User-centred design means not only designing for users, but also designing with users (Kontogiannis & Embrey 1997). Pheasant (1996) defined the principle of user-centred design as follows: if an object, a system or an environment is intended for human use, then its design should be based upon the physical and mental characteristics of its human users. The user-centred design process is based on user feedback and user participation (Koivunen 1994). Feedback from users reduces misunderstandings and improves the quality of the user interfaces. User involvement is important especially at the early stages of design (Koivunen 1994). The R&D and marketing departments of companies use a number of methods to probe their customers’ minds and reactions and to elicit feedback and to use it in further development (Tang 1991, Koivunen 1994). For instance, observations of users using the system in their normal work give a lot of information on the frequency of the performed tasks and the problems related to those tasks. Also, the efficiency of the system can be measured by measuring the time spent on various activities.
User participation should be comprehensive and not simply part of an end-process evaluation procedure, which seeks to identify design faults (Catterall et al. 1991). It should be part of an evolving and iterative design process. Users may, for instance, be actively involved in the specification procedure and in rapid prototyping trials. User involvement in the process also affects the success of implementation (Eason 1988, 1995).
User-centred design descriptions include four basic steps already defined by Gould and Lewis (1985, quoted by den Buurman 1998):
know your users,
incorporate the current knowledge of the users in the early information stage of design,
confront the users repeatedly with early prototypes for evaluation purposes, and
redesign as often as necessary.
Designers are often engineers, who know their products in and out and find them easy to use. What an engineer believes to be the knowledge, experience, and needs of the target user population, i.e. the objective view, may be very different from the subjective view of the actual end-users (Bogner 1998). It is almost impossible even for an experienced designer to think of all the different ways people understand the cues a product gives about its intended way of use and to think of the different ways people actually use products. This discrepancy can be corrected by involving actual, typical, if not worst-case, users in design discussions and in ongoing testing during product development. When users are incorporated in the whole design process, design errors are detected and corrected before they propagate and become expensive or impossible to correct (Lanning 1991). Errors can be as much as 50 times more costly if not corrected early (Boehm 1981). Further, Keinonen et al. (1998b) pointed out that ”user tests reveal patterns of behaviour and problems with the use, which the designers would never have come to think of without the test. By acting with a prototype, the user is able to communicate in a natural way. There is no need for the user to master the design jargon.”
Lanning (1991) has divided the user’s role in user-centred design into three types: the user can be a subject, an evaluator, or a designer. The user, as a subject, is a creature to be studied, whose needs should be satisfied. As an evaluator, the user is a living tool used by professional designers to measure the user-perceived value of design alternatives or design decisions. The evaluator may be used to predict the acceptance levels of the final design and to provide performance, preference and intent-to-purchase information. Experimental usability and utility studies are typically made to collect evaluator information. The user, as a designer, is a member of the design team. In this role, the user actively makes strategic design recommendations, suggests alternatives, and helps to define acceptance criteria. Users, as designers, can help to select relevant background information, alternatives to be considered, and alternatives to be selected. Users, as evaluators, help to ensure that the design process is producing user-centred designs by validating the selected alternatives.