Communications of the ACM
Communications of the ACM
The Status of Helpdesk Support
Helpdesks have long been the standard source for end-user computing support. Though they are also called information centers, call centers, and PC support centers, their primary function of helping users help themselves has not changed. In the early days of PC computing (late 1970s and early 1980s), end users viewed helpdesks as a godsend. But today, that attitude seems to have changed, as computing enters the new millennium. Academic studies consistently show that end users are dissatisfied with helpdesk support [1]. This dissatisfaction has driven end users toward other support sources, including local MIS staff, friends, and online help. In a survey conducted in the spring of 1999, I sought to identify the support services offered by helpdesks and determine whether end users actually use these services.
A questionnaire was designed by modifying the instrument developed by Mirani and King [3] (see the sidebar "Instrument Development and Data Collection"). I collected data from 197 helpdesks, mostly in the U.S., representing a variety of industries. Results showed that end users used helpdesk support only minimally. Despite this lack of interest by the people they are intended to help, helpdesk support managers can still control key computing resources, while increasing localized support for end users suggests the need to restructure helpdesks.
Results
A total of 209 responses were received, of which 197 were usable. The figure categorizes respondents by industry type; most were from educational institutions, medical firms, and manufacturing firms. The age of the helpdesks ranged from less than a year to more than 25 years. The responses were aggregated from all 197 helpdesks to help identify the most popular services. Moreover, to study whether these services varied between novice and veteran helpdesks, the responses were divided into three groups based on helpdesk age. I conducted similar analyses by dividing the helpdesks based on staff size and number of employees supported. The main support services offered by helpdesks can be grouped into 11 categories (for details and percentages, please see ecommerce.lebow.drexel.edu/CACM/Helpdesk.html).
Compared to previous years, as reported by respondents, more helpdesks in 1999 offered services in certain categories, including application development support, standards and guidelines, variety of software supported, and post-development support. However, over half of these helpdesks had never offered services in application development, operational support, and variety of software support. At the same time, over half of them offered services in data provision, purchasing, staff characteristics, Y2K (a critical support service at the time), and telecom.
It should be noted that many helpdesks did not support application development, despite it being a type of service that made helpdesks popular among end users during the early days of end-user support. Related categoriesoperational support, post-development support, and variety of software supportedalso seemed to be low-priority offerings among the surveyed helpdesks. This gap between availability and demand might have been due to the transformation end users have undergone, as personal computers became standard office hardware for everyone. Prior research showed that data support was the main priority for end users [4]. My findings indicated that helpdesks recognized this need and offered the service. Though many helpdesks did not give support to users acquiring and managing corporate data, some data-related services were being offered by some helpdesks at the time of the survey. For instance, support services maintaining data integrity and security and listing companywide data resources were offered by more helpdesks at the time than by their counterparts in the early 1990s. The surveyed helpdesks also seemed to have adequately covered end-user concerns on Y2K issues ahead of the date change January 1, 2000.
The most interesting aspect of the results was that end users tended to use helpdesks only minimally, possibly due to the existence of local support staff in each department supporting only users of that department. Compared to about 4% of the helpdesks offering this service earlier, 81% in 1999 was a tremendous increase. Local staff supporting one or a few departments might better understand the related business functions and be more knowledgeable about the computing requirements and software used in the departments than the staffs of centralized helpdesks.
Helpdesk services based on helpdesk age. Responses to the survey were organized into three groups based on helpdesk age to identify the differences in services offered: level-1 (up to three years, or 37%); level-2 (three to six years, or 34%); and level-3 (more than six years, or 28%). Table 1 outlines the most popular services with differences and/or similarities among the different groups. More than half of the helpdesks in each group offered certain services in the support category of standards and guidelines. For example, in data provision support, while there are no major differences in items involving data integrity and security and data backup/recovery service, more level-2 helpdesks offered items involving companywide data resources and maintaining subject databases. Understandably, helpdesks in each group appeared cautious in providing corporate data-extraction facilities.
Data sharing among users seemed to be a popular offering among all groups. More level-1 and level-2 helpdesks offered operational support than the more experienced level-3 helpdesks. Again, this might have been due to the transformation of the user's role, as end users might need more advanced than basic support. A large number of helpdesks in each surveyed age group offered the service categories of support staff characteristics, Y2K support, and telecom support. Table 1 shows that more level-3 helpdesks offered crucial services, including maintaining data integrity and security, providing data backup/recovery, and online support, than level-1 and level-2 helpdesks.
Helpdesk services based on helpdesk size. To study the services provided by different-size helpdesks, the 197 responses were divided into three groupssmall, medium, and largewith staff sizes of fewer than six, six to 15, and more than 15, respectively. Table 2 shows that services provided by all sizes of helpdesks are similar to the results in Table 1. However, more small- and medium-size helpdesks offered data provision support than larger helpdesks, specifically for maintaining data integrity and security, providing data backup/recovery service, and facilitating data sharing among users. Such huge differences were not apparent for this service when surveyed helpdesks were compared by age. One can assume that the size of a helpdesk is directly proportional to an organization's size, which in turn may be related to the need for extra security for corporate data. Hence, users in larger organizations may have to deal with more restrictions on data access. Applying the same reasoning, it is understandable that more medium- and large-size helpdesks than smaller helpdesks provide telecom support, including online support, and help with mainframe access and telecommunications hardware/software. The number of small- and medium-size helpdesks providing operational support is greater than the number of larger helpdesks providing operational support, similar to the data in Table 1, which showed more level-1 and level-2 helpdesks provide this support than level-3 helpdesks.
Helpdesk services based on number of employees supported. I also analyzed helpdesk services by dividing the respondents into three groups based on number of employees supported (for more detail and statistics, please see ecommerce.lebow.drexel.edu/CACM/Helpdesk.html). They were: small (up to 500 employees), medium (500 to 2,000 employees), and large (more than 2,000 employees). The number of helpdesks offering application development support decreased as the number of employees increased. The same appears to hold for many services in the categories of standards and guidelines and operational support. Fewer helpdesks supporting large numbers of users, that is, helpdesks of larger organizations, offered data-related services compared to small- and medium-sized organizations, in accordance with the earlier discussion. Results for the item of coordinating applications across users of post-development support indicated it is easier for helpdesks supporting fewer users to coordinate applications across users than helpdesks supporting large numbers of users. Finally, more helpdesks supporting relatively large numbers of users seem to provide telecom support than helpdesks supporting fewer users.
Implications
Besides inviting you to draw your own conclusions based on the numbers presented here, I would like to point out that several findings warrant special attention. The most interesting is that few helpdesks reported that end users actually requested help, and when they did, only for some services. The logical explanation for this gap between likely user need and the services helpdesks provide would be that end users might be using other forms of support, including local MIS staff support, as well as informal, online, and other external sources, including vendors. This conclusion is supported by the fact that most helpdesks offer online support (68%) and local MIS staff support (81%). Moreover, earlier research showed that users prefer decentralized support (local MIS staff) to centralized helpdesk support [2]. These results have implications for helpdesk management and structuring. For example, helpdesks can be redesigned to offer support for resources that are better controlled centrally, including corporate data, purchasing, and mainframe access, leaving the rest to someone else. Many helpdesks indicated they provided only basic support; in other words, users were redirected to the appropriate department or other helpdesks for additional help.
Another interesting finding is that most helpdesks seemed to maintain tight control over corporate data, an absolutely necessary function. End users are not trained computer professionals and might easily compromise data security and integrity. Though end-user computing applications have benefits, they also pose potential threats to the organization. Also interesting is that many helpdesks do not view application development support as critical. Many also ignore other support services related to application development. Possible explanations include end users already having thorough knowledge of application development techniques, though this explanation is unlikely, as most end users are not trained in application development, and end users (more likely) relying heavily on other forms of support.
Conclusion
When I conducted the survey in 1999, helpdesks seemed to offer some data-related services, online support, Y2K support for end-user applications, purchasing support, and telecom support. At the time, support for application development and related support categories seemed a low priority for helpdesks. However, I also offer a word of caution; since 17% of the survey responses were from educational institutions where application development by end users may be less apparent than in other organizations, the survey results might have been skewed. Still, the results do reflect the status of helpdesk support. My data showed that helpdesks are cautious in offering data-related services. Just because end users seek only minimal support from helpdesks does not mean they do not need support. End-user computing applications can be a potential source of threat to any organization with respect to maintaining data integrity and security. Therefore, resources should be redirected toward strengthening other support sources, including local MIS staff and online options, so end-user computing is as useful as possible to the organization.
References
1. Govindarajulu, C. End-user computing support: Status and directions. In Proceedings of the 1996 Southwest Decision Sciences (San Antonio, TX, 1996).
2. Govindarajulu, C. and Reithel, B. Beyond the information center: An instrument to measure support from multiple sources. Inform. Mgmt. 33, 5, (1998).
3. Mirani, R. and King, W. The development of a measure for end-user computing support. Decis. Sci. 25, 4 (1994).
4. Rainer, K. and Carr, H. Are information centers responsive to user needs? Inform. Mgmt. 22, 2, (1992).
©2002 ACM 0002-0782/02/0100 $5.00
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
The Digital Library is published by the Association for Computing Machinery. Copyright © 2002 ACM, Inc.
No entries found