Do you have BI Performance Anxiety ?
BI is key to enabling companies to turn oceans of data into predictive models and actionable decisions. However, a survey of 353 executives in large companies, reported that their chief BI concern was the performance of various BI solutions.
Development, support and enhancement teams are typically deployed to address BI performance challenges with varied success. But most companies don’t have a dedicated focus on performance.
A BI Center of Excellence (BI CoE) measured by performance KPIs and service metrics is one solution to this problem. This is not an area that traditionally draws high-level attention or is featured in a dedicated CoE initiative, yet in the right circumstances it offers unique value.
BI COE for Performance
An effective BI CoE consists of a combination of troubleshooting skills, knowledge of new industry trends, new developments in the BI environment, solid grounding in the development function, and architectural expertise. A BI CoE operation can provide ongoing production support as well as routine maintenance. Some tasks that can be performed by a BI COE are:
- Performance Tuning: Identifying bottlenecks, troubleshooting and fixing
- Enhancements: Modification to existing reporting systems
- Production Support: Data Fixes, Ongoing Support, Scheduling, Monitoring and Admin Activities
- Maintenance: Planned Recycling Refresh and Synchronizing environments
A typical Business Objects implementation may get 250 enhancement requests a year along with 2000 production support requests depending on the complexity and maturity of the environment. Responses to these requests can take anywhere from a week for an enhancement request to a few hours for production support.
A significant benefit from the adoption of a BI CoE model is the identification of numerous functions that can be automated. These include time-consuming manual tasks, such as:
- Migration and version control
- Granular level of audit log (sub-object level to know exactly what was changed)
- Report Impact analysis for a given change to schema or Universe
Bottom Line Impact
It is not uncommon for companies with about 1000 users and 20 functional departments to spend in excess of a $1M in support and enhancement of a BI implementation. Some of these can be hidden costs such as opportunity costs of specialized resources which are not directly measured. Implementing a dedicated BI COE with efficient processes can improve BI performance while savings of up to 50% over traditional approaches.
The fundamental goal of a BI implementation is to create valuable information that directly impacts business decisions. BI performance can make the difference between a good decision and a bad one. Corporations who have invested considerable resources on solutions are looking to get their money’s worth in performance. A well-executed BI CoE can directly impact the ROI of this investment.
BI is high on the strategic agenda of CIOs as companies require more insight into information as data keeps accumulating. Better decision-making and competitive intelligence is well beyond building data warehouses and data marts.
Transforming raw data into meaningful and useful information, trends and forecasts that can be used to enable more effective strategic, tactical, and operational insights and decision-making requires a solid architecture, development, QA/testing and a support/enhancement structure. A BI CoE can address both requirements.