This suggests that a measurement regime, to be really effective, should include predictors of future behavior and performance, as well as offering comments on the past. Thus the measurement regime has to identify the "drivers" of performance in any situation. It also suggests that a measurement regime should include effective presentation of results as a key component; data should be presented in a simple, consistent form which allows readers to focus on the important issues. More data is not necessarily better! The data should highlight issues and help in assessing priorities.

The balanced scorecard attempts to overcome the deficiencies of existing measurement systems - it is used to measure performance and develop strategies by analyzing results across a range of activities. At the very heart of the balanced scorecard method is the belief that organizational success can best be achieved and measured when viewed objectively from four perspectives:

1 Financial. How will we look to our stake holders?

2 Customer. How must we look to our customers?

3 Internal. What internal processes must we excel at?

4 Innovation. How can the organization learn and improve?

This well-rounded assessment provides management with a "balanced" view of the business.

The concept was developed by Robert Kaplan of Harvard Business School and David Norton of the Renaissance Group. It was titled to reflect its attempt to provide a balanced presentation of both financial and non-financial measures.

Unlike many management techniques and methodologies, the balanced scorecard concept avoids the over-dependence on simple "bean counting" - using only financial measure to determine success. It recognizes that operational measures - such as on-time delivery, order cycle time, product returns, etc. - should have an equal status with these more traditional financial measures. Often referred to as "leading indicators", these operational measures can often identify performance trends well in advance of crude profit and loss results.

Derived from an organization's vision and mission, strategy is at the heart of the process (see Figure 1). The strategy determines what is to be measured. Measures - often referred to as key performance indicators (KPIs) - are those things which will be used to determine progress towards achieving the strategic aims. For example, if an organization's strategy is to improve customer relations, one thing it may wish to measure is the amount of face-to-face time its salespeople spend with customers each month. Or, if another includes improving employee skills, it may wish to measure the hours of continuing education and training undertaken each year per employee.

The key to success with the balanced scorecard concept is the appropriateness and quality of the measures. Identifying measures which can track progress towards goals is often as hard as reaching the goals themselves. Some might ask why it is necessary - if we are making progress towards the goals, do we need to know? The important point, of course, is that we definitely need to know if we are not making progress towards our goals.

However, as is true with most things in life, things of real value are often not easily attained. If the balanced scorecard has received criticism, it is related to the effort involved in its implementation. On completion of the identification of the KPIs, the organization must begin an ongoing cycle of measuring the KPIs, and analyzing the performance results data. Thus, implementing a balanced scorecard presents a set of challenges. However, as is also true in life, modern technology has risen to the occasion to provide the ability to automate both the implementation and maintenance of a balanced scorecard. This is a natural evolution of the concept - since it involves an iterative cycle of measurement and analysis of a relatively static set of indicators, it is natural to use technology to automate parts of the process.

The technique has been around for several years, but only in the past year has interest in automating the process grown as many companies have established centralized in data warehouses that provide the corporate information necessary for a balanced scorecard analysis.

Performance data

The process of measuring performance data encompasses several tasks. The first step is to identify the sources of performance data. Such data may be of two kinds - internal and external.

Internal data is information about internal financial and operational activities. It is captured throughout the course of day-to-day operations and is often contained within existing computer applications and databases. For example, financial, sales, service and other types of operational transaction-based data arise naturally in the daily routine of operations. Measures should be part of an identified cause and effect chain that represents the strategy of the organization. Productivity is a key objective under the internal process quadrant of the system and it is important to include reliable productivity measures as part of the overall regime.

External data is information about the industry, about competitors, etc. Although reasonably easy to identify, capturing this data can present some interesting challenges. It usually comes from external industry analysis, and can increasingly be found on the World Wide Web. It also arises from benchmarking activity.

Of course, the fact that performance data has been captured by an operational system does not mean that it is usable by scorecard software. The scorecard software may not be able to access data contained within the inventory files of a manufacturing application. Some balanced scorecard solutions require that all data be fed into a proprietary database before it can be used by the scorecard software. So, even though an inventory transaction captured important information, that information will have to be re-captured (re-entered) into the scorecard software. This is typically a manual process and only serves to exacerbate the difficulty and complexity of implementing a balanced scorecard solution. It will also deter the continued use of the scorecard since the tedium and cost of continually recapturing the data will take its toll on the enthusiasm for the technique. Therefore, it is highly desirable to implement a scorecard system that provides direct connections into your existing computer applications and databases. This allows the scorecard to retrieve such performance data automatically at regularly scheduled intervals.

External data may be more difficult. While it will typically comprise only 10 per cent to 20 per cent of the input to the scorecard, it may be the most strategic in nature. External data is often not available electronically - in a standard database or file format. In such cases, it must be entered into a database or spreadsheet before it can be used by the scorecard system. If such data is available electronically, it can usually be imported by most scorecard systems. Again, it is highly desirable to select a scorecard system that can accept data from standard spreadsheet and ASCII file formats : this will dramatically reduce the amount of overhead required to implement the solution.

Once performance data has been identified and captured, the "measurement" must be determined. This "model" will reflect the organizational strategy. The measurement model will:

• set performance targets;
• normalize scores to agreed scales;
• establish the relative importance of, and weightings for, KPIs;
• identify the time periods within which performance data will be summarized and presented;
• establish a process for the provision of commentary for users of the scorecard.