Why You Don't Need a Usability Lab Options

[Ruud]

http://www.sitepoint.com/article/dont-need...d-usability-lab

Although the James Bond-esque appeal of a high-tech lab cannot be denied -- video cameras mounted on the walls and ceiling with pan, tilt, and zoom controls allowing you to follow participants' movements! Video editing equipment so you can create a concise presentation to summarize your findings! Auxiliary keyboards and mice to control participants' computers from the adjacent room! Keystroke logs of all user input! -- it's probably not necessary for your purposes.

A funny, and true feeling intro to a good down to earth article on getting out there and doing it versus ... well, everything else :-)

[Paul_H]

The company i worked for a few years ago had a usability lab, made up of psychologists and other non technically minded people. Would of been great if they had understood what’s feasible with browsers, and known anything about web standards – but they were to busy monitoring users eye movements with 1000’s of pounds worth of equipment to learn anything useful like that.

No that i’m bitter(was the most fun job i ever had), but they played a big part in the project that finally nailed the company. -hmmm maybe i am bitter after all

[cre8pc]

I'd seen this article too. It kinda skims the surface of what really is a big issue in some companies, depending on what they make and who they're making it for, budget and employee experience, and a host of other things.

To do it right (lab testing) a company has to qualify the expense. There's always a need, before, during and after any build, whether it's a web site build or Internet software app. How badly they need to examine actual hands-on user experience is up for debate. Obviously if a company invests a ton of money on design, development and R &D costs, it would make sense to hire usability consultants to come in and set up a full blown lab to see how people use the final or even the spec'd out product.

I see a need for both a lab and the alternative, which is budget friendly user testing and/or usability evaluations by a usability consultant trained to look for problems.

Whereas a corporation can afford to hire companies that charge $600 PER PAGE for web site usability testing, the majority of the people who also need usability help have many affordable options that will increase their bottom line and help meet their goals without breaking their piggy bank.

Inviting friends over for dinner or in Stock's (aka Grumpus) case, 50 rounds of beer, is often all it takes to get volunteers to sit in front of a computer so you can watch them use a web site, form, shopping cart, etc. :wink:

[mugshot]

I find it useful to just send an email out to your immediate friends and letting them review your site.

Spending a hige budget on usability testing versus buying beers, will yield the same results Regardless of the form of payments, if a site is usable it is usable, if it is not, heck you dont have to be drunk to realize the site isnt functional

[dlumerman]

In working in a large corporation I've found the usability lab indispensable, but not for the actual use of the lab.

The lab adds to to propaganda as well as validation of usability in the company. My lab is not elaborate, 2 machines, a wall and camera, but having a dedicated space that you can convert and cheerlead about usability has meant the difference between being involved in projects or not being involved.

Usability is as much a hearts & minds problem as is a science.

Years ago my company had a big lab with 2 way glass the whole deal. It was shut down and disbanded because they filled it with phd's that did not connect to the business customers, thus had no support.

Show the ROI, show how it's easier to do usability, show how it can help them. The lab just becomes something else in the toolbox.

[cre8pc]

Interesting. I heard, a few years ago, that IBM (I think it was them) had a glassed in lab room in a section of a building that visitors could access. Their usability testing was promoted as a public relations thing this way. But, who did they actually impressed is something I always wondered about.

Welcome the forums by the way. I've been enjoying your thoughtful posts!

[fisicx]

A usability lab is like any other form of market testing. You take Joe Public and plonk him (or her) in front of whatever product you want to test and let them get on with it.

The problem with all testing is how the test is carried out. Ask me to test beer whilst shopping with all 73 members of the family in tow and you will get some very skewed results. The same with a website. Sit me in front of the latest creation by 'posh designs ltd' and the results will be very different from someone searching the web for a particular product or service and finding the new site. Test at home, in the workplace, in an internet cafe, on the train. Test in the environment in which the thing will be used. Back to the beer, do the testing in a pub on a saturday night. That's when people will appreciate the product (might not like it but they will be more comfortable using it)

A last analogy. Peoples blood pressure goes up in the doctor's waiting room. Do the same test at home in front of the TV and the results are less likely to be biased.

[RonZ]

What are their expectations, based on the design?
...
What would they expect to be able to do on this site?
...
What does the language mean to them?

What the author describes, given the questions she poses, is not a usability test. So her points about not needing a lab, remote testing, etc are useless. If you don't know what it is you're trying to accomplish, anything goes. :wink:

What's the main point of a usability laboratory? To simulate a normal office environment, and find out how real users are experiencing your site.

Nope. Not even close. The point of a lab is to control the factors that can lessen or invalidate test results.

Not that I think a lab is always appropriate, but you need to know what it is you are trying to accomplish with a test before you can figure out how to best go about testing.

And you might want to figure out what you are going to do with the test results as well...

[cre8pc]

There's different testing. Some is simply user interface, and for web designers, is a great way to understand how someone moves around a site.

I think user testing is going the way of focus groups, in that the findings aren't reliable indicators for painting the whole picture. You only get parts of it.

The kind of user testing I find valuable is highly targeted and specialized, where the actual **intended** users are watched, in environments that simulate reality, such as rushed lunch hours, interruptions from other people, bosses looking over shoulders, laptops on train rides, whatever.

User personas are one approach to designing for the intended user, when the actual human or humans aren't available.

But, mainstream design has far to go. Most design is still what the designer likes and thinks will work, without investigation or followup testing of any sort.

This lack of follow through, by the way, contributes to the failure of search engine results conversions. The designer brings in the SEO/SEM and they haggle for keywords, but there's less consideration on who chooses those keywords when searching, why they did, and what the expected action is upon delivering the page once found.

For example, keyword "bike". What country? What age? What for? What type? What gender? etc etc etc Who is the end user and what do we expect them to do when we make a page about a bike?

Another example is something I thought of last year, while sitting in my Saturn car dealership. They asked me how I knew about something and I said "Your website." But it was the mechanic who asked me, not the designer. It's unlikely the mechanic gave any of my usage feedback to anybody. They should have user testing done in their dealerships, like a testing cafe, where they can test their user interface, order process, etc. with their customers who are more likely to use their web site.

[fisicx]

If you want to observe how the monkey peels the banana then you do need a laboratory of some sort. If however (as Kim says) you want to simulate how somebody uses a product then you need to reproduce the environment in which they will use the product.

And as already been eluded, there are different types of testing. Many 'consultants' however will have their own preferred method. If you really want to do the job properly then you need to employ different techniques and compare the results. But a warning - one consultant the company employed would not give us the raw data (commercial in confidence!) only the analysis which was so biased towards the directors viewpoint it proved to be worthless as a R&D tool.

So in a round about way - I agree with Ron, you need to know what you want the results for before you decide on how you are going to test for those results.

[phoomp]

fisicx
Test at home, in the workplace, in an internet cafe, on the train. Test in the environment in which the thing will be used

Of course, the real world is the best lab, but it's also the most expensive. How many users do you think can be cost effectively tested in the home, workplace, cafe and train? Maybe two?

For a 5 page site this might be acceptable, but when you get into sites that number in the hundreds of pages it's a much different story. One or two users will not provide a very good sampling of the users that may visit such a large site, especially if the site is intended to sell a product to a large cross-section of society with users of widely varying skill levels.

In this case, one needs to test many users. Doing so in the actual use environment would, of course, be idea, but it would be extremely cost prohibitive. Enter the usability lab.

[cre8pc]

Of course, the real world is the best lab, but it's also the most expensive.

Definitely! Which is why the fall back measures you describe are relied upon.

I think that up until now, any user testing was enlightening because this is all so new still. In general, web dev is still experimenting and the whole computer/human connection is still being explored. From this perspective, any user testing is going to blow the obvious problems out of the water.

But, more and more companies, with the money to play with, want to know more. Some were inspired by user personas, and wish to see those user personas in action, for real, rather than the imagination or on paper.

It can be done. One way is remote access, and mouse tracking. The "user" can be anywhere and their mouse movements recorded as they attempt a task.

We'll be hearing more about this in the coming year, and other ways usability testing companies are coming up with alteratives, though like anything new and wonderful, we'll have to pay out the nose to get it :roll:

Welcome to the forums by the way!

[-Phoenix-]

Of course, the real world is the best lab, but it's also the most expensive. How many users do you think can be cost effectively tested in the home, workplace, cafe and train? Maybe two?

Agreed.

But then again you can get creative to lower the costs. I can usually get about 20 people in a night at the local coffee house to do user testing for a while if I buy them a cup of coffee. I put a sign out near the cash register stating for 15-30 minutes of your time to do user testing on websites I will pay for your coffee. I get about 70% people that walk in do it. They are mainly curious to see what its all about, they could probably care less about the coffee. I usually have two-three that end up hanging around all night. So it becomes party. Total cost about $37 for the night. The information I get out of it, is extremely valuable and often makes change right then and there as they are give me feedback.

[cre8pc]

Ah HA! So it's you where I first got that idea about the cafe :doh:

Hadn't you written about it before? I know I've seen this and thought it was brilliant

Me and Starbucks. heh.