Next-generation BI is not just about technology. It's a new way of thinking about your data -- picking and choosing what data has value and what's worthless. So some experts say the first step is turning to master data management -- and, in turn, a data governance program. Others say the first step is establishing a corporate performance management strategy.
Above all, next-generation BI is about putting the right data in the hands of the right users across an organization. This involves a new way of approaching information worker involvement and controlling how workers gain access to critical information.
Why are MDM, data governance and CPM considered next-generation BI approaches?
Because they are key areas that organizations are failing to pull together. Many organizations grab data from various sources with the only common link being spreadsheet reports.
There are plenty of vendors playing in the master data management space, including SAP AG, IBM, Oracle Corp., Microsoft, TIBCO Software Inc. and Informatica Corp., and smaller companies like Purisma Inc., Siperian Inc. and Initiate Systems Inc. But establishing a data governance program first is key, and it is not that difficult to do.
"All you have to do is get the individuals who own the data sources to actually take ownership of them," said Joe Bugajski, an analyst at Burton Group Inc. in Midvale, Utah. "They don't have to know the nitty-gritty about the technology; IT will back them up on that. Just acknowledging who owns what data sources and what they should be paying attention to can be an enormous step forward that results in positive changes."
Gartner's Rayner recommends creating a corporate performance management (CPM) strategy first. This involves the following four steps:
Monitoring performance using historical data.
Implementing an enterprise management framework to choose the financial or performance metrics that will help move the organization toward more predictive BI.
Using predictive planning and modeling systems for forecasting to predict and identify the "what-ifs" and facilitate scenario planning.
Developing a strategy to identify patterns and make changes based on them.
What will next-generation BI dashboards and interfaceslook like?
Envision a video wall in which employees can point to a particular object, maybe a map that covers a customer region, touch the screen and drill down to see sales performance in that region at that moment.
It's not as far off as you may think. Customer-facing business intelligence dashboard displays are popping up in airports and retail chains. It is not inconceivable to think that they will become the business intelligence dashboards of the future.
"The world of old-line business intelligence software in which 10% or less of the employees slice and dice and shrink-wrap reports for other users will be replaced by multimedia presentation dashboards," Bugajski said. In this world, anyone in the company can see what's going on, whether it's a news feed about a competitor, internal profitability information or a fraud alert displaying ATM activity.
And BI tool vendors are stepping up as everyday employees reject the traditional interfaces for business intelligence systems. Users are instead favoring visualization mashup capabilities similar to those they use when searching for information on the Internet.
Is anyone actually doing next-generation BI yet?
Some CIOs are starting to lay foundations for predictive analytics and real-time business activity monitoring, or BAM.
For example, The Hillman Group Inc., a distributor of fasteners and maker of engraving systems and signs for large retailers such as Home Depot and smaller hardware outfits, has integrated its multimillion-dollar application for planning and scheduling orders into its BI portal. Dashboards refresh order status in real time. (The portal is built on IBM WebSphere, which is fed data from Information Builders WebFocus, which in turn lies on top of disparate data sources such as Excel, SQL Server and ERP applications.)
Next up at Hillman Group will be predictive analytics to identify the company's leading and lagging indicators and the relationship between those indicators. For example, a leading indicator would be the cost of zinc or other raw material; a lagging indicator would be margins.
Doesn't next-gen BI involve search technology?
Some cite search as the business intelligence reporting tool of the future. And the thought of being able to use search-like functionality to gather data across your organization, or even unstructured data across the Internet from news feeds, blogs and even tweets, is appealing to both business users and the CIOs in charge of business intelligence projects.
But the debate on how search and BI will interact, and whether mingling the two technologies is a good idea, will continue for years to come.
Forrester Research Inc. analyst Boris Evelson describes a BI utopia in which an appliance crawls both structured and unstructured data and places all information in optimized storage, where it becomes instantly available.
Not all elements of this are fantasy. Business Intelligence vendors are working on technologies to integrate unstructured data such as email seamlessly with BI. For example, Panorama Software Ltd. integrates Microsoft Outlook with BI tools.
Search technology that does more than spit back relevant links is also not so far off. Information Builders has Magnify and Microsoft has Bing, both of which promise to make users' search experiences more intuitive. Taking it a step further, why not just marry BI and search and call it guided search?
Other experts aren't embracing search technology as a business intelligence tool quite as much. Gartner's Rayner argues that the worst thing a business can do when it comes to business intelligence systems is to throw another tool at users. The problem, he says, is users already have too much access to information and too many BI tools that they don't use.
