The Design Week survey published this week shows that companies working in design for the Internet are making a significant impact on the UK design economy. This picture is very different from my first observations on the scene in late 1996 when UK-spawned companies such as Webmedia, Hard Media, Online Magic, MetaDesign and Obsolete were experimenting with the concept of being an Internet design company.

Designers have moved from creating what were essentially glorified hypertexts with pictures to working at the cutting edge of the new economy, conceiving interfaces to complex products such as online banking, e-commerce shopping sites and share dealing systems. This has tended to put technology in the driving seat, a situation exacerbated by the current rush to market of our share option-fuelled dotcom twenty-somethings, leaving the people who are expected to use these products confused and frustrated. The reputation of design, currently on a high, will suffer if this situation isn’t reversed, and the potential of our brave new network society will also be severely undermined. Where do we start?

Design is a combination of creation and inspiration combined with rational analysis, problem solving and iteration. These latter elements need to be emphasised, not only because clients can engage with them more easily but also because the people who use the final product will benefit. The clearest examples of a measured approach to design for the Internet are usability and usability testing.

How do we approach assessing usability? “Does the product communicate a sense of its purpose, how to begin and how to proceed?” is the first question Santa Cruz-based designer Lauralee Alben asks. “Are the product’s features self-evident and self-revealing? How well does the product support and allow for the different ways people will approach and use it, considering their various levels of experience, skills and strategies for problem-solving?” (Read on.)

A product can only be considered to be usable (or unusable) in the context of its target audience and the specific tasks it is intended to be used for. That said we are all familiar with poor usability on the Web. Sites with graphic-heavy pages that are aimed at users on modems. Forms on e-commerce stores that force Londoners to fill in the county (or state of the union) in which they reside and then fail to flag up clearly the ‘mistake’ they have made. Navigation systems that reflect a companies’ organisational structure rather than the users’ tasks. Search engines that by default search only a limited part of a site and present any results you are lucky enough to get without the context needed to evaluate them. (For a hands on guide to poor Web usability see Jared Spool et al’s dated but still relevant ‘Web Site Usability: A Designer's Guide’.) (Read on.)

Poor usability undermines the overall all ‘quality of experience’ (a term coined by Alben) that design can bring to a product, and it is relatively easy to avoid. Clients and designers need to think clearly about their intended users (and do research where appropriate) so they can model their objectives and ‘headsets’, their way of working and the competing pressures they are under, as well as their level of experience and access to the Internet. (At Palo Alto-based Cooper Interaction Design these user-avatars are given names and talked about in person.)

Usability testing is the method for checking the extent to which the resulting design solutions work. (Note that it is usability testing, not user testing; users are never wrong, it is the design that is under scrutiny.) Testing needn’t be done on a finished design and generally works best when it is applied from the early sketch stage and throughout the design process. Neither does it have to be expensive or involve lots of users. According to globe-trotting usability guru Jakob Nielsen “elaborate usability tests are a waste of resources. The best results come from testing no more than five users and running as many small tests as you can afford.” (Read on.) It is remarkable what insights usability testing can bring, even with the most well-considered designs. As Nielsen observes: “The most striking truth… is that tests with zero users give zero insights.”

Getting usability and usability testing into the product development and design process is no easy matter, but organisations will save on the bottom line if they do it properly. Telephone and email support enquiries will be reduced, fewer shopping trolleys will be abandoned before the checkout and customers will get a stronger brand experiences resulting in greater loyalty.

We should see usability as part of a wider set of skills and disciplines that make up the profession of ‘experience design’ (a term promoted by the US-based Advance for Design forum [now Experience Design community of interest]) and it shouldn’t be championed to the exclusion of those other elements.

Neither should usability be used as are reason to ‘dumb down’ design to some lowest common denominator in pursuit of universal accessibility, unless a product is really intended to be used by everyone on the planet.

To their credit a number of design companies in the UK have made usability an integral part of their process. Most bring in outside experts and a few have gone as far as creating usability laboratories. If other companies put usability further up their agendas they will win respect and influence with clients and help justify design’s reputation in the wider world.