As data volumes and varieties grow, more organizations are looking for ways to get value out of their big data.
Data discovery, an advanced analytic capability, allows business analysts to discover insights in their data without depending on IT to create subsets/business views or merge/cleanse the data. Using new data discovery and visualization tools, analysts can now conduct discovery themselves, find insights, and share them with other users, speeding up “insight to action” for competitive advantage. After all it’s the business users who really best understand their own data and can readily see threats and opportunities through that data.
As you review these new tools for use in your own organization, keep in mind the key capabilities and caveats and common business applications for self-service business intelligence (BI) listed below.
Critical Capabilities and Caveats of Data Discovery Tools
- Agility and high performance for faster data exploration and combining large data sets
- Real-time analysis for operational business intelligence to find insights and take immediate action on them
- Flexibility to combine multiple data sources easily—without IT’s help
- Easy to use and learn
- Visual appeal and a large choice of graphic display options to communicate insights effectively
- Dashboards that are easy for users to create
- Access via smart phones, tablets, and other mobile devices
- Collaboration so it’s easy to share and take action on insights
- Built-in data manipulation (e.g., hierarchy creation, etc.) without scripting or programming
While all the above end-user features should be included in your selection criteria, consideration should also be given to a tool’s ability to integrate into your broader business intelligence suite for ad-hoc reporting, formatted reporting, predictive analysis, and other BI actions—all of which complement self-service capabilities in data discovery tools.
Organizations should also beware of selecting multiple self-service tools across end-user area, which can create more islands of data, different user interfaces, and interoperability issues.
Common Business Applications of
- Increasing revenue: Two common, cross-industry marketing applications are customer and distribution analysis. Both are used to identify characteristics for market and channel segmentation in relation to marketing campaigns. Similarly in sales, product revenue and margin analysis allow you to see into volume, profitability, and profitability drivers, so you can sell more profitable products and less unprofitable ones and determine the necessary changes to make products more profitable.
- Reducing risk: Another common application across many industries is fraud detection. Whether for claims fraud (insurance), inventory “leakage” (retail), or credit card fraud (banking), data discovery can help identify outliers that may indicate fraud and clusters of fraud characteristics before you run your data mining.
- Decreasing expenses: Data discovery also helps optimize business process efficiency and workforce productivity through best- and worst-performance analysis and benchmarking. In fact, a placebo effect is often realized just from sharing of best and worst performance benchmarks, resulting in a performance lift. You can make substantial improvements when you apply the actions/findings in business process reengineering, training, or incentive/talent-management and workforce-scheduling applications.
Regardless of the tool you choose, the most important thing is to get started. Use it against your data, analyze, share/collaborate on your findings, take action, and measure the results. This closed-loop process allows you to leverage the value in your data assets and move the needle to improve your market performance.
Bottom line: most organizations have well conquered descriptive business intelligence (looking backward at what happened). The value in big data lies in prescriptive business intelligence and using self-service BI tools like data discovery to determine what will happen and how to capitalize on it.