16

UXers here probably audit existing websites or pieces of software that are already established. Sometimes, we are involved in designing them from the start, but for times when we are not, we just have to audit and list recommendations for the existing work.

In your UX/usability audits, do you follow standard procedures or do you just write down whatever UX or usability issue you see, then list your recommendations?

I'm thinking of making a standard usability audit. Right now, the quick things I can think of are Fitt's law, web page title, and visual consistency.

Should there be a standard process for usability audits? If so, what do you recommend to include in that standard process?

Let's limit this to websites to be more specific.

Improve this question
2

7 Answers 7

Reset to default
10

There are several heuristics you could check, such as Jacob Nielsen's. Or use normative rules and standards like the dialogue principles from ISO 9241-10. Those are helpful to look at different aspects of usability separately.
You should probably develop something like a checklist, instead of just randomly listing anything you found - chances are too high to miss something. Also, I find it better to not only list usability mistakes, but also things that work well.

Nielsen's heuristics:
http://www.useit.com/papers/heuristic/heuristic_list.html

... explained with examples from the web:
http://designingwebinterfaces.com/6-tips-for-a-great-flex-ux-part-5

ISO Principles:
http://www.userfocus.co.uk/resources/iso9241/part110.html

It seems I can not add a comment to my own answer, so in answer to Allans comment: I actually have no information on free information about the ISO standards -- their content is not free after all. I think any company that develops user interfaces on a regular basis should have them in their library and for company the prices are not too high. I recommend at least the 9241-10 and 9241-110, they have a lot of valuable information that I look into from time to time. It's another thing for freelancers or learners, where money is a greater issue. As a student I would maybe recommend looking into the university library or ask in design departments. Anybody else know good free places to learn about iso standards?

Improve this answer
1
  • Very useful info! As for the ISO standards, you think, userfocus is the most complete free source for this? :)
    – Allan Caeg
    Commented Dec 7, 2009 at 2:17
5

I came across a rather useful spreadsheet from Userfocus with 247 web usability guidelines. It's organized into several tabs, each covering a different category, such as Task Orientation, Navigation & IA, Forms & Data Entry.

I liked it as it gave some structure to the criteria I wanted to evaluate. The summary shows a score for each tab, as well as an overall score.

Improve this answer
2

I came across this 1-page, 25-point Website Usability Checklist (includes a downloadable pdf version) which you may find useful in terms of potential format and review points/buckets. It's not as comprehensive and detailed as the one Janel suggested above but might be useful as a quick checklist:

http://www.usereffect.com/topic/25-point-website-usability-checklist

Good luck!

Improve this answer
2

I maintain a short list of key heuristics and a collection of specific checkpoints designed to test a website against each of those heuristics. I choose the most appropriate combination of checkpoints depending on the type of website and its audience. I review those heuristics and checkpoints from time to time, adding and removing as appropriate. I took a couple of ideas from Psychological Usability Heuristics and Web Applications (RIA) Usability Heuristics this morning, in fact.

Improve this answer
1
  • This is the best answer on this page ^ Much more generally applicable and scientific.
    – kbwatts
    Commented May 13, 2016 at 5:19
2

Doing the Audit

I think the standard process you choose depends largely on the resources available for doing the audit. Bare audit could be something as simple as expert review done by one usability expert. The expert should at minimum base his evaluation at least partially on some heuristics/criteria as there is evidence that this leads to more problems found (Bastien & Scapin, 1995). Regarding heuristics Nielsen's heuristics are a reasonable base as are Shneiderman's Eight Golden Rules. Heuristic evaluation could be more effective if you modify the heuristics to suit the domain.

The thing with heuristic evaluation is to get reliable results you should have several experts doing the evaluating. The best results require the usability experts to be also domain experts. (Nielsen, 1992.) Other problem is that heuristic evaluation does not find actual usability problems, but rather the evaluator infers the problems from various qualities of the UI, which can lead to false alarms. Usability testing works the other way around where UIs problematic features are inferred from the problems users are having.

Generally more expensive way to audit UI would be to do usability testing with representative users. I haven't done too much usability testing, but not counting the cost it seems to me to be the method with least problems and most reliable results. If various stakeholders can be brought to observe the testing it can also be easier way to convince them that changes are needed.

I would in general go with usability testing augmented with as many experts doing the heuristic evaluation as possible. Or even non-experts doing the evaluation as many novices can trump one expert.

Regarding the Recommendations

In essence writing recommendations is all about making them as usable as possible. Bad report isn't useful.

Dumas et al.(2004) recommend that when writing recommendations, you should:

  • Emphasize the positive.
  • Express your annoyance tactfully.
  • Avoid usability jargon.
  • Be as specific as you can.

I would add to those that remember to include concrete improvement suggestions with the recommendations. Just pointing out the problem isn't that helpful.

Improve this answer
2

Do, just like the other fellas already pointed out a heuristic evaluation but you should also do a task list and then a cognitive walk trough.

A cognitive walk trough is basically you as an visitor trying to accomplish the stuff the site is intented to help the visitors accomplish. For example finding a phone number, buying a green sweater, returning it etc. Meanwhile you are trying to observe yourself and taking notes. What obstacles did you hit and so fort.

The benefit is that you will get a better understanding of how good the site is at solving problems and is a good complement to the checklist. The disadvantage is that you are probably more experienced and will solve stuff your visitor wouldnt.

Improve this answer
1
  • Yes its better for the person doing the Cognitive Walkthrough to not have been involved in the site before. You get one shot at this, as you'll 'learn' the site the more time you spend on it - so you need to take careful notes from the first time you look at it.
    – PhillipW
    Commented Jan 4, 2012 at 22:45
1

Pay attention to the "F" pattern as well to make sure the important page elements are present in those areas.

http://www.useit.com/alertbox/reading_pattern.html

Improve this answer

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.