There is a movement within the insurance industry to consolidatedatabases from multiple departments in order to gain a completepolicyholder view. Typically, insurers with one database can seethe different channels and lines of businesses within the companyin which policyholders interact. In addition, it allows insurers tobetter comply with regulatory reporting and prevent premiumleakage.

But, consolidating databases has always been a challenge, whichis why so many insurers choose to maintain separate databases fordifferent departments or business units. The question is whether itis worth going through the business challenges of combining adatabase to get the business benefit of centralization.

While there are many considerations to take into account, onechallenge that faces insurers with any migration is data quality.Questions often arise about datastandardization, duplicates,outdated information, and missing fields. All of these challengesare surmountable if stakeholders properly consider how to mitigatethese processes prior to migration.

Data Standardization

Unfortunately, there is nostandard set of fields, abbreviationsor formats for data that are universally recognized. Dataconsistency is a free-for-all of sorts, with differences fromdepartment to department and insurer to insurer. In addition, humanerror is often cited as the top data quality error.

Due to this degree of inconsistency, migrating data into acentral database can be concerning. To minimize risk,administrators should decide what fields to keep and what format touse in each field.

Once the fields are set, as much information as possible shouldbestandardized. Contact information can easily be transposed to aset ofstandards through software tools. For instance, a mailingaddress can be changed to the country's national postalauthority.

Standardized and clean information makes it easier to identifyand remove duplicates and append information later in the migrationprocess.

Eliminating Duplicates

Duplicate accounts are problematic during any migration process.It is common for databases within different departments or evenfrom different business units to have overlapping policyholders orprospects.

Given this data landscape, eliminating duplicates can bedifficult. However,standardized contact data can be an ally. Sincecontact information is typically found in every database, it can beused to help household information and identify duplicatecontacts.

Insurers need to select matching elements to determine whereduplicates can be found. Administrators need to determine the levelof matching they want to accomplish, as well as the tolerance levelfor what is considered a duplicate record in the organization.

This process can be ridged or flexible, depending on the numberof records, how much an insurer wants to review potentialduplicates, or how much they want automatically merged.

Updating Data

Personal information is very fluid and changes on a frequentbasis. Policyholders are constantly moving or experiencing othermajor life events. Because of this, contact records can beincomplete and become outdated quickly.

The age of data may be unknown and cause administrators toquestion its accuracy. However, certain contact fields can beupdated with current information.

First, address information can be updated. The USPS has aNational Change of Address file that businesses can utilize throughthird party vendors. Insurers can identify whether an individualhas moved and send a request for updated address information withinthe last four years.

Additionally, email address information can be updated. Thirdparty vendors can remove syntax errors, but they can also check tosee if an email address is still active. Updating this type ofinformation is beneficial, as it aids in creating a database ofconsistent fields and updated records.

Preparing for success

The key to succeeding in any project is proper planning and afull understanding of the task at hand. Database migration is nodifferent. Insurers need to fully understand the information theyare looking to consolidate, along with its strengths andweaknesses.

Data quality is always a concern, but this can be mitigatedthrough the use of proper tools and techniques prior to and duringimplementation. Insurers shouldstandardize as much information aspossible before consolidation and then work to remove duplicates.Finally, they can look to update outdated information.

Overall, a centralized database can provide insuranceorganizations with valuable insight and a complete view of theirpolicyholders. While many are looking to take the centralizationplunge, insurers need to be confident that the move will improveoverall data quality.

Thomas Schutz is senior vice president and general managerat Experian QAS North America. He serves as the company's topexecutive for all strategic business decisions in the UnitedStatesand Canada. Schutz can be reached at [email protected].