Friday, February 16, 2018

However beautiful the strategy, you should occasionally look at the results. (Winston Churchill)

And isn’t that what usability testing is all about? 

Of course, there are other ways to get that information as well, especially these days. For one thing, you can simply put it out there and monitor the heck out of it. 

A kind of old-fashioned way to do that is Voice of the Customer – popping up surveys, including “give feedback” links, and so on. These days, you can even monitor social media. It’s all good stuff, but there’s also no shortage of possible drawbacks. I already wrote a post going over those in great detail, so I won’t repeat all that here.

You could also take a look at web analytics. (I wrote about that, to some degree, in another post.) The main thing with analytics, though, is that, though you might have tons of data showing exactly where users went and what they did as they actually used your system, you totally lack why they did so. And that can be pretty darn important.

It’s the same problem with A/B testing. Once again, you get reams of real data, all tied directly to the bottom line, and in this case, totally actionable. You are, however, relying totally on correlation – which, in my opinion, means relying totally on conjecture. We know that A performed better, but why? How can we tell with just the numbers? What were users thinking when they went with A?

And then there’s usability testing. It’s the prime way to get feedback before release, but it also provides super-rich data that can be used to completely understand your users, identify and understand their issues, and then come up with legitimate solutions to address those issues. 

That said, if you are getting some kind of feedback, you are way ahead in the game. What really worries me are those companies out there who basically don’t do any of this.

By the way, when Churchill was talking about strategy and results he was actually not talking about usability testing (I know, hard to believe). I would imagine he was talking about something like war plans, or economic strategy, or political campaigns. Unfortunately, we may never know. There is actually no evidence that Churchill ever said that (and I’m not sure who did). But it certainly does sound like something he would have said.


Bet you had no idea that the Daleks were a secret weapon during WWII
(or that Churchill looked anything like that)

Thursday, February 15, 2018

Mobile is a magnifying glass for your usability problems. (Josh Brewer)

There are designers and developers and UX researchers out there whose whole world seems to be totally encompassed within the mobile space. Now, that is definitely a fact of life these days. At the same time, though, that’s also a tad unfortunate as well.

For some of these folks, mobile is a brave new world, where they are intrepid explorers creating their own rules. Now, mobile definitely is terra nova. It is not, however, terra incognita.

For someone who’s been doing this as long as I have, mobile is really just the latest whatever. All the things that we learned about green screens, CLIs, GUIs, the Internet … they pretty much all still apply. Yes, there are new things like gestures, but affordances, mental models, proximity, hierarchy, information density, legibility … all that great stuff has not somehow magically disappeared overnight.

Reinventing the wheel is not the point with mobile. In fact, designers need to think, not about being innovative so much, but more about handling the old principles within the additional constraints that mobile involves. 

And what are those? Well, the main one by far is simply the huge difference in screen size. And what does that mean relative to UX? More than anything, I think it takes the old KISS formula and makes it absolutely paramount. Complexity that you might have been able to get away with on a nice big screen is simply going to blow up in your face on mobile.

Less central – but certainly not unimportant – issues include awkward input, new environmental contexts, and several others. There’s also one thing I’ve noticed which probably doesn’t need to be an issue at all. For some reason, some designers took the new context of mobile and decided to just ditch the idea of affordances altogether. 

What do I mean by that? I just seem to see missing pieces on every mobile test I run – or as simply a user myself. In fact, just this morning, I was testing out a prototype for a test I’m running next week on an iPhone …

First, though, I had to clean the screen up a bit, and get rid of some old icons on the home screen. Now, how to do that? Why press and hold those icons, of course (I somehow remembered that from the last time I had to test an iPhone). That puts you into delete mode, where the icons wiggle and have litte x’s in their top-left corners. Okay, this I can figure out – an x is a pretty darn clear affordance. But what do I do to return to normal mode? I would imagine I would press a wiggly icon again. Nope, you press the Home button (had to really think about that one).

Now I can type in the URL of my new prototype. I do that, bring it up, then wonder how I can put an icon for it on my home screen. Oh, it’s the little box icon with an arrow in it (had to ask an iPhone user for that). What do I click now (interestingly, the iPhone user couldn’t help me there). Oh, that little bar of icons on the bottom – looks like I can slide it over (had to ask an iPhone expert for that one). Now, why didn’t they signal that somehow? You know, like with a >, or a little bit of the next icon, or practically anything? 

And all this is before I even get to start in on that prototype. And after that, I then get to try it all over again on an Android, which I least am familiar with, but which also has its own way of doing things, and its own set of affordances to ignore as well. Ugh, it’s gonna be a long day …


Josh was once the principle designer at Twitter