Feedback process

To clarify the feedback process, a model proposed by Sink and Smith (1999) is used (Figure 3.6). The model consists of three parts: who is doing the leading and managing, what is being led and managed and what we manage with. Between these, there are three interfaces: the measurement-to-data interface, the information portrayal-to-information perception interface and the decision-to-action interface. They argue that for any management system, the goal is to optimize the performance of the next-larger system. That is, for a production line manager the goal is to optimize the performance of the factory, for the factory manager the goal is to optimize the supply of the organization, and for the supply manager the goal is to optimize the supply chain, and so on.

Figure 3.6. Model for performance feedback systems. (Sink and Smith 1999) WHO’S LEADING AND

WHAT IS BEING LED AND MANAGED Information

Using the above model, the feedback process could be described as follows: first, the data will be collected in the measurement-to-data interface, then analyzed and compared against the standard, after which the analysis will be reviewed in the information portrayal-to-information perception interface. After this, the managers next process the information with the knowledge they have, and make decisions impacting the process that originated the data. Figure 3.7 illustrates how the model fits into the cybernetic feedback model.

Figure 3.7. Feedback systems as a part of the cybernetic feedback model. Adapted from Simons (2000) and Sink and Smith (1999).

According to Crawford and Cox (1990) performance must be measured in ways that are simple to understand by those who are being evaluated. They even argue that performance data should be collected by the ones who are being evaluated. This is because the employees need to understand what they are evaluated on, and what kinds of changes are needed to improve on the measures. By giving the responsibility of data collection to the employees, it forces the measures to be simple enough, and also forces the employees to use the measures. It also makes the performance evaluation data more reliable to the employees.

In the analysis, or “what we manage with” -phase the data is compared to set standards and converted into information. Feedback information should flow in two dimensions:

vertically and horizontally. Vertical flow of feedback information refers to the flow of information between organizational levels. As has earlier been stated, strategic objectives should flow from company strategy to the plant level. The exchange of feedback information between levels of organization is part of creating organizational focus towards the objectives. This process can be facilitated by creating formal interactions and reports between organizational levels. (Lockamy and Cox 1995)

WHO’S LEADING AND

Horizontal flow of feedback information refers to the flow of information between functions such as procurement and sales on an organizational level. This is to promote organizational focus within the organizational level. A performance measure is seldom affected by only one function even though it might be owned by that function, and therefore it is important to ensure support from other functions as well. For example, manufacturing lead time might be the responsibility of production manager, but the support of procurement to get the materials on-time is needed for success. (Lockamy and Cox 1995)

Information portrayal refers here to how the analyzed information is presented. Topics such as how will it be visualized, in what forum it will be reviewed, how often, etc.

should be addressed. Crawford and Cox (1990) recommend using visual trend graphs to portray the information. They are easy to understand, and deteriorating performance is quickly identified with the aid of graphs. This information should be readily available.

The need for the information to be openly available, not only tracked and stored is echoed by Gomes et al. (2006). A decision that has to be made with any report is how frequently it should be updated. The consensus on performance measurement reporting in the literature seems to be that the reporting interval depends on the kind of measure, and organization level (1990).

Lockamy and Cox (1995) propose that at the plant level, the feedback should be as fast as possible. From the moment of measurement, the relevant stakeholders should have the information with a minimal passage of time. This is to provide an appropriate level of control. At this level, measurement systems should strive to be real-time (Vorne 2007). In higher levels, it becomes more difficult to have the information as quickly:

information about a product defect found at factory level is quickly acquired, but its impact to the profit and loss statement can only be seen later, when the financial data is published. Crawford and Cox (1990) who studied designing performance measurement systems for JIT operations propose that the kind of measure also affects to the reporting interval. They found that time-related performance-to-schedule measures should be reported more often, for example daily or weekly, than other measures, such as quality or inventory measures, which can be measured monthly instead.

The last phase of the feedback model, “who is leading and managing” and the decision-to-action interface will be discussed in length in chapter four. This is the phase where the performance measurement system becomes management control, and thus the concept of management control systems will be used to discuss the topic.

4. PERFORMANCE MEASUREMENT SYSTEM