Selling usability

I took a new job a few months back. The initial people I met wanted to bring in someone that would improve usability in the product lines and didn’t seem to have any concept of usability. Later on, I met some decision makers that are not clear they want usability help on the products, and have very definate opinions of what a usable product should look like. From this experience, I have a few questions:

  1. I know that fewer buttons do not necessarily make an easier interface. However, is there any data that consumers perceive less buttons as an easier device?

  2. I need to measure how valuable features on our products (how often they are used, how important the end users think the functions are). I have a few ideas on how to get data, talk to our CS dept., talk to customers looking at our products in stores, talking to sales dept. Does anyone have any other good techniques at learning what to keep?

  3. I can do usability testing and task analysis etc. What are some effective ways of communicating these specialized results to people who are not designers?

Thanks!

I’m sure there is, but couldn’t point you to it. It’s pretty deductive: the fewer buttons, the fewer perceived features. What gets you in trouble is when there are in fact more features than there are buttons, creating ‘modes’ of operation. Don Norman details this problem in ‘The Design of Everday Things.’ So you can make things appear easier to use, but actually make them behaviorally more difficult. This tradeoff is best determined via user testing: ask them what they think at first sight (visceral) then after they’ve used it (behavioral.) A great example: is a one button mouse easier than a two button mouse when you factor in the use of modifier keys? Apple (and their customers) still seem to think so!

I loved how Don Formosa from SmartDesign characterized the iPod’s genius: "A CD has approximately 12 songs. A Sony CD player remote has 27 buttons. The Ipod can play 10,000 songs with 6 buttons. (Sony = .44 songs/button, Ipod = 1667 songs/button.) “that is an amazing invention”

  1. I need to measure how valuable features on our products (how often they are used, how important the end users think the functions are). I have a few ideas on how to get data, talk to our CS dept., talk to customers looking at our products in stores, talking to sales dept. Does anyone have any other good techniques at learning what to keep?

Most designers will do features analysis in tandem with marketing. I like to carve our responsibilities so it’s clear we’re focusing on the Usable/Desirable aspects of features.

QFD is a well respected technique: QFD - Wikipedia

‘Co Creation’ techniques are all the rage right now for good reason. Put users in the designers seat and have them ‘collage’ cut-out components together to form their ideal product or experience. Add up the results and you have data to present back.

Also, don’t forget to do a competitive analysis–you can get pretty far by choosing a few ways to differentiate without necessarily innovating.

  1. I can do usability testing and task analysis etc. What are some effective ways of communicating these specialized results to people who are not designers?

Thanks!

  1. Stories: The most powerful! Use quotes and pictures of actual users.
  2. Data: quantify what you can with rating scales, timing users, measuring their ability to correctly complete tasks etc.


    Best of luck!

Re: question one take a look over at one of dons essays…which leads to some interesting pointers for certain markets

Maybe a better question for #1 would be: how is perceived complexity related to number of button. I’d like to research this, but I would suppose it is not a directly linear relation. Common sense would say people would think devices with 1-3 buttons would be equally easy. 3-10/20ish would be more difficult. Over that would be immensely complex.

Of course, it difficult to say for sure without taking into account the interface design. Take the iPod for example. I have a CD player with seven switches on it. An iPod, really, has eight distinct controls. However, one would perceive the iPod as simpler, because it looks like it has 1-3 controls.

Another example would be a Tivo remote control. It has many switches on it, however they are arranged into groups. This must have an effect on perceived complexity.

One other factor would be familiarity. If I show people a 3x4 arrangement of buttons, it may seem quite complex. If I add the familiar numbering of a telephone, I’d bet people would list it as a very simple interface.

I smell a thesis project here! I’m definately going to see if its already been done!

Now I’m offended! Didn’t you read my post that made this exact point?


To search for whitepapers on this topic, go here:

If this topic has been covered, it’s likely to have been posted via SIGCHI.

http://scholar.google.com/ is also very good

I haven’t found anything yet. I searched through thesis papers last night trying a variety of search terms. I also tried the link from cg, but didn’t search too deep into the results. Google scholar seems interesting, although I need to get my search down from a million results!

It’s interesting…I’m finding alot of results for research into creating easy-to-use interfaces, but not one paper looking at user perception of different interfaces. Maybe it’s a case of not bothering to research “conventional wisdom”?

In “Emotional Design” Don Norman covers this and references a study having to do with an ATM machine. The study proved the correlation that ‘easier looking’ = ‘easier to use.’

This article covers some “concept testing” methods you could use:

Those refs are:

Kurosu and Kashimura (1995) Apparent usability vs. inherent usability:
experimental analysis on the determinants of the apparent usability

Tractinsky, N. (1997). Aesthetics and apparent usability: Empirically assessing cultural and methodological issues.

These we studys in to good looking things work better, the first in Japan the 2nd in Israel .

There is also an interesting point raised in the Universal principles of design regarding the 80/20 rule:

“The 80/20 rule asserts that approximately 80 percent of the effects generated by any large system are caused by 20 percent of the variables in that system”

Thanks Wolfman!

I checked out both. Both are posters from SIG CHI conferences, 1995 and 1997 respectively.

http://acm.org/sigchi/chi95/proceedings/shortppr/mk_bdy.htm
http://acm.org/sigchi/chi97/proceedings/paper/nt.htm

Look at the second one, which references the first and is more descriptive.

I think this is what you’re looking for: “Higher apparent usability was associated with less groups, defying conventional advice in the usability literature, which calls for the separation of functionally unrelated controls”
and “the results provide further support for the contention that perceptions of interface aesthetic are closely related to apparent usability and thus increase the likelihood that aesthetics may considerably affect system acceptability.”

I also recommend digging into their bibliography, as this is likely to contain most of what you’re looking for.

I never thought of searching “apparent usability”. Just the titles alone would broaden my search. Thanks wolfman for the references!

I have to say, it’s remarkable that this stuff has been studied to this degree. It just goes to show there is nothing new under the sun.

My only complaint is that design school never taught me to read…oh well.

Thanks again for all of the help.

Well, I read the articles that wolfman sites in the topic, plus another few related articles. I think it is safe to conclude that an attractive looking product is perceived as easier to use. However, the studies, plus Don Norman’s article on simplicity, leave me wondering about my first question even more.

How would users have responded if one ATM had one button less because two functions were mated to the same key? Would the results have varied at all? What if the difference were larger, 3 keys versus 12 for example. Argh! I want more data!

I’ve been thinking of trying to do my own online survey to test this, however, I see all the same problems that those researchers had…namely how do I make sure I’m testing buttons and not brands or aesthetics or some other criteria.

I was thinking of presenting a widget type product with various amounts of buttons and asking people ot rate them. However, I don’t think people could relate to a non-product. So many products are either button laden or button deprived however. If I choose a telephone or remote and remove the numerical keypad, obviously this is going to influence the results. If I attach 30 buttons to an MP3 player, their innate superfluousness will undoubtedly through the results way off as well.

Any ideas or warnings would be greatly appreciated!

I think I’ve found my product to test my theory: the clock radio. As we found with cg’s problem, these things can come with any number of buttons, switches, knobs, sliders etc. What a perfect platform to see how people react to various amounts of control!

Expectations of performance vs. Actual performance are a huge problem in retail, where consumers rarely have the opportunity to try before they buy.

I got burned on my Philips alarm clock, which I sought to replace, but then again on the Oregon-Scientific Starck clock. As a result, I’ll think twice about buying products from either company (and so perhaps will anyone else reading these forums!)

In the end, you should just mock up some options and get some consumer feedback on both dimensions: expectation of use vs. actual use. Your design goal should be to develop a concept which scores high on BOTH dimensions.

Reducing visual/percieved complexity is only half of the solution, you need to consider the complexity of the complete interface when talking about making things easy to use.

Just an update to the literature which we’ve compiled in this thread. I found this book to be describing all of my problems and offering some solutions:

The Simplicity Shift by Scott Jenson http://www.amazon.com/Simplicity-Shift-Innovative-Tactics-Corporate/dp/052152749X

I haven’t heard of this book in ID circles before. Unlike Don Norman’s book, this guy comes at usability from the corporate culture side rather than the user side. Anybody out of school needs to deal with these problems inside an existing culture, therefore, this book feels more relavent to me.

My 2c worth I pose that the optimum number of buttons for a complex device and one that not only appears simple and common is the 10 or 12 layout, think about it adding macnine, phones, and calculators all use this layout so it has a familar feel to it…just a notion.

That’s a good point.

12 key layouts are “conventional” and convention = viscerally and behaviorally usable.

I think the 3-4 button “highlight, enter, back” system is also extremely conventional (iPod etc.) and is preferred despite the fact that it often requires more discreet interactions and the dreaded narrow/deep IA.

It would be interesting to explore perceived and measured usability of various key layout design patterns.

the 10’12 key pad also has a advantage of being within the old 7-10 digit number set that is easy for us humans to remember. When you look at your 12 key pad on your keyboard you see a system that is also scaled to have large easy to use keys even with in a small foot print. I am certain that when apple did the design for ipod they said "ok whats sized for the hand, and the 10/15 key pad was laying right in front of everybody. Reinvention of the wheel can be a lot of fun for some, but in the end what has been around for a long time works pretty well or would not still be around.