Ensuring that information in corporate systems is as accurate as possible is becoming more and more central to business success, which increasingly is giving business users a role to play in the data quality process. And to put them in a position to play that role well, companies first need to develop and implement effective data quality training mechanisms that outline best practices and actions to avoid, according to data management analysts.
Like everyone else who is involved in data quality improvement efforts, business users should grasp the value of high-quality data, the analysts said. The quality of an organization’s data can make or break it, they added, and a training program for end users should start by illustrating the business benefits of sterling data quality compared to the problems that bad data can cause.
It’s vital that front-line workers see the big picture on data quality, said Andres Perez, president and senior information management consultant at IRM Consulting Ltd. in San Antonio. “Business users -- the people who actually touch the data -- need training to help them understand how information is used and how [companies] get impacted when it is not correct,” he said.
Perez added that real-world examples and statistics can help illustrate the data quality imperative, especially the potential downsides of not paying proper attention to data accuracy and consistency. “Anecdotal evidence helps people visualize the problem,” he said, “and statistical evidence drives the issue home to show the depth of the problem.”
Ted Friedman, an analyst at Gartner Inc. in Stamford, Conn., said a well-executed training program is an essential element in shifting the mind-set and culture in an organization to embrace the importance of complying with internal data quality standards. And effective data quality training typically isn’t a one-time thing, Friedman said; desired practices should be reinforced over time, and the training content should be revisited periodically to make sure it isn’t out of date.
Ideally, the business user’s role in data quality is part of an overarching plan that also includes responsibilities for IT and data quality professionals, data stewards and corporate and business executives. Analysts said it’s useful for users to understand the roles of the other players in the context of an overall data quality workflow in order to convey a sense of the broad scope of enterprise awareness about the importance of accurate data.
No uniformity on data quality training
Obviously, the specifics of training business users on data quality best practices are predicated on the makeup of a particular organization. Defining exactly what data quality means to a company is part of that, and in many cases the answer might not be self-evident. David Loshin, president of consultancy Knowledge Integrity Inc. in Silver Spring, Md., said the degree of data quality that’s required for different types of information should be determined in accordance with the data’s function and importance, in keeping with the catch phrase fit for use. “If you don’t know how information is being used, it isn’t possible to monitor any perceived improvement in the quality of the data,” Loshin said.
Because of limited resources, it’s rarely feasible to mount a full-bore effort to perfect all of an organization’s data, Friedman noted. As part of the training process, “it’s critical to set priorities and a proper scope in data quality efforts so that you are applying your resources in an optimal manner given the potential payback,” he said. “That is, [telling users that] by improving the quality of this data, you generate the most possible positive outcomes for the organization.”
Increasingly, enterprises are handling data that comes from social networks and other outside sources. Trainers should alert business users to regard external data with some skepticism when it comes to accuracy, according to Loshin. “We live in a world where half the data in an organization comes from an unknown origin, which creates a little bit of complexity,” he said. “You can’t go to Twitter and tell people that they need to spell their names correctly.”
Shifting to actions that business users should be trained to avoid, the chief culprit is viewing data quality as an occasional exercise. To help prevent that, Gartner analyst Lyn Robison recommends building language into job descriptions to explicitly spell out the need for consistency on data quality as well as specific data quality responsibilities.
Business users should also be taught to not mentally wander off into data quality isolation, analysts said. Done right, data quality management is teamwork personified; going your own way often leads to bad habits, such as a lack of discipline. Taking an eye off the ball and neglecting critical details is often the first step down a slippery slope of data mistakes that can compound into big business problems.
Roger du Mars is a freelance writer based in Redmond, Wash. He has written for publications such as Time, USA Today and The Boston Globe, and he previously was the Seoul, South Korea, bureau chief of Asiaweek and the South China Morning Post.
This was first published in July 2012