Getting to Know Your Audience

Essays in this series about understanding your audience:

The Essential Audience

Most sites are built for an audience, with the audience representing any person who will interact with, view, read, navigate, search on, order from, submit to, download from — and so on — the web site. Since practically every measurement of a web site’s success involves the audience — hits, sales, click-throughs, customer comments, customer complaints, positive community, etc. — then it makes a great deal of sense to develop a deeper understanding of your audience and use this understanding to design a site that accommodates your audience.

Generating an understanding of who your users are, what they want to do, and how they tend to behave doesn’t necessarily have to be a complicated process, but it does return the best results when approached systematically and consistently.

Looking Closer at Audience Composition

The term audience is useful as a conceptual label, but the truth for electronic commerce sites — and any large-scale web site — is that audience is a very high-level term describing what is in fact a range of hard-to-define groups and sets of users, each with their own goals, motivations, and tasks. Your audience is unlikely to be homogeneous or monolithic, and you’ll probably have difficulty trying to pigeon-hole parts of your audience into convenient “like” groups.

A natural step in understanding an audience is the grouping of individual users into sets of users, according to a classification scheme. This process of categorizing users is useful because a mass of individuals is more easily understood if generalizations about behavior and goals can be made: dealing with a few trends is easier than dealing with many individuals.

The problem with categorizing users is that the point-of-view of the people doing the categorizing plays a huge role in how users are defined and divided. Major commerce sites are built by teams and departments that are coming from different places concerning users. Programmers may tend to think of users along certain lines, while marketers will probably have entirely different views. For a successful quality assurance process, you need to find some more objective way of describing users based on their characteristics, not on what your expectations of them are.

The Observable Audience

Beyond understanding the motivations or goals of your audience, you must also understand the basic mechanics of how your users will interact with your site:

  • How will the site render on a user’s client?
  • How will the user view the site? Visually? With graphics enabled? With a Braille reader? With a screen reader?
  • Will the user’s client be able to support all of your site’s functionality?

The best place to start digging for information to answer these questions is to look at statistics on browser, operating system, and computer platform use. All web servers can record examine and record the client application’s user-agent information, which is sent by the client with each HTTP request header. A good strategy would be to look at statistics from web sites that gather user-agent information across a network of participating servers, because this information will be more general and will give you an idea about what’s generally available and popular, and then look at your own site’s server logs to see what your specific audience members are using.

User-agent information is very valuable, but is also often ambiguous. User-agent information for any client application reflects what the programmers thought to include as a description of the application. This usually is a brief indication of the browser name and level, the operating system, the language, and various other descriptive elements. The problem is that there is no standard explaining how the information should be formatted, and no standard describing common definitions of terms, browsers, or operating systems. In addition, many applications don’t describe themselves, but instead use the description of another application; for example, most Microsoft Internet Explorer versions spoof themselves as being Netscape browsers. So, some statistics may not make sense, and should be discarded or combined with another statistic category.

[ Check out a sample user-agent log from philosophe.com. ]

With that warning, you can gather from server user-agent logs important information about common operating systems, common browsers, and common computer platforms. This information is useful to your quality assurance process in two ways: first, it shows you what kinds of environments your site should be designed for, and second, it shows you what environments you should be testing the user experience with.

Identifying Your Audience’s Operating Systems

Identifying and understanding the operating systems on your users’ machines is important to maintaining a quality user experience.

User-agent information will tell you what operating systems are being used by your audience. Even though the web and its component languages and protocols are intended to be platform independent, the web is dominated by designers valuing appearance over information structure and relationships, with the result that the “look and feel” of a page is usually more important than the structure of the page’s information content. Operating systems are not identical in terms of their graphical interface rendering capabilities or their support of data transfer and processing functions.

Once you know which operating systems are being used by your visitors, you can test the important browsers on those operating systems, and you can include support for these operating systems in the design and code process for your site.

Identifying Your Audience’s Browsers

After operating systems, browsers are the next bit of important information to gather about your users. While most of your audience will be using one of a very small set of operating systems, there are many, many more browsers and browsing applications available to users. Many of these browsers will not support the same feature sets, HTML versions, or even the basic functionality we expect of user-agents.

Server logs will show which browsers are being used by your audience, but you should approach the process of identifying important browsers a little more systematically.

Using Your Understanding of Your Audience

The work and research into understanding your audience(s) will allow you to improve your site in several ways.

First, knowing the users’ requirements and environments enables your team to design for the audience, which will foster a better user experience.

Second, an objective understanding of your users will allow a smoother process among all teams and departments for maintaining and enhancing your site because you’ll all be more likely to agree about who the users are and what they want.

Third, you will be better able to write use cases and test plans because you’ll have a more realistic and consistent understanding about user goals and tasks.

And finally, you will have enough information to create a suite of test environments that reflect what your users will be using. In turn, a test suite will allow you to consistently and accurately test the ability of your site to accommodate your user sets.

[ Read the next essay in this series about understanding your audience, Categorizing Users.]