The IT Department and Business Intelligence

Key Points

  • The IT department should support all three pillars of the business intelligence strategy. However, we should appreciate that business intelligence environments have different characteristics to operational systems and adjust the engagement model accordingly
  • User communities should have greater freedom to make their own mistakes in a business intelligence environment, especially in the initial stages of a program

IT managers may not have a clear picture of where they sit in the business intelligence technology strategy. Vendors support tools sold straight to the business units and key databases sit outside of their control. Intuitively, they feel that this lack of centralisation and coordination will eventually manifest itself in a maintenance nightmare for their department.

About CMBI

The IT department  and business intelligence

Sometimes, I find that IT managers do not have a clear picture of where they sit in the business intelligence technology strategy. Vendors support tools sold straight to the business units and key databases sit outside of their control. Intuitively, they feel that this lack of centralisation and coordination will eventually manifest itself in a maintenance nightmare for their department.
Driving this trend are vendors who exploit the frustration of user communities by claiming their solution requires no support from professional technologists. All too often, the user community just become dependent on the vendor’s consultants rather than internal IT resources.
This can be a difficult chicken-and-egg type of problem. The IT department are reluctant to invest money in dedicated BI resources if the business units would prefer to work with vendors. Business units will continue to use external resources if there is no one in the IT department to support them.
Obviously, a complete reliance on vendors misses the opportunity of building the skills and knowledge within the organisation. It means that we must formally scope and cost each new BI project rather than allowing requirements to flourish organically through closer cooperation between departments.
Realistically, even power users may need assistance from time to time, particularly with ETL (Extract, Transform, Load), data modelling and performance tuning. If technology department personnel are available for review and advice then this will enhance the quality and turnaround of BI solutions.

Business intelligence environment administration

It is important to differentiate business intelligence environments from operational systems where it is necessary for the IT department to take a hands-on role in day-to-day maintenance and control, to ensure availability and smooth running of the business.
User communities should have greater freedom to make their own mistakes in a BI environment, especially in the initial stages of a program. The IT department should provide the safety net, for instance by creating backup schedules for key data stores. But we must accept that users may crash the BI server, accidentally delete reports, run ETL processes at inappropriate times, create inefficient cubes, and cause a certain amount of technology chaos as part of a natural learning curve. If we reprimand or patronise these experiments, users will quickly return to spreadsheets and personal databases and BI opportunities will be lost.

Table – Division of responsibilities between the IT department and the user community for implementing the three pillars of the business intelligence strategy
BI strategy BI task IT department User community

Process

Defining objectives

Advising on data availability and existing solutions

Defining business goals

Data

Increasing data availability

Increase visibility of data across business units. Provide development support for data integration

Own and manage data

Technology

Database and ETL technology

Choose technology. Support configuration and provide training and support. Development work if required by users

Power users can develop solutions. Other users may just specify requirements

Technology

User facing BI tools

Create environment and infrastructure. Provide support and examples of best practice if able to

Choose tools. Develop solutions

 

See Also