DIY BI: A guide to self-service business intelligence implementation
A comprehensive collection of articles, videos and more, hand-picked by our editors
More and more companies are looking to deploy self-service analytics tools that enable business users to make more data-driven decisions without having to rely on IT or business intelligence staffers to write queries and create reports. But the self-service approach also introduces some risks, and experienced BI managers and consultants said organizations need to think carefully about the analytics functions they move out of IT and push onto business departments.
The latest Gartner Magic Quadrant for Business Intelligence and Analytics report, published in February, predicts that self-service data discovery software is going to be a top need for businesses in the year ahead. At the same time, the report says that few BI vendors offer tools that grant end users the freedom to explore data as they please while providing IT and BI teams with the ability to manage effective data governance policies. As a result, many organizations feel tension between ease of use and governance requirements.
Speaking at the 2014 TDWI BI Executive Summit in Las Vegas, Peter Mueller, head of the global business analytics program at Lonza Pharma & Biotech, said people often refer to data as a corporate asset. But that view of information's business value can be jeopardized by out-of-control BI and analytics applications, particularly when the analysis is done by workers with little training. If a company isn't careful, the analysis process can be mismanaged or misinterpreted, leading to potentially damaging data inconsistencies and faulty findings. Putting in place a self-service BI tool allows more users to access and analyze data, but it also increases the likelihood that problems will occur, according to Mueller.
"Data is a proxy for reality," he said. "It has little relevance unless a tool or person does something with it. So if we are trying to treat it as an asset, it becomes very difficult."
Peter Muellerglobal business analytics program manager, Lonza Pharma & Biotech
Mueller stressed the importance of user education at businesses that are implementing self-service analytics systems. Simply turning the technology over to business users without appropriate training increases the chance that they will mismanage data, he said. Even though people are still likely to store files in the wrong location or copy data for their own use, providing instructions on the responsibilities that come with being a data owner may help limit such occurrences.
IT and BI teams should take the lead on training, said David Cole, chief technology officer at Claraview, a BI consulting services provider in Reston, Va., that is owned by technology vendor Teradata Corp. Cole acknowledged that self-service tools can help speed up the process of data discovery for business users and make the findings they generate more relevant. "If it's stuck in some back office somewhere, it's not going to take off," he said, referring to BI software.
But an organization also needs to understand the limitations of its business users. In many cases, they don't know about data management best practices or have experience using data to support decision making -- IT and BI professionals are more likely to have those skills. So when a company begins rolling out self-service software to business users, the IT department or BI team should take some time to share its knowledge up front, Cole said.
Integration of new data sources is another thing IT should be responsible for, said Cindi Howson, founder of BI Scorecard, a research and consulting company in Sparta, N.J., that publishes technical evaluations of BI and analytics tools. She pointed out that a lot of self-service tools now make it easy for users to import new sources of data. That's fine when a user is pulling in Excel spreadsheets or conventional structured data, Howson said. But when the type of data being brought into a BI system starts getting complex and varying in structure, it may require a greater level of technical expertise to integrate the information effectively.
Still, businesses implementing self-service analytics tools will have to live with a degree of messiness in their data, Howson said. That's the tradeoff: You might not have perfectly managed data, but you will empower a greater number of people to find meaningful correlations in the data, which ultimately may be more valuable. Self-service governance policies need to find the proper line between preserving data integrity and not slowing users down.
"Think about that balance," Howson said. "That agility has business value."
Learn about the balance between self-service flexibility and governance
See where BI teams come up short in managing self-service analytics programs
Read why a customized approach to self-service BI software works best for business users