INFORMATION DESIGN: DEFINING BY DOING
Information designers and experts in content complement each other in a lengthy process of analysis designed to ensure the optimum way to create effective information material. User experience is also considered crucial to define the visual results of graphic information.
Most people I meet have little or no idea what information design is, or how it differs from graphic design. The result is often confusion - ironically, something that information designers try to eradicate. So I m going to try to eradicate at least some of the confusion by explaining what information design is, and how it differs from graphic design.
My colleagues and I at the Communication Research Institute do a great deal of information design, but in an unusual context. We are a not-for-profit research group whose primary goal is to take existing design practices, review them critically and conduct research to find ways to improve them for the common good. Thus as we design, we also experiment, research, and reflect on the effectiveness of what we do and its social purpose.
Our "laboratories" for this work are the many organizations that become CRI Members to support our work and gain the benefits from it. With over 250 members over the last 20 years, we have had a lot of "laboratories" to experiment in, and we have made some fascinating discoveries, some of which have led to improvements in practice. We are now at the point where we think the professional practices we and others have developed are sufficiently stable and mature to be shared widely with our fellow designers. Normally, it would take some time (possibly 20 to 30 years) for these new professional practices to find their way into design teaching, and from there into design practice. But I am impatient. I would like many people to be doing this type of work NOW. The need for it is overwhelming, and the demand is not far behind.
What we design and why
Most of the things information designers design are part of ordinary life: forms, bills, contracts, labels, instructions, waygiving systems, web sites, text books, etc. We design them so that they can be easily used.
Our work is often at its best when it is "invisible" - people use the information we create to perform ordinary tasks easily, without even realising that the information they are using has been "designed". Our job is not to stamp our identity on our designs; on the contrary, it is to lend a quiet dignity to the tasks of ordinary life without in any way coming between people and these tasks.
We think of our work as a public service, rather than as an opportunity to express our creativity; though we are not averse to being creative when it is appropriate.
Taken together - the things we design, and our motivation for doing so - mark us apart from many, though not all, graphic designers. But the differences do not stop there.
How we work in the studio
Graphic design is almost wholly a studio-based practice which mainly focuses on the graphic aspects of documents, leaving the text to other professionals. Indeed, many graphic designers are trained with dummy text to enable them to focus on the "look" of the text without dealing with its content.
In contrast, information design is only partly a studio-based practice, and what we do in the studio is different. In our studio we design layouts, much as graphic designers do, but we also structure the content and write the text. Sometimes this is done as a team, with a writer and designer working together, but more often than not the person doing the designing is also doing the writing.
The reason for this integration of writing and graphic design is because we have found through research that this is the best way to create documents that are easily usable. Moreover, readers try and make sense of documents as a whole and don t separate text and graphics in their reading. We take responsibility for crafting the total reading experience.
Thus in the "studio" an information designer uses skills in graphics, in writing, and in transforming the esoteric into the publicly accessible.
Studio work accounts for only about 10-15% of the time we spend on a project. What the research has taught us is that to have a successful outcome, we need to embed the studio work in a much more elaborate process.
The information design process
Early in our research into information design practices we found that design projects (both our own and others) for which we could collect data were unsuccessful either in part or whole. There were four main reasons for the lack of success. I will take you through each of these and the methods we introduced to make our work more successful.
1. UNUSABLE DESIGNS
Our studio-based designs were never as usable as we thought they would be. No matter how much effort, sensitivity, skill, and creative energy we put into our designs, we found that people had problems using them, often problems that we could not have foreseen from within the studio. Repeatedly we found that getting a design right first time was highly unlikely.
So we developed our own methods for diagnostically testing designs to identify the problems, fixing them, and then retesting them to make sure that the faults were no longer there. Our methods were, in a sense, clinical - much in the way that a doctor looks for symptoms, applies a treatment, then looks to see whether or not the symptoms have gone. We were very tempted to get other professionals more skilled in testing to undertake this work for us. But after many trials we discovered that the best people to do this work - most cost effectively, and achieving the best results - were the information designers responsible for creating the designs. This meant that we had to expand our skills and undertake a proportion of our work outside the studio.
We began experimenting with this iterative process of designing, testing, and refining in the mid 1980s. At first it was slow, expensive, and cumbersome because we were dependent on phototypesetting and offset printing to create each new version of a design for testing. But from 1985, with Apple Macs and laser printers, we were able to massively speed up the process and reduce the costs; what would take weeks with hundreds of dollars printing costs per iteration took hours with negligible printing costs. And now, with Internet-based designs for the Web, the costs of iterative redesign and creation of new prototypes is even cheaper.
Because testing is new to many designers, it is often greeted with scepticism and many questions: is it hard to do, must I test lots of people to get useful data, won't it inhibit my creativity? I will deal with each of these briefly.
Is it hard? We have found that it is possible for designers to learn the basics of testing in a simple two-day intensive training program. As long as this is followed up by supervised practical work, most designers can become proficient at this type of work fairly easily.
How many people do you need to test? The answer is none! You are not testing people, you are testing designs to find design faults that you can correct by better design. Thus, regarding participant numbers, we ask a different question: how many people do I need to help me find all the faults in a design? The answer is: the first 5 or 6 people you invite to use the design will identify most of the faults. To err on the side of caution, we often invite 10 people to participate in this type of testing, and we choose people who we think may have some difficulty with the type of information we are testing, so that we maximise our opportunities for finding faults.
Does testing inhibit a designers creativity? Far from it; indeed, the very reverse. The first time you perform diagnostic testing is like an epiphany; for the first time you get a glimpse of what it is like for someone else to use your design. This is a profound moment of insight and revelation. Few look back. Moreover, the interaction between other people and your design leads to fresh insights and often to different and better ways of designing information. In other words, testing leads to more, not less, creativity and originality.
The lessons from testing are many, but two stand out. First, we can now achieve a much higher standard of design from a user s viewpoint than has ever been possible. Moreover, we can support that claim with evidence. Second, through testing we have developed a much clearer sense of how to formulate information design briefs and demonstrate to clients that we have met their brief, and that they have gained substantial benefit from our work. This is worth elaborating.
Testing is highly specific. To use an analogy from medicine, if you want to have your blood tested, you have to be quite specific about the kind of test you want: is it for cholesterol, liver function, blood sugar? The same is true of diagnostic testing in information design. So we work jointly with the client and other interested parties to draw up a list of things we should test for. Most commonly in functional documents like labels or bills, it is a list of tasks (performance criteria) that we have agreed people should be able to perform with the document. The testing then consists of asking people to perform these tasks, and finding out whether or not they can perform them and how well or badly. If people can perform these tasks well, we have done our job. If they cannot, it s back to the studio for another iteration. In the end we can demonstrate that we have met the brief.
2. IMPRACTICAL DESIGNS
Even when we achieved successful designs for the people we cared about most, the eventual users, we kept discovering practical constraints - usually after we finished the work - that had major impacts on the implementation of our designs. Even if we had been thoroughly briefed by a client, we found that constraints originally considered unimportant could be critical to the outcome. We started to itemise these constraints, probing our laboratories to find out where they were and how they might affect outcomes, and found patterns of constraints across the various organizations. Today, we have a check list of some twenty items that we routinely investigate in an organization before we commence work in the studio. These investigations take place during the Scoping of our process. Again, they take us out of the studio and use about 15% of the total effort.
To feed at the hand of capitalism you must demonstrate that you have added value by providing a Return on Investments (a ROI). For many designers, this is the Holy Grail of professional design practice in a capitalist system.
At CRI, we did not specifically set out to discover the Grail, we just wanted to help people through their daily drudge; but we found it, and the discovery led to benchmarking.
Usually, clients ask us to redesign an information document because they are vaguely aware that it is not working. Some years ago, before starting on any redesigns, we decided to find out how well or badly the information worked; so as part of the Scoping stage, after we had agreed on our performance criteria, we decided to test the existing designs. The data from such testing was often quite dramatic: 100% of forms filled out incorrectly, fewer than 50% of customers able to use their phone bill to perform simple tasks, fewer than 50% of patients able to follow the instructions for using a medicine bought from a supermarket. Such poorly-performing information was the norm, but it had never been measured before, and no one knew exactly how bad the designs were or how costly it was to industry.
This testing is now routine. The testing done before a redesign shows us where we are now, and tells us how much improvement is needed for a design to match the agreed performance criteria. We have called this testing Benchmarking. It provides data which is compared to the data from testing after the redesign; the difference between the two is used to calculate the ROI. So when we tell a client that our new design has made an improvement - that we have added value and provided an ROI - we have the evidence.
Benchmarking has another useful function: it provides us with many clues on how the new design might make the information more usable. This then saves us time and effort later, leading to fewer iterations in testing and refinement - one of those moments where economics and good design practice coincide. Benchmarking accounts for about 10% of our project time.
Given our careful scoping, benchmarking, integration of text and graphics, testing, and refinement, one might think that our work is successful more often than not; but this is not necessarily the case. The major risk to a successful outcome is, and always has been, the politics that surrounds the design process.
Even such humble organisational documents as forms - the beasts of burden in the information society - are subject to the tugs and pressures of sectional interests within organizations. None of this is apparent until you try to change them. Suddenly, as if from nowhere, erupt a host of clamouring voices: the marketing department wants to impose its branding requirements and marketing messages onto the design; the information technology managers insist that the order of information cannot be changed because of their processing software; the legal department won t let you change the wording for fear of exposing the company to litigation; the call centre is worried that any changes will lead to additional customer complaints or inquiries; and so it goes on.
We contribute to these voices by representing, often for the first time, the interests of the consumer or citizen. And this new voice at the table - an outsider - makes quite different demands. Managing all of these different voices to ensure an outcome that serves the outsider's interests as well as the internal constituencies takes up the remaining 50% of our project time, and requires skills of statecraft, diplomacy, and negotiation.
In the end, we are left with a question. Why does information need to be constantly redesigned? We discovered that even well designed information deteriorates quite rapidly. In 1988 we redesigned a phone bill so that 100% of literate customers could use it appropriately. By 2003 the same design could only be used appropriately by 42% of literate customers. The social environment in which such documents have to work, like most of our social arrangements, is highly changeable. For this reason we introduced our final stage, Monitoring, where we can keep an eye on the changes and recommend action where it seems appropriate.
This then has been my attempt to remove some confusions about information design and how we have developed the craft at CRI. I end with a plea. People are routinely disenfranchised because forms and other public documents are difficult to use. People get frustrated and lost when they cannot find the information they need. And people suffer or die because they do not have usable information on the medicines they take. I invite you to join me in the important work of information design.
All of the work mentioned in this paper can be found at our website.
About the article
This article was orginally published in tipoGrafica magazine 71: year XX, June, July, 2006, pp. 08-15. It is reprinted with permission.
David Sless is an information designer specialized in the critical review of the graphic communication of everyday objects by means of research conducted on daily practical communication problems. Director of the Communication Research Institute and professor of Science Communication. Advisor in corporate communication, and guest lecturer in the US, Europe and Asia. In addition, he is the author of over 180 publications.