Last month, I discussed BI as a service. In this month’s article, I dig deeper into the different processes of ITSM and align them with the business intelligence (BI) way of working.

Business Intelligence Service Support

Configuration management is the physical and logical perspective of the business intelligence infrastructure and the services being provided for business intelligence.

This process stores a complete overview of the BI infrastructure and configuration items. It is not only necessary to draw up an inventory of all the different configuration items, but you need control of changes that are made to the hardware, software and documentation throughout the complete business intelligence lifecycle.

Change management is the process of effective management of all changes in your BI environment.

Change management needs standardized methods and procedures for all changes to facilitate the handling of change quicker and more efficiently. Each change needs to asses the impact of a proposed change in terms of cost, benefit and risk. One of the most interesting features to measure a BI change is to assess the impact throughout the business intelligence environment. For example, what is the impact, end to end when we change one of the ETL-flows, data definitions or reports?

Release management is testing, verification and release of changes to the BI environment.

The objective of release management is to preserve the integrity and availability of your business intelligence production environment while deploying new functionalities (software or hardware). Effective release management ensures timely and error free rollouts. Planning releases for business intelligence is a difficult task. Some users (supported by IT) always have an urgent need either to break the release management plan or make changes directly in the production environment. Release management plans should be clearly communicated. Each stakeholder needs a clear view of the schedule and impact of releases to be able to align his activities with these plans.

Incident management is the day-to-day process that restores normal acceptable service with a minimal impact on business continuity.

The definition of “normal service operation” is a necessity for effective incident management. This definition will be set out in the service level agreements (SLA) of the provided business intelligence services. The first question that needs to be asked is the definition of a business intelligence process. Is it a critical process? If not, the importance of handling incidents will have low priority and not handled during normal office hours.

Problem management is the diagnosis of the root causes of incidents in an effort to proactively eliminate and manage them

A BI knowledge base is an important instrument for problem management. Such a knowledge base needs to capture all the problems with common solutions and workarounds. This database is the source of information to rely on to easily identify symptoms of previous incidents and speed up a solution resolution for certain problems before they escalate into serious issues.

IT Service Delivery for Business Intelligence

Availability management optimizes IT infrastructure capabilities, services and support to minimize service outages and provide sustained levels of service to meet business requirements

How to we define availability? For stakeholders, it is usually the perception that the system is available or not. Availability is more inclusive than that. It also entails performance, reliability, maintainability, serviceability and security (cfr. ITIL V3 foundation). In practice, it will be a balance between the benefits of availability and associated cost.

BI service continuity is managing an organization's capability to provide the necessary level of service following an interruption of service.

This process is under pressure when upgrades, releases or new functionalities need to be installed. Each new component can impact the continuity of a company’s BI service. A backup plan/plan B is a necessity to manage this process effectively.

Capacity management enables an organization to tactically manage resources and strategically plan for future resource requirements

Effective capacity management requires constant measuring the limits or levels of capacity of any system. It should allow planners to make the right decisions about current and future capacity demand. A research study from META Group stated that many servers are utilized for less than 25% of a 24-hour day. It demonstrates that under- as well as over capacity must be controlled. For example, just-in-time capacity management is cost-effective but may be a risk if you can’t meet customer needs.

Service level management process maintains and improves the level of service to the organization.

Once we define the different business intelligence activities in a process, we are able to manage this process. It is a complex combination of several parameters to define SLAs for business intelligence. We must be careful that the definition of various SLAs takes into account the changing business environment: projects on hold, new projects, urgent changes, etc.

Financial management for BI services provides the organization with a way of managing the costs associated with the resources needed to meet requirements in an effective way.

Business as well as IT need to find a balance between projects, support and technology. A BI environment must move with the times and adapt new technologies. Oftentimes, BI projects are underestimated in time and cost. An efficient budgeting and planning approach should be in place to manage this process.

I didn’t handle service desk because it is a function and not a process. A service desk provides a central point of contact between users and IT. The service desk is a crucial function in BI as a service. It is vital that the service desk understands the needs of the customer and provides excellent customer service. The help desk person needs good facilitation skills and must be able to bridge the gap between business and IT.

Next month, I will discuss the roadmap to service management for business intelligence.

Recent articles by Ivan Schotsmans

• BI-Day Hot Spots
• Business Intelligence as a Service: Service Level Agreements
• Road Map to BI as a Service
• Introduction to BI as a Service