The challenges for performance management are reflected in the challenges set for IT systems supporting the automation of performance measurement and monitoring. These systems must provide accurate and actionable information and ensure organisation-wide consistency in decisions. Professor Stijn Viaene and senior research associate Luc Lutin explain that haphazard data collection will not do.
What are the typical challenges for BI applications?
Viaene: “It’s no longer enough to be able to look back, which is what traditional reporting does, or to monitor in real time. Your business intelligence (BI) applications should also support your planning process, which is what most companies do actually focus on when rolling out a system. In my view, too little attention is paid to how these applications can be integrated into day-to-day activities, especially at the strategic level. Balanced scorecards are used to kick off a meeting, but not to streamline it. It’s less of an issue at the operational level, where scorecards and dashboards are more concrete and specific.”
Could you give an example of successful use of BI in daily operations?
Lutin: “The Amsterdam-Amstelland police force is a wonderful example of an organisation that has managed to integrate BI applications into its daily activities at all levels. More specifically at the operational level, they use BI to optimise resource allocation. Using historical data, they’ve built models to predict criminal activity. The police force now uses this information to plan patrols in problem areas. As a result, the crime rate in these areas has dropped.”
Viaene: “It illustrates what we call ‘operationalising BI’: making BI available at the operational level of front-line employees so that it can trigger action. After all, no matter how accurate it is, information is useless unless it’s put to good use. KLM is another good example. They’d been collecting a lot of customer data from their frequent flyer programmes, but then hardly ever looked at it. Today this data is used to streamline all their customer-facing processes to ensure consistent customer experience and brand perception.”
What do you feel is the most important requirement for an automated performance management system?
Viaene: “It needs to provide the big picture, vertical as well as horizontal integration. It has to support, for example, balanced scorecards at the highest strategic level, as well as the cascading of these scorecards to lower operational levels with increasing detail. At the same time, it has to bridge organisational silos. All this assumes an end-to-end process view across levels, departments and functions.”
Lutin: “The Amsterdam-Amstelland police force looked at their end-to-end chain right from the start. This end-to-end view was replicated in their central data warehouse BinK!, which integrates information from the police, the courts and other parties in the chain, such as the prison service. It is used as the single data source for analysis and reporting at all levels. When someone is released from prison, for example, the police also know and can have a local officer check up on them to prevent them relapsing into crime.”
 BinK! = Boef in Keten(en), or Villain in Chains.
Checklist for practitioners
What should you bear in mind if you want automated and integrated information-driven performance management to be a success?
Invest in an enterprise data warehouse
An enterprise data warehouse as the “single point of truth” is essential to support organisation-wide consistency in decisions.
Secure senior management’s support and commitment
Projects of this scope and scale have a big impact on your organisation and require substantial investments. Senior management’s backing is crucial.
Set up a support organisation
You will need a competence centre to propose and roll out best practices and procedures. This has been one of the critical success factors in the Amsterdam-Amstelland police force.
Think big, but start small
Start with a well-chosen pilot project that you can use as a stepping-stone to further development. Invest in a flexible and scalable IT architecture that is responsive to changes in your business.
Approach BI as an end-to-end value chain
Failure to do so may result in solutions that are locally optimised, but that provide functional value only, rather than enterprise value.
Focus on right-time, actionable information
Remember: information comes at a cost. And information, even correct information, is useless unless you do something with it.
Business Intelligence (BI)
A set of technologies and processes that use data to analyse and understand business performance. BI includes data access, reporting and analytics.
Enterprise-wide consolidated and standardised data, grounded in agreed definitions, business rules and registration requirements. Not necessarily physically set up as a single centralised database.
For more practical tips, read “Managing for Performance Excellence”. This also provides a detailed description of the technical building blocks and applications supporting performance management, with special emphasis on enterprise data warehousing. The case study of the Amsterdam-Amstelland police force describes how an intelligent combination of information technology and organisational change has led to greater efficiency and effectiveness at all levels. Viaene and Lutin discuss the critical success factors and lessons learned.