Self-service Data Integration: The Key Piece of the BI Puzzle
True self-service BI depends on self-service data integration to combine disparate data across all platforms.
By John Whittaker, Quest Software
The ultimate goal of business intelligence is to drive informed action faster while eliminating some IT headaches. IT is up to its eyeballs these days in creating reports and dashboards. The IT team doesn't conceptualize the data, they control it, make sure it's backed up, and ensure governance, but they are challenged by disparate data from a host of sources spanning a variety of platforms and formats.
The monolithic data layer of legacy BI platforms and the distributed data layer resulting from the emergence of big data and cloud computing function independently of each other, and it's impossible for data analysts and provisioners to access, cleanse, and format all of this data as fast as business analysts need it.
The traditional approach to BI has called for centralized control by IT -- "one version of the truth" -- with a wall between IT and the business. To facilitate access to this data, vendors of data warehouse platforms have added a pseudo "self service BI layer" to their tools. These tools -- such as Oracle Business Intelligence Enterprise Edition (OBIEE), Business Objects (BO), and IBM Cognos -- are designed to allow the business user to view and create ad hoc queries of the centrally controlled data. However, with today's explosion of data and data's many homes -- there could be hundreds of sources in a large organization -- the time IT requires to integrate the data to provide useful, actionable information does not synch up with how quickly business stakeholders need it to drive effective business results. The process is not just slow, it's also costly.
Because they must try to integrate customer, financial, and operational data from a variety of sources onto one playing field to get the information they need as fast as possible, frustrated business analysts try to go around IT with their own self-service BI tools. This can cause problems for IT, which must maintain governance of the data for quality control, management, and resource planning. How do they do that while also matching business analysts' needs for the speed of business change?
The modern business analyst must be able to integrate data from many sources and to view that integrated data from a number of angles to discover insight and actionable results. There are data integration tools that let business users circumvent IT and enable them to manually integrate the company's disparate data so it can be used all together, regardless of what kind of data it is or where it comes from.
These manual efforts often lead to multiple versions of the truth, however, which means business users are making decisions based on data that may not be the most relevant or timely. Compounding the problem is that these tools are not "IT sanctioned." The data they feed back to the user isn't cleaned up by IT, or even controlled by IT in any way, causing many governance and security headaches.
The perfect data integration tool should span the gap between IT and the business by providing high speed, heterogeneous integration of data from both IT controlled platforms and other data sources, which enables business users to produce new views of the business without having to initiate a slow, costly IT project thus delivering immediate business insight and actionable information. At the same time, it would reassure IT that when users query IT-sanctioned platforms, the data will not be removed from the platform and that it will be viewed with an IT-approved tool from an IT-trusted vendor.
In short, self-service data integration is the key to success both for business analysts (who need to quickly capture data to provide the right insight for good business decision-making) and IT management (which must ensure governance standards are not compromised). Essentially, self-service data integration allows IT to partner with business analysts to provide actionable information to the enterprise's executive officers in a more responsive and timely manner. It releases IT to focus on adding value to the enterprise with such activities as improving performance levels, boosting availability, and developing new applications that better align with business activities. Self-service data integration also is a cost saver, reducing the need for an army of IT specialists down to just one data analyst to cleanse and integrate the data.
The growth and success of any business depends on the ability to action insight gained from information about the company's performance quickly. The ability of business users and IT users to collaborate on self-service data integration is the key piece that completes the critical jigsaw puzzle that is business intelligence. It also closes the gap between IT and the business, enabling IT to quickly provide the right information to the business user while enabling the business user to include data sources required to complete the analysis, thereby reducing costs and eliminating the risks inherent in non-IT-sanctioned data integration tools.
This type of tool is critical to the self-service BI model that has emerged in recent years. Self-service BI enables the business user to answer business questions using a BI asset that has not been previously prepared by IT or a developer. The reality, however, is that true self-service BI depends on self-service data integration to combine disparate data across all platforms. Down the road, there will be no self-service BI without self-service DI.
John Whittaker is the director of product marketing at Quest Software (now a part of Dell) within the database management practice. John brings over 20 years of experience in high-tech organizations, previously holding positions with companies such as AlertSource, Trinent Internet Solutions, and KnowledgeCentrix, among others. Areas of expertise include, but are not limited to database management, information security, business continuity and disaster recovery planning, managed services, and e-commerce. You can contact the author at [email protected].