expert usability review vs. usability testing

Lisa Halabi: ITWales.com [22/02/2007]

One question we're often asked is which method is best: usability testing or expert usability reviews? Well, if they were sports cars, expert usability reviews might be a Porsche (pretty decent car and better than no car at all), but usability testing would be in a different league, namely Formula 1.

So, what's the difference?

• An expert usability review is when a usability specialist inspects a website to identify potential usability problems
• Usability testing involves getting people from the target audience to evaluate your site whilst performing tasks

They're quite similar in many ways in that both:

• Find and prioritise usability problems
• Evaluate designs in the context of tasks

Do they find the same problems?
The answer depends on how precisely you measure this, on balance no, not reliably. Generally speaking, expert usability reviews tend to find high level breaches of design rules and consistency. Usability testing is better at finding issues related to special domain knowledge and task flows.

If you were to carry out an expert usability review before conducting usability testing, and compare the two sets of findings, you can see how the issues you found compared to their real experiences. Did you find all the problems? Which ones did you miss? Which ones that you thought might be issues turned out not to be? This will give you a good feel for how expert usability reviews hold up against usability testing.

Often, expert reviews will:

• Miss usability issues that arise during usability testing
• Find some issues that usability testing didn't
• Report false alarms (i.e. not real issues)

And that's the problem. No matter how good an expert you are if you just rely on an expert usability review you'll find different sorts of issues and overlook some potentially serious ones.

Consider the case of a health website aimed specifically at new mums. The expert usability review found plenty of great issues, but might have missed the main killer one that caused participants to pass comments like "That's just so insensitive!" and "I wouldn't use this site anymore". And all because of a seemingly benign bit of banner advertising for a local gym placed next to an article about breast feeding. A reviewer would have difficulty predicting the indignant reaction of new mothers who felt the site was rubbing their nose in it. After all there would be little chance of them going anywhere near a gym with a newborn.


How much overlap is there between expert usability reviews & usability testing?
Naturally, there is some overlap in the problems found in both expert usability reviews and usability testing. Research conducted by Fu, Salvendy and Turley in 2002 estimated this to be around 41%. Not as high as you might expect. Interestingly, expert usability reviews tend to find more problems compared to usability testing. But it's quality over quantity.


Expert usability reviews are good...
Expert usability reviews are more common than usability testing simply because they're cheaper and quicker to do. Typically, it'll only take a few days to inspect a site and write a report and because you don't need to hire labs nor recruit users, costs are significantly cheaper. Because of this, they're often preferred to usability testing, especially when time and budget are squeezed.

There are however some pitfalls. For a start, a lot depends on the reviewer's level of experience. No two usability experts will find exactly the same issues so the method is also open to practitioner variation. The biggest problem however is simply that they sometimes miss the real problems that cause users to fail tasks. This can be especially true when the target audience has a particular skill set. For example, an accountant may be better placed to diagnose certain types of issues related to online accountancy tasks than a non-accountant reviewer.


...But usability testing is da Daddy!
Usability testing takes more time to plan and organise and is more expensive too - recruiting and incentivising people from a target audience can be costly. But it's definitely worth the effort! The results give a truer picture of the real problems people encounter because they're derived from real users in the first place.

It's inconceivable how sites go live without checking to see that they're actually usable. It's like building a car but not test driving it. "Yes, we've built this great website. Umm, well no, we've not checked to see if people can use it yet. But don't worry because all the links work and the pages download quickly."

Another advantage of usability testing is that there's less conjecture and feedback comes straight from the horse's mouth. Videos of user sessions in particular, can provide a powerful persuasion tool for reluctant stakeholders. It's hard to argue with recordings of people in tears of frustration.


Use real users and accept no substitute!
Both expert usability reviews and usability testing have their time and place. In practice, people often use expert usability reviews early on to straighten up their design in preparation for usability testing. (Indeed this is critical for accessibility testing, since it's a waste of time to recruit and test users who can't even get into your site in the first place.)

Whilst an expert usability review is better than nothing, it's still vital to put your website in front of users as early as possible too, rather than leaving it too late. Key elements such as site structure and navigation should be assessed with users at the earliest opportunity. Otherwise you risk having your website completely overhauled at a late stage.

Ultimately, the key is to appreciate that expert usability reviews and usability testing are different beasts. The most effective approach is to try and integrate both techniques. Remember though, you must test your site with real users - accept no substitute!

About the author
This article was written by Lisa Halabi. Lisa's crazy about usability - so crazy that she's head of usability at Webcredible, an industry leading web usability and accessibility consultancy. She's very good at conducting card sorting and really enjoys wireframe design.

IT drives wedge between workers

Computing, [15/01/2007]

A lack of user-friendly technology in the marketplace is exacerbating a digital divide in the workforce between those who can use technology effectively and those who can't and is likely to provoke a backlash among users, according to a new Technology Predictions for 2007 report from consultancy Deloitte, released today.

translating the web: web site development for an Asian audience

The UPA Voice [December 2006]
By: Jacqueline Sinex

In all aspects of marketing, the Eastern world is breaking through to the West. More and more businesses are expanding product lines and services into a new market that involves countries in East Asia. Whether a business forms an alliance with a Chinese company to use its resources for a project, or it sells directly to Japanese consumers, it is clear that key media materials should be appropriate for Asian audiences.

websites face four-second cut-off

BBC NEWS [09/11/2006]

Shoppers are likely to abandon a website if it takes longer than four seconds to load, a survey suggests.

The research by Akamai revealed users' dwindling patience with websites that take time to show up.

It found 75% of the 1,058 people asked would not return to websites that took longer than four seconds to load.

The time it took a site to appear on screen came second to high prices and shipping costs in the list of shoppers' pet-hates, the research revealed.

Akamai consulted those who shop regularly online to find out what they like and dislike about e-tailing sites. About half of mature net-shoppers - who have been buying online for more than two years or who spend more than $1,500 (£788) a year online - ranked page-loading time as a priority.

It found that one-third of those questioned abandon sites that take time to load, are hard to navigate or take too long to handle the checkout process.

The four-second threshold is half the time previous research, conducted during the early days of the web-shopping boom, suggested that shoppers would wait for a site to finish loading.

To make matters worse, the research found that the experience shoppers have on a retail site colours their entire view of the company behind it.

About 30% of those responding said they formed a "negative perception" of a company with a badly put-together site or would tell their family and friends about their experiences.

Further research by Akamai found that almost half of the online stores in the list of the top 500 US shopping sites take longer than the four-second threshold to finish loading.

The survey questioned 1,058 net shoppers during the first six months of 2006.