Page 254 - Psychological Management of Individual Performance
P. 254

238        enhancing performance through goal-setting and feedback interventions
                                    CASE STUDY



                                    APROMES FEEDBACK SYSTEM FOR SERVICE TECHNICIANS
                                    We will illustrate our experiences at the hand of one of the cases mentioned earlier, namely
                                    that of field service technicians in a region of a company that supplies and services pho-
                                    tocopiers (Kleingeld, 1994). Each service region is headed by a supervisor and consists
                                    of approximately 20 service technicians, carrying out repairs and maintenance. Table 11.2
                                    shows a feedback report which contains an overview of one technician’s performance in a
                                    one-month period (as well as a six-month average).


                                    TABLE 11.2 Example of a feedback report
                                          ProMES feedback report page 1-Technician: Johan Karelse-Period: 01/99
                                                                Month           Moving average 6 months

                                                       Calls  Ind. score  Effect. Calls  Ind. score  Effect.
                                    1. Quality
                                    1.1. Mean Copies Between Calls (100)
                                    Copier model A      25   26300 copies  −1  164  23300 copies  −20
                                    Copier model E      40   29500 copies  −24  226  27900 copies  −32
                                    Average                             −15                     −27
                                    1.2. % Repeat calls (70)
                                    Copier model A      25    8.0%        17   164  17.1%       −21
                                    Copier model E      40   17.5%      −17    226  17.3%       −15
                                    Average                             −4                      −18
                                    2. Cost
                                    2.1. Labour time per call (50)
                                    Copier model A      25   97 min       12   164  87 min       22
                                    Copier model E      40   90 min       17   226  90 min       17
                                    Average                               15                     19
                                    2.2. Parts cost per call (55)
                                    Copier model A      25   108,-      −20    164  75,-         0
                                    Copier model E      40   123,-        6    226  121,-        5
                                    Average                             −4                       3
                                    Overall effectiveness               −8                      −23

                                    From the feedback report, it can be seen that the technicians consider themselves respon-
                                    sible for two responsibility areas: ‘Quality’ and ‘Cost’. For each responsibility area, two
                                    performance indicators were developed. High-quality performance is demonstrated by a
                                    high average number of copies between two breakdowns (Mean Copies Between Calls) and
                                    by a low repeat call percentage (a repeat call refers to a breakdown followed by another
                                    breakdown within a very short space of time). A technician’s performance on ‘Cost’ is
                                    measured by the average amount of time spent on each visit and the average price of spare
                                    parts used per visit.
                                      A crucial element of each ProMES system is the performance effectiveness curve, also
                                    called ‘contingency’ (Pritchard, 1990). These curves are developed for each indicator
   249   250   251   252   253   254   255   256   257   258   259