The willing, the indifferent, and the unwilling;
Call center employees’ reactions to control
Paper for the 28th ILPC at RutgersUniversity, 15-17 March 2010
Roel Schouteten (RadboudUniversityNijmegen, the Netherlands)
Jos Benders (TilburgUniversity, the Netherlands)
Els van den Bosch (Radboud University Nijmegen, the Netherlands)
Correspondence to:
Roel Schouteten
RadboudUniversityNijmegen
Institute of Management Research
P.O. Box 9108
NL-6500 HK Nijmegen
T: +31 24 361 1267
E:
PLEASE CONSULT FIRST AUTHOR BEFORE QUOTING
Abstract
Call centers are known for operating intensive control regimes. As labor process researchers have stressed and repeatedly empirically demonstrated, this does not mean that call center agents are completely controlled.
Highly influential in this respect are Bain and Taylor (2000) who argued against an image suggested by Fernie and Metcalf that call center employees were completely controlled, as if in a perfected authoritarian regime. Instead, Bain and Taylor showed, there is always room for resistance, however all-encompassing the control system is deemed to be designed. Following this work, many researchers showed how employees actively but foremost passively resisted the attempts to control them. The image that this strand of literature conveys is that agents’ reactions are relatively homogenous: they resist control, albeit in different ways.
Based on a case study (van den Bosch, 2009) we argue that agents react in different ways. We distinguish three groups of agents:
- control has a positive effect on performance and motivation
- control has a negative effect on performance and motivation
- control has no effect on performance and motivation
Although the actual level of control was the same for all agents, the effects were different. The main determinant of the differences is the level of perceived control, which is determined by individual characteristics and work characteristics, including ambition, self-confidence, autonomy, communication, and supervisory support. For instance, whereas group 1 agents are ambitious and experience control as a necessity to enhance call centre service quality, group 2 agents feel that control is a management tool to assess and sanction underperforming agents. Group 3 agents don’t care about the level of control and perform at their best regardless the level of control. As a result, groups 1 and 3 experience low levels of control and group 2 agents experience a similar level of control as high, which has a negative influence on their motivation and performance.
Our results suggest that it is advisable to distinguish at least three groups of employees: the unwilling, the indifferent and the willing. This sketches a much broader spectrum of employee reactions than is the case in much of the literature.
Introduction
As a growing number of workers are employed in call centers (Taylor & Bain, 1999; Schalk & van Rijckevorsel, 2007; Strandberg & Wahlberg, 2007), an abundance of articles is being published on the (quality of) work and working conditions in this sector. Organisations are using call centers to communicate directly with customers, manage customer complaints and maintain customer loyalty (Schalk & van Rijckevorsel, 2007). As such, call centers are an important means for organisations to stay in touch with large numbers of (potential) customers and gain competitive advantage. As a result, the level of service quality, and therefore employee performance, are important issues in designing call center work.
In order to improve call center employee performance, electronic performance-monitoring systems are extensively being used (Bain & Taylor, 2000), resulting in intense levels of control. Features of electronic control systems by the use of information technology include (Wickham & Collins, 2004: 11):
- work can always be located,
- mistakes can quickly be tracked back to their originator,
- jobs can no longer be ‘parked’ or postponed,
- it is no longer possible to ‘pass the buck’.
The use of such systems results in more and more detailed information on employee performance (Sia et al., 2002).
Besides, the work in call centers is often, though not always (Batt & Moynihan, 2002), short-cyclic and offers few challenges to employees. This is the result of a continuing Taylorization of the work, which in call centers is embedded in the key technology of Automatic Call Distribution Systems (Wickham & Collins, 2004). The interaction with the customer is broken down into standardized sections and is carried out according to predetermined routines. Not for noting have call centers been characterized as “electronic assembly lines” or “an assembly line in the head” (Taylor & Bain, 1999). Like in assembly lines, the limited challenge provided by the work means that there is little intrinsic motivation. Especially the limited amount of autonomy – an important determinant of employee motivation (Oldham et al., 1976) – is a negative consequence of high control intensity (D’Cruz & Noronha, 2006).
Especially labor-process researchers have discussed employee reactions to the control regimes. Highly influential was a study by Bain and Taylor (2000) who argued against an image suggested by Fernie and Metcalf that call center employees were completely controlled (in an ‘electronic panopticon’), as if in a perfected authoritarian regime. Instead, Bain and Taylor showed, there is always room for resistance, however all-encompassing the control system is deemed to be designed. Following this work, many researchers showed how employees actively but foremost passively resisted the attempts to control them. Employee resistance became a predominant theme, arguably replacing the view of total control held earlier.
However, it appears plausible that different reactions are possible. For instance, Boiral (2003), writing on employee reactions to ISO, distinguishes three groups: the believers, the active resisters, and the indifferent. In a similar vain, call center agents are likely to show different reactions to control regimes. Could control actually be welcomed? Being aimed at improving the agents’ calling skills by providing feedback based on system information, control enhances the development opportunities (training on the job) for the agents. Training and career opportunities offer a strong motivation for agents with a strong career orientation. If employees are active agents and reflect a diverse population in terms of class and norms, it seems likely that more reactions than just resistance occur in praxis. We conducted a case study in the in-house call center of a Dutch insurance company who is, at the time of the study, at the verge of introducing a new more sophisticated Call Center Management System. The call center management intends to use the system’s control opportunities to enhance the level of service quality. It even developed a handbook and protocols on how to use the new system in order to improve call center performance.
Method
In the case study we investigated the call center agents’ responses to the introduction of a new call center management system. As opposed to the old system the new system – called Voice Logging and Quality Monitoring – contains more enhanced control systems, because all calls, as well as system use (in particular the computer screens used) are digitally recorded. This system enables managers to check calls easily, without agents noticing when they are being controlled. In this setting we investigated the effects of the level of control on absenteeism, performance, turnover intention and motivation.
Interviews
The third author conducted interviews with eighteen respondents: fifteen call center agents, two supervisors and the floor manager. These are representative of all employees regarding age, gender and seniority (length of service). These respondents were interviewed before the introduction of the new system. Shortly after the introduction, six call center agents were interviewed again to determine the effects of the new system. All interviews were recorded, transcribed and coded. The interview reports were also send to the respondents to check the validity of the interpretations.
Observations
Next to interviews the third author conducted observations to investigate the supervisors’ ways of using the possible control mechanisms embedded in the technology. During the observations additional information was gathered through questions and listening in to calls. The observations increased the insight in the technical control possibilities, the ways that supervisors actually used them, and the employees’ reactions to control.
Document content analysis
Most protocols and processes of the in-house call center, as well as the performance criteria are documented in an Agent Handbook. Besides, regarding the Voice Logging (VL) and Quality Monitoring (QM) system, there is a protocol on how to use the system. These documents were studied as background information for the interviews and observations.
Measures
Control: technological control emerges from the use of automation to structure and monitor the work in ways that make compliance a virtual certainty (Wicks, 2002: 672). It is aimed at ensuring the desired behavior by the call center agents in order to deliver the necessary service quality. In the observations and interviews with the supervisors and floor manager, control is measured as the extent to which supervisors actually use the available control mechanisms. In the interviews with agents, control is measured as the extent to which agents are aware of the fact that call duration, clearing times, idle time, etc. are being registered and the fact that calls and computer screens are being recorded and listened in to by supervisors.
Dependent variables: as dependent variables and as effects of control we measured absenteeism, turnover intention, motivation, and performance of call center agents. The call center sector is often characterized by high levels of absenteeism (Mulholland, 2004). One of the explanations for the high levels of absenteeism is the work organization. Work that is characterized by short cyclical work and high control intensity (like an assembly line in the head) results in high levels of stress (Taylor & Bain, 2001), resulting in absenteeism (Lankshear et al., 2001; Schalk & van Rijckevorsel, 2007; Wickham & Collins, 2004). Absenteeism, as reported by the insurance company, is the number of hours employees are absent as a result of illness, divided by the total number of hours worked. In the interviews the respondents were asked whether the illness was related to the work or working conditions.
Besides high levels of absenteeism, the call center sector is also characterized by high turnover levels (Lankshear et al., 2001; Schalk & van Rijckevorsel, 2007; Wickham & Collins, 2004). Job dissatisfaction, as a result of high experienced levels of control, results in high turnover rates (Strandberg & Wahlberg, 2007, Mulholland, 2004). In the interviews the respondents were asked whether they intended to leave the company and why. As a result, we measured turnover intention; not actual turnover.
Motivation is an important concept in explaining employee behavior (Steers et al., 2004; Locke & Latham, 2004) and can be divided into intrinsic and extrinsic motivation (Ryan & Deci, 2000). Knowing that the work is being controlled by the supervisor may result in higher levels of extrinsic motivation (Vinke & Schokker, 2001). On the other hand, (too) high levels of control resulting in work stress decreases intrinsic motivation (Bain & Taylor, 2000). In the interviews the respondents were asked to what extent they are motivated, what their most important motive in the work is, and what influences their motivation. They were also asked whether the control mechanisms influenced their work motivation.
It is common business in call centers to measure all kinds of performance indicators. Examples include number of calls, call duration, waiting times, number of sales (Taylor & Bain, 1999). The combination of technical and supervisor control helps disciplining the call center agents and increasing the level of professional service delivery (Lankshear et al., 2001). However, too much control might have a negative effect on performance (Callaghan & Thompson, 2001). In our study performance measures are based on the monthly reports on service times (call duration, waiting time, clearing time) per agent. In the interviews the respondents were asked to what extent the performance criteria were being used for appraisal. They were also asked to what extent the level of control had an influence on their performance.
Results
Based on the documents and interview data we can distinguish three groups of agents for whom the effects of control on performance, motivation and turnover intention differ:
- control has a positive effect on performance, motivation and turnover intention
- control has a negative effect on performance, motivation and turnover intention
- control has no effect on performance, motivation and turnover intention
We could not find an influence of control on absenteeism. None of the respondents reported to have ever been absent because of the control intensity in the call center. A common response was that
“control is just part of the work in any call center, there are rules to adhere to, but that is the same for any call center” (respondent #2).
On the whole, the level of absenteeism in this call center is rather low; 4.9% in 2008. That is close to the average absenteeism level of all Dutch workers (4.8% in 2005; de Grip et al., 2005) and much lower than the average absenteeism in Dutch call centers (approximately 10%; de Grip et al., 2005). Therefore, in the remainder we only focus on performance, motivation and turnover intention.
Although the actual level of control was the same for all agents, the effects of control on performance, motivation and turnover intention were different. The differences result from differences in the perceived level of control. As perceived control seems to be the most important distinctive characteristic, we can label these three groups as “the willing”, “the unwilling” and “the indifferent”. We discuss the effects of (perceived) control on performance, motivation and turnover intention per group.
The Willing
Four out of the fifteen interviewed agents experience mostly positive effects of control. A common response in this group is
“control is part of the deal, I’m used to it (…) it is a learning opportunity and an incentive to improve performance; not a means to be punished” (respondent #10).
As a result, their perception of control is positive. While these respondents are intrinsically motivated to improve the quality of their service, they experience control as a means to improve their performance. Supervisor feedback based on the logged calls and screens is welcomed:
“If my work is being controlled, I become more focused; it motivates me” (respondent #1).
“If there would be less control, I know that colleagues who shirk would get away with it and I would have to work harder” (respondent #10).
“If control would be less intensive, I would be less motivated” (respondent #3)
From these responses we learn that control has a positive influence on the agents’ (extrinsic) motivation. Moreover, because of the agents’ motivation, they also deliver better performance. Knowing that supervisors can listen in inspires them to perform at their best and improve their calling skills:
“When calls are more often taped, you will pay more attention to your calling techniques. Then you will automatically do better, because you paid more often attention to it” (respondent #5).
For this group, unlike Strandberg and Wahlberg’s results (2007), control has no influence on turnover intention. They do not feel threatened by control and supervisor feedback. Some respondents would even like more feedback in order to improve their skills. None of them intends to search for another job if the level of control would increase:
“I will not look for another job when control increases, it is not a problem” (respondent #10).
As a result, for this group of agents, control has no influence on absenteeism and turnover intention, but it has a positive influence on (extrinsic) motivation and performance. They are intrinsically motivated to perform at their best and the amount of control and an increasing amount of supervisor feedback would stimulate them to improve their skills. Note that their perception of the level of control is low.
The unwilling
Six out of fifteen agents experience mostly negative effects of control. These agents pay a great deal of attention to (the level of) control and they perceive the level of control as high. A typical quote would be:
“I think it is terrible that calls are being controlled, especially since specific controls are not announced in advance” (respondent #14).
Because of the control intensity these agents feel little freedom and they experience pressure. They understand that control is necessary, but they think the level of control could be lower.
“Now I am used to the high levels of control, but is not a nice experience. (…) I know that I do a good job, that I do my best and that I am trustworthy” (respondent #11).
Some of these respondents dislike the level of control so much that they are actually looking for another job, or will start looking for another job if the level of control will increase:
“there is already a high level of control and it is getting worse. I hear a lot of colleagues complain about the control. It is too much. (…) if the level of control increases, I will definitely look for another job” (respondent #13).
These results are in line with the results of Strandberg & Wahlberg (2007) and Mulholland (2004) who found that turnover intention increases as a result of dissatisfaction about the level of control. However, the terms of employment (salary) and working relations with colleagues moderate the negative influence of control on turnover intention:
“the job is well-paid, the office’s accessibility by car is good and I have very nice colleagues” (respondent #15).
For these agents the high experienced level of control has a negative influence on their motivation, and an increase in control (e.g. when screens are also going to be monitored with the new system) will further decrease their motivation:
“even when you are one minute late they [supervisors] will appear at your desk. This works at the expense of job satisfaction. The most important investment is in job satisfaction. Satisfied employees are more motivated; that is something I do not experience right here” (respondent #13).
About the extra control possibilities in the new system, respondent #14 states:
“this is just another means to control us, which I do not like at all”.
As a result, these agents motivation decreases with increasing levels of control.