Is the customer always right?

The latest advertisement campaign from Microsoft highlights how their new operating system, Windows 7, has been developed based on user feedback. Each advert centres on a user who introduces the audience to a new function that they thought of and communicated to Microsoft. Although clearly intended to be light hearted, these adverts raise an interesting question, ‘is the customer always right?’, and how much should we rely upon what they tell us? Turning to Window’s main competitor, Apple, the story could not be more different. As Steve Jobs is keen to point out, Apple does no market research. Instead, they rely on a small team of designers with a detailed domain understanding to develop their products. This begs the question, what is the role of end-user engagement in product or system design?

In the 1950s Henry Dreyfuss, helped to publicise the importance of fitting products to people, instead of the other way around. The ‘user-centred design’ movement gained momentum as consumer electronics became cheaper and widely used in everyday products. In the twenty-first century, the ease of use of websites and consumer products has proved to be a key factor in consumer choice.

As Human Factors specialists, we are often asked to ‘enhance’ existing products – to make them ‘more intuitive’ or to ‘enhance the user-experience’. Engagement with the user community often provides us with potential solutions from the outset. With limited resources, it can be tempting to rely upon these individuals, to follow them blindly. Moving icons, adding functionality, all tailored to a specific individual and task. While there may be times when it is appropriate to follow this guidance, for many complex sociotechnical systems the solution is not so simple. What may work for a specific individual, performing a specific task, under specific circumstances, may not be appropriate in all situations. In larger systems, end-users rarely have a complete understanding of how the organisation is structured or the implications of this on systems performance. Nor necessarily should they if it does not directly influence their primary task.

To leverage significant system enhancements, the role of human factors practitioners should go beyond optimisation of the man-machine interface. Whilst this may remain an integral part of system optimisation, there are other factors at work, such as organisational design, that are also worthy of our attention. While many will consider this blatantly obvious, it is not universally acknowledged or accepted. Much of Apple’s success has been attributed to their meticulous control of every aspect of the system. They control the hardware, the software and much of the supply chain.

Considering large socio-technical systems, the diagram above highlights many of the aspects that should be considered in design. Each level of the diagram offers opportunity to enhance overall performance. As the centre of the diagram shows, overall system performance can be changed through component design or by providing the end user with additional skills. These can be viewed as two separate components, or they can be considered as a human-machine subsystem. Stepping out further, the system can be viewed at the team level. Here we can start to consider and model how the design of a system influences the way a team functions. Stepping out further still, larger, more extended, networks can be considered. Finally, at the outermost layer, we can start to consider the wider domain independently of the system currently in place. We can consider its overall purpose and how we might assess this.

In summary, effective design is more about domain understanding than end-user engagement. Clearly, talking to users can help in developing this understanding; however, it should not be used as the sole design driver. Effective systems design should consider the physical and functional elements of a system along side the social and structural ones. It is contented that this approach will lead to safer, more effective systems.