Advances in data management and other technologies have made real-time data collection and analysis possible, leading to new types of business intelligence applications. Operational BI collects and analyzes data on the fly, putting it directly into the hands of front-line workers.
To learn more about operational BI and its benefits, Satori interviewed Wayne W. Eckerson, a major proponent of operational BI and author of "Best Practices in Operational BI: Converging Analytical and Operational Processes." Eckerson is director of research for The Data Warehousing Institute, a worldwide association of business intelligence and data warehousing professionals that provides education, training, research and certification.
Q: What is operational business intelligence and how does it differ from other types of BI?
Eckerson: Operational BI gathers data from the processes that drive an organization daily, hourly or minute-by-minute with an eye to optimizing them. This is a radical departure for BI, which in the past was used solely by business analysts who examined historical data to discover trends and patterns that could be used to make decisions, improve plans and increase business efficiency. In contrast, in operational BI, managers and front-line workers use just-in-time data culled from multiple systems to monitor processes and take proactive action.
Q: What are operational BI's chief benefits?
Eckerson: It empowers managers and front-line workers to proactively fix problems before they escalate out of control or to exploit opportunities before they disappear. It helps executives keep their fingers on the pulse of the organization and adapt more quickly when conditions change.
Q: Can some business units benefit more than others from operational BI?
Eckerson: Every department can benefit. The operations group can track activity on a just-in-time basis, whether it's a shipping department tasked with improving delivery times or a customer service organization that must improve service while reducing expenses. The sales department can use it to track call center activities or on an e-commerce site and make changes to optimize revenues and profits. Marketing can use it to track Web traffic patterns and improve cross-sell and upsell opportunities through various sales channels. Even human resources in a large organization can benefit by tracking the movement of employees into and out of the workforce and tracking the skill level of its workforce.
Q: What are some real-world ways companies are profiting from operational BI projects?
Eckerson: Many companies are building dashboards to monitor operational processes where data is updated on a near-real-time basis. One application I keep seeing is the use of operational dashboards to manage inside sales activities. The dashboard becomes the system of record for hundreds of salespeople and their managers to track individual and group progress towards achieving daily quotas. The dashboard tracks all incoming and outgoing calls and the status of each customer or prospect within the sales cycle and displays green, yellow or red lights across a variety of metrics for each salesperson and sales team. The dashboard also uses daily activity to project end-of-day results and compare against plan, rank salespeople by performance and summarize commissions for each salesperson.
These dashboards are a fantastic way to motivate salespeople since they - and everyone else -can instantly see the impact of their actions every 15 minutes. It's even better for managers, who can spot which salespeople are having difficulty and provide assistance when working with a difficult customer or unique situation.
Q: What advice would you offer to companies considering an operational BI project?
Eckerson: It's much easier and cheaper to build operational BI capabilities into your architecture up front, so you don't have to rebuild or revamp it when your business eventually requires it. If you don't do this, you'll be tempted to create separate environments for historical data and real-time data, which means that you are splitting the content and creating potential problems with data consistency.
Usually, people want to view just-in-time data within the context of historical data so they have some context for what is happening now. While it's possible to meld two different data sources together to do this, it's easier in the long run if you can support both streams of data from the same system.
Learn more about "Best Practices in Operational BI: Converging Analytical and Operational Processes."
 |
 |
