Inaccurate, outdated and incomplete data is bad for business, especially in terms of profitability and competitive advantage. What are the consequences of relying on poor quality data? Claudia Imhoff, one of the most recognized experts in BI and analytics, shares her insight – and offers tips on how to improve your data quality.
Imhoff talked about the impact of poor quality data in a recent webinar, “Fix It Yourself: Improving and Monitoring the Quality of Your Data.” Imhoff is founder and president of Intelligent Solutions, Inc., a data management consulting firm.
The Causes of Poor Data Quality
There are several factors within the enterprise that can lead to poor quality data. One factor is the increasing use of technology. Changing from manual to more automated processes increases the risk of technology failures and user input error, Imhoff says. But technology also makes detection and audit trails more efficient.
Another cause of poor data quality includes systems integration, as a result of large-scale acquisitions and mergers. Consolidating data from multiple systems into a single platform can be a total nightmare – especially when the data is inconsistent, inaccurate, outdated and duplicated.
Imhoff also described another cause of poor data quality – organizational structure failure. “Breakdowns in established procedures and the failure to follow processes … all contribute to poor quality data,” Imhoff says.
The Consequences of Poor Data Quality
Imhoff cited a recent Gartner report that stated that an average organization loses $8.2 million annually through poor quality data, 22 percent estimated their annual losses to be $20 million and 4 percent report losses were $100 million.
“That’s astounding,” she says of the report’s figures. “Much of this loss is due to lost productivity. When we have to compensate for inaccuracies and have to work around figuring out how to deal with poor data quality – that’s a loss in productivity.”
Imhoff stressed the need for trustworthy data, saying it’s critical for competitiveness, operational efficiency, risk reduction, customer satisfaction and protection.
What Businesses and IT Need to Do to Combat Poor Data Quality
Incentives, incentives, incentives. That’s what Imhoff stresses to businesses. To promote overall data quality, offer incentives to employees who input data correctly, Imhoff suggests.
“Order entry clerks are being paid by the hour, they’re not being paid on the quality of the data,” she says. Imhoff also advices that businesses should view ALL data as important – to support targeted knowledge, information sharing and collaboration.
In terms of the IT sector, Imhoff says that most companies have a mix of disparate or poorly integrated legacy applications. This leads to inconsistent and fractured processes. She suggests being more “proactive” on improving the data quality process, instead of being “reactive.”
Imhoff says that IT should focus more on its data users (customer-focused) rather than just on technology alone.
She suggests asking the following questions to improve the data quality process:
- What are the processes to identify and react to risk?
- What data is needed and what shape is it in?
- How can proactive processes be established?
- What training is needed to improve collaboration, information sharing and data quality?