Master Data Management Tips: Fitness for Purpose

by | Apr 5, 2018 | Master Data Management

Data Quality should be measured within the context of how it will be used. The concept of “Fitness for Purpose” is central to any data quality metrics within a data governance program.

Many data governance programs struggle right out of the gate as organizations attempt to define data quality standards for enterprise data. The governance team will often discover that individual business stakeholders have different, sometimes conflicting needs for some enterprise data.

There are many reasons why people have a different view of enterprise data. 

  1. Legal or regulatory policies may require that an enterprise application preserves the information that was gathered when the contract was signed even if this information is now out-of-date
  2. System limitations and/or the prohibitive cost of system modifications result in data anomalies such as creating duplicate customer accounts to handle multiple billing options
  3. A customer-facing web site relies upon email for registration and communication, but it also gathers a physical address during registration. The web site uses the customer’s state to distribute the leads to sales-people so it does not really care about the validity of the other address elements resulting in some “bad” addresses in the database.

All three of these examples may impact the number of customer records held in each system and/or the demographic information captured on each customer record. Each individual system owner considers that their data is of sufficient quality even though errors and out-of-date data exists. The data governance team will find it challenging to get alignment around policies for customer addresses from these three system owners without factoring in the purpose of the data.


Data governance teams can avoid much frustration by defining all data quality measures in terms of the intended use of the data, and by recognizing that data created for different purposes can rarely be combined unless the data model recognizes the context of the origination of the data.


Opt-in with Gaine for More Insight

Keep ahead of the rest with critical insight into Healthcare and Life Sciences MDM and interoperability technique, best practices, and the latest solutions.