Often when clients have a relatively low budget for usability testing, or a short amount of time in which to conduct it, an ‘expert’ or an ‘heuristic’ review will be run by an experienced usability practitioner. There are slight differences between the two, with the expert review entailing a less formal evaluation process than the heuristic review. But all things considered, they’re pretty much in the same ballpark time wise… So I suppose that means cost wise too.
There are a number of advantages in conducting either type of review. As I mentioned above, where resources are limited, they can be an effective and efficient method of assessing a site; the time and cost associated with recruiting, interviewing and paying test participants is negated. These evaluation methods can also be conducted very easily and consistently throughout the life of a site, providing a benchmark as well as a periodical health-check.
In conducting an heuristic review, a series of guidelines or checkpoints is used by the usability expert to assess a site (or application). In conducting an expert review however, these specific guidelines may not be utilised, with the practitioner relying on their expertise of general usability principles to review the site at hand.
Arguably, the most well known usability heuristics are those developed by Jakob Nielsen, who outlines ten rules of thumb. There are numerous variations of Nielsen’s ‘standard ten’ including an extremely detailed checklist developed by Deniese Pierotti for Xerox. Pierotti’s heuristics use Nielsen’s ten rules as a foundation, but then goes further by providing individual checkpoints specific to each rule. It’s definitely an extensive list of usability checkpoints, and one which I expect was developed for a particular environment.
The thing is…
I believe there are a whole lot of things that haven’t been considered in these lists. If we’re going to take a best practice, user centred, holistic, may the force be with you approach to web development, surely there should be a broader focus; and one that incorporates as many areas of front end development into an heuristic review as possible.
In 2004, Louis Rosenfeld published a list of heuristics for the IA where he said:
Every information architect should always have a set of favourite questions in their back pocket; they really do come in handy.
I categorize mine into groups that correspond to the five areas that a user is most likely to interact with a site’s information architecture:
- Main page
- Search interface
- Search results
- Site-wide navigation
- Contextual navigation
A couple of weeks later, he expanded on these by developing a series of heuristics specific to Search Systems . These are very specific, and don’t necessarily apply to every site or application. But what they do offer is an opportunity to build a more tailored and well rounded set of web usability heuristics. In summary, the search system heuristics focus on the review of:
- Locating search
- Scoping search
- Query entry
- Retrieval results
- Query refinement?
- Interaction with other IA components
- Finishing search
Both articles offer far more detail defining each heuristic with a series of related questions. They are essential in understanding how IA heuristics can build on the standard ten that have been churned about for long enough.
- Does the site use (script based) pop-ups?
- Does the site use device independent scripts?
No doubt there are others that could be included here, but I think it’s a good start… Can you add to this?
The Web Standards Developer…
The quality of front end coding (CSS and HTML) also has a significant impact on site usability and should certainly be a consideration when developing a set of usability heuristics. ‘Coding Heuristics’ might include questions such as:
- Does the site incorporate a favicon?
- What is the naming convention used in title tags?
- Is the site coded semantically?
Each of these heuristics improves the general usability of a site for all users, but there are other, more specific coding techniques that will improve site usability and accessibility.
Quite a few of the accessibility guidelines outlined in WAI WCAG 1.0 also improve site usability for users who are not disabled. In addition to those above, I will typically include these as coding heuristics:
- Does the site implement fieldset and legend?
- Does the site use frames?
- Is colour contrast adequate?
- Are visited links obvious?
- Are link destinations clear?
- Is the file type of downloads obvious e.g. pdf’s?
- Have relative font sizes been incorporated?
- Can the site be read with style sheets disabled?
- Does the site validate?
I’m not suggesting that this would negate the need for a complete accessibility audit. Clearly that is a stand alone event. But in conducting an heuristic review of sites usability, these are elements that directly affect all users.
This list is by no means exhaustive and in some cases it might be appropriate to refine these usability heuristics so that they’re more in tune with your own approach. But in conducting an expert or heuristic review, we need to think a lot deeper than the GUI. We need to move forward from the ‘standard ten’ and incorporate heuristics that address holistic development and user centred design. Beauty is, after all, only skin deep!
Making Computers-People Literate. Copyright 1993. By Elaine Weiss ISBN: 0-471-01877-5
Usability Inspection Methods. Copyright 1994.By Jakob Nielsen and Robert Mack ISBN: 1-55542-622-0