There is No Spoon and there is No Usability
Usability isn’t anything at
all. It doesn’t exist. Instead, usability is a label given to the results that we see or want. Just as osteoporosis isn’t a disease, usability isn’t the outcome of usability testing. You probably didn’t know it but osteoporosis is just a label given to a cluster of medical symptoms. Analogously, usability is just a label given to a list of desirable features or outcomes associated with a product or service.
Let’s not mistake the outcome of the usability process (e.g., testing) as usability itself. And, let’s not clump usability methods together and call them usability either. Putting this another way, we should stop using the outcome as the label. Usability methods don’t create usability, but they do produce outcomes that are beneficial to users. Usability methods help foster satisfaction, efficiency, desirability, and more, but usability methods don’t produce usability itself.
You can measure temperature but you can’t measure usability. For example, I can say that it is 78 degrees Fahrenheit in my room but I can’t say that my room has any type or level of usability. There is not a usability metric. At best, we can throw together several usability outcome measures together, such as level of satisfaction and completion rates, but we can’t measure usability itself. There is no metric for it. It just doesn’t make any sense.
Summary: Usability as a term should be thrown into the pit. It isn’t real. It doesn’t exist.
March 9th, 2005 at 9:58 pm
I completely agree. I say the same thing about UCD to my coworkers. Within the area we sit together we can discuss UCD all day long, but comes the time to go out into the big, bad world of project work I say UCD doesn’t exist. Don’t talk about it. Don’t try to sell it. Just do it for yourself because the project will appreciate your outcome.
I used to hate being called “The Screen Guy.” I don’t care anymore. To them I will be the screen guy.
Meh, way too many words. I agree.
March 9th, 2005 at 11:25 pm
You gave the room tempature example. Well usability is “comfortable room tempature”. You cannot measure it but you know when you are in it and if you are not in a comfortable room tempature, you start to complain. That’s how you measure it.
Not directly related but an interesting article: http://www.sitepoint.com/article/quantify-user-experience
March 10th, 2005 at 7:57 am
Usability smells like cheese.
(Coming soon: The follow up to UsabilityMustDie.com will be UsabilityIsDead.com)
March 10th, 2005 at 9:25 am
Random thoughts …
Teach the usability tools to get people to think about the end user as part of their work - developers, marketing, product strategists etc.
Get deep into the data to understand how it can assist the business strategically or “make a buck” and at the same time find opportunities to ask is it “useful” (why do we need it?)
Suggest some of these tools are becoming commoditized, but how you think about the findings, what you do with them and how these are communicated effectively to the stakeholders is something that a) comes with experience and b) Continue to learn …
Good to see some discussion back on Webword …
March 10th, 2005 at 6:09 pm
John,
Provocative statements. While usability may not exist, many constructs also don’t exist in the same sense (such as intelligence or personality) but this doesn’t stop us from making very important decisions on these constructs. We are constrained by the limits of our instruments and the method of empirical verification, meaning we need to measure proxies for constructs (the loathsome IQ tests, personality batteries and the metrics from our often maligned usability tests). We should be cautious about putting too much faith in our use of any construct and their corresponding results. This caution however, shouldn’t prevent us from making statements about what we observe.
Can we say people are intelligent based on the formation of their arguments, their ability to solve complex math problems, the words they use or a record of their scholastic achievements? Or should we withhold such statements since there’s no such thing as intelligence, “it’s merely a label given to the results of seeing things that we see or want.”
I think you’re touching on an important topic, that of reification, or the treating of an abstraction as if it had concrete or material existence. The late Stephen Jay Gould had a lot to say about this issue in his seminal book “The Mis-Measure of Man.”
Your temperature statement is appropriate and is similar to a point made by Joe Dumas. “There is no thermometer-like instrument that can provide an absolute measurement of the usability of a product (Joe Dumas User-based Evaluations Human-Computer Interaction Handbook, 2003). Usability is an emergent property that depends on the interactions among users, products, tasks and environments.
With that said, while you may be able to look at a Thermometer to measure temperature, you’re actually measuring just the movement of mercury or alcohol—a proxy for the temperature. And temperature is often used as a proxy for a measured comfort level to help us know if we should wear a light coat, heavy coat or crank the AC. Even this gold standard of measurement is flawed as thermometers measure temperature whereas humans care about comfort. “In Washington DC, 90 degrees with 90% humidity just doesn’t feel the same as 90 degrees in Flagstaff Arizona with 10% humidity. True measures of comfort should take into account humidity, elevation wind speed and temperature.” (Richard Saul Wurman, Information Anxiety p 286.)
So while we certainly have a long way to go in measuring usability, if done the right way, throwing together a bunch of things that purport to measure a construct is often the best we can do. This is in fact what is done for measuring intelligence (Stanford-Binet IQ Tests), measuring credit worthiness (FICO scores) and personalities (Myers Briggs and the like).
While there is no metric for usability, there is often a need. We’ve attempted to put one together. We’re presenting those results this year at CHI in Portland. Out paper and talk are entitled “A Method to Standardize Usability Metrics into a Single Score.” I’m looking forward to your comments.
Jeff Sauro
March 10th, 2005 at 9:49 pm
I always thought the measurement for usability was ROI.
March 11th, 2005 at 1:16 am
Well, yes. But then there is no such thing as information technology either. Information is a far, far, more abstract notion than usability. And nobody within the industry seems to care much whether anyone ever measures information, the suposed product of IT. Except perhaps to say everything is information without qualification or quantification.
So they actually managed to make the word information meaningless. Good work.