop a single source of truth for a particular type of information and provide all stakeholders with access so they can use that information in their analytics and decision-making. To achieve this, organizations must facilitate data sharing across departments, agencies, and any other group that contains end users. Only then can a unified system of record be leveraged for more accurate and streamlined decision-making.

In P&C insurance, many departments use geospatial data to understand property risk. However, many carriers do not have a unified system of record across these disparate departments, leading to discrepancies in risk assessment, claims management, and more. These different departments often onboard different sources for the same type of data, leading to data redundancies and misallocated resources. By implementing a firmwide data sharing program, carriers can prevent these discrepancies and redundancies throughout the insurance customer lifecycle, reducing expenses and optimizing workflows for accuracy. 

For instance, geospatial MDM enables an underwriter to query a database for one specific address and see the entire parcel it’s on as well as any structures associated with it. The underwriter can then assess risk based on the property attributes and environmental hazards located nearby. When the policyholder one day files a claim for that property, the claims department can query the same database and see the same risk profile, making it easier to respond to the claim more efficiently and accurately. Similarly, when the carrier is reviewing its reinsurance strategy, analysts can identify the underwriting and claims history of that property record and determine the best way to mitigate financial losses across their portfolio. This unified property analysis is possible through geospatial MDM.

5. Keep data up-to-date with real-world change

The reason why MDM is so important, especially in the geospatial industry, is due to how dynamically data needs to be updated in order to reflect real-world conditions. MDM ensures that information can be efficiently refreshed without breaking data models or creating duplicate records. By leveraging unique and persistent identifiers for each record, attributes can be updated an infinite number of times without disrupting the integrity of the database.

Geospatial MDM is particularly challenging because of how quickly the physical world changes. It can be difficult to keep that data up-to-date as a true reflection of the real world, but data needs to be that fresh in order for it to serve as a single source of truth for all stakeholders. Fortunately, advancements in artificial intelligence (AI)-based mapping have made it easier than ever for physical changes to be observed and recorded in geospatial databases. Every day, millions of images are captured from satellites, airplanes, drones, street-view cars, and more, providing the basis for updating geospatial data. While traditional data creation methods require hours of manual digitization, AI-based mapping has revolutionized how we extract insights from imagery.

The P&C insurance industry is rapidly adopting this AI-based imagery analysis as a way to facilitate stronger geospatial MDM. From imagery, carriers can observe changes in roof conditions, tree cover, and other property attributes that factor into risk profiling. AI-powered mapping detects these changes from imagery, giving carriers the ability to update databases in near real-time so that all stakeholders across departments are consistently working from a single source of truth for the physical world.

Get started with a geospatial foundation for MDM

At Ecopia AI (Ecopia), we’ve been extracting geospatial data and insights from imagery for over a decade. In 2018 we developed the first and only complete map of buildings in the United States, comprising over 173M building footprints and 240M+ primary and secondary addresses in addition to parcel boundaries. Our AI-based mapping systems keep this database up-to-date and append unique and persistent identifiers to each of these property elements to provide our P&C insurance clients with the strongest foundation for geospatial MDM, enhancing risk assessment workflows throughout the customer lifecycle.

Many companies claim to be experts in AI these days, but we’ve been laser-focused on AI-powered mapping for over 10 years and have the track record to prove it. Ecopia’s global partner network of over 30 imagery providers ensures we always have access to the freshest, most high-resolution geographic information from which to extract insights and detect change, resulting in the foundational data needed for effective geospatial MDM.

To see why Ecopia’s AI-based mapping solutions are trusted by P&C leaders for MDM at Tokio Marine, Chubb, USAA, and more, get in touch with our insurance team.