[Excerpts from "ATM Design and Training Issues" by Wendy A. Rogers & Arthur D. Fisk, in ERGONOMICS IN DESIGN, Volume 5, Number 1, January 1997. Copyright 1997 by Human Factors and Ergonomics Society, P.O. Box 1369, Santa Monica, CA 90406-1369 USA; 310/394-1811, fax 310/394-2410, http://hfes.org. To obtain a copy of the entire article, which includes full references, contact Lois Smith at hfes@compuserve.com or the address above.]
The concept of automatic banking is deceptively simple: You walk up to the machine, enter your card, input your personal identification number, and then do your banking. However, as many of us know, things can (and do) go wrong. Automatic teller machines (ATMs) were designed to increase the flexibility of banking for customers (and to save banks money). When you open a new account, most banks will give you an ATM card, a personal identification number, and sometimes a pamphlet describing how the system works. More often than not, however, bankers assume that you can learn how to use the ATM on your own.
One goal of human factors professionals is to ensure that if individuals wish to do something with a system (e.g., use an ATM), sufficient analysis and design have been carried out to remove impediments to successful system interaction. Thus the goal of our involvement with ATM systems has not been to increase ATM use per se. Instead, it has been to ensure that if people wish to use an ATM, the system is designed to enable them to do so efficiently. Also, we are interested in giving potential users the opportunity to obtain sufficient information and/or training.
ATM use is likely to increase in the near future as the available functions are expanded. Already, ATMs offer 24-hour service and are conveniently located. They can be found in bank lobbies, drive-up windows, shopping malls, grocery stores, airports, universities, hospitals, convenience stores, service stations, hotels, and other tourist areas (Riffelmacher, 1991; Schmitzer, 1984). Most current ATMs enable cash withdrawals, deposits, transfer of funds between accounts, and balance inquiries. Functions that have become available more recently, or are available in limited areas, include
- purchasing U.S. postage stamps (Riffelmacher, 1991;Strachman, 1994),
- obtaining stock information ("Citibank," 1994),
- applying for an automobile loan (Melia, 1991),
- receiving a detailed bank statement (Leander, 1990),
- purchasing cinema tickets (Riffelmacher, 1991), and
- opening an individual retirement account (Melia, 1991).
An additional incentive for using an ATM comes from the recent trend to charge bank customers a fee for using a human teller (Meier, 1995)!
As a result of the increased use of ATMs, their developers are faced with a challenge. A recent survey of 1475 ATM cardholders in the American Southwest revealed that 70% of the respondents claimed they would not open an account at an institution that did not offer access to ATMs (Iida, 1991). Thus, access to convenient and reliable ATMs has become a criterion for choosing a bank, and financial institutions will need to optimize their ATM functions, especially from the perspective of the user.
Who Uses ATMs?
"Know thy user" is a basic tenet of human factors. Several large-scale surveys of ATM users have been conducted (El-Haddad & Almahmeed, 1992; Gilly & Zeithaml, 1985; Rogers, Cabrera, Walker, Gilbert, & Fisk, 1996). We focus on the most recent data of Rogers, Cabrera, et al., who distributed 9000 surveys in Memphis and Atlanta and received 1562 completed surveys (17%).
Relative to nonusers, ATM users are younger, are more willing to use other types of technology, earn more money, and are more educated. By inference, then, nonusers are less familiar with technology, have less formal education, and thus might require more intuitive user interfaces as well as instructions about how to use ATMs. Based on these data, one appropriate design strategy would be to de-emphasize the technological nature of ATMs in order to persuade more people to use them.
ATM Users' Difficulties and Concerns
Millions of people use ATMs. What kinds of difficulties do they have? In our survey, more than 10% of all respondents rated the following as problems that occurred frequently:
- The machine ran out of money.
- I forgot to record my transactions.
- I was unable to see the screen well.
- I put the card in the wrong way.
- I had to wait in line to use the machine.
- The machine worked too slowly.
In addition to these problems, older ATM users (aged 65-91) also reported frequently having difficulty getting the amount of money they wanted and understanding how to do what they wanted on the ATM.
The majority of these problems could be corrected by a combination of design changes and ATM training. New technologies currently being incorporated into the design of ATMs include touch screens, pen input, multimedia, and speech recognition (Henneman & Rubini, 1993), although these technologies may pose new problems. Larger screens, voice guidance, and Braille indicators are also being incorporated into some recent models of ATMs (e.g., Fassett, 1991) in an effort to comply with the Americans with Disabilities Act of 1990 (see also Melia, 1991). Advances in ATM training are discussed later in this article.
Providing Training
An important finding of the Rogers, Cabrera, et al. (1996) survey was that 21% of the nonusers said that they would be interested in learning to use an ATM if they were provided with training. In their interviews with older adult nonusers, Rogers, Gilbert, and Cabrera (in press) reported comments such as "I don't know how to use it and I don't want to embarrass myself" and "I don't like to feel rushed when I am trying to learn." Even ATM users believed that "the bank should offer training or a chance to practice with the machine" (42% overall, 52% of adults over age 65; see Rogers, Cabrera, et al., 1996).
Rogers, Fisk, et al. (1996) developed an ATM simulation that runs on an ordinary personal computer. They compared different training procedures for older adults (aged 60-80). One group of trainees received only a description of the ATM, another group received more detailed text instructions, a third group received a pictorial guide to the ATM, and a fourth group completed an on- line tutorial (see Figure 3). The experimenters assessed trainees' acquisition of skills on the ATM as well as the degree to which the participants could successfully transfer their new knowledge to a novel ATM.
Accuracy and transaction times were best for the on-line tutorial group, intermediate for the text and pictorial guide groups, and worst for the description-only group. These data demonstrate the importance of providing older adults with ATM training and the fact that the type of training influences the level of performance. The on-line tutorial, which provided specific practice on the task components, best facilitated performance, retention, and transfer.