Intuition: it’s not what you think.
You might feel intuition is the best way to understand the problem of designing an interface. You rely on your intuition, you trust it, maybe you even feel that it’s the source of deep insight and creativity.
But you’d be wrong.
Intuition is really your first impression, that little bubble of apparent understanding that comes unbidden in response to a situation. When we build our UIs, intuition is often what guides our initial work. “Ah this one…it’s child’s play,” we think to ourselves. And it’s too often where we stop.
Turns out you can’t quit with just the translation of intuition into an interface. That’s the point of user testing – you’ve got to watch your pride and joy actually be used by real people in real situations. It’s not enough to have a brilliant, intuitive idea that you think will work. (After all, it’s intuitive, isn’t it?) You really have to watch someone else trying to solve a problem with the child of your intuition. You have to put the test users under a bit of stress, so that the natural beauty of your interface design isn’t the thing they’re noticing and appreciating – they’re focusing on the work itself, and not the tool.
Great user interfaces might spring from an intuitive understanding, but they’re crafted by hard work; honed by carefully removing everything that doesn’t add to the tool’s value, deleting anything that gets in the way, or whatever it is that causes confusion.
Making the user interface “intuitive” is often our highest goal when we do our professional work.
But what does that mean, exactly?
Mostly it means that people can quickly understand the UI, or that they get the feeling the UI disappears in their work. In psychological terms, the interface has great “positive transfer” from other skills or it has a great “flow” character. That is, if you can drive a Toyota, you can drive a BMW – that’s a great positive transfer experience. If you can move smoothly from task to task without many bobbles, that’s a great flow experience.
And sometimes it’s the case that intuition informs both the design of positive transfer and flow; but just as often that’s not the case as well.
At this point in my career I’ve seen thousands of user interface ideas: they pour unendingly from the minds of software developers. I love ‘em all. Look at them all trickle by in the thousands of wildly creative and inventive ideas.
But as a consequence, I can no longer look at an interface with a fresh, unbiased, open mind. I'm corrupted. I’d like to think that I can still see an interface as though for the first time, but I can’t.
What’s more, you probably can’t either. If you’ve been working in the software field for any length of time, you’ve grown accustomed to the look and feel of tools that wear their development history on their sleeves. A command line might feel natural to you, but there’s no positive transfer for anyone who hasn’t already spent thousands of hours working with one. The same thing is true of many of our prized productivity tools – even knowing where to look in a busy GUI is a skill that comes only with a lot of experience in the trenches. There’s nothing natural about a GUI in the least.
(Repeat this after me: “There’s nothing natural about it in the least.” Good. Repeat every day until you reach satori.)
As interface designers we often achieve our greatest successes when the interface disappears, and getting to that point is what makes us professionals at this game – our willingness to go beyond our personal intuitions and see what really works for our target audience. Great user interfaces are ultimately about creating something that’s ego-less, something that works well for people who are not you and not just the same as you.
Now that’s intuitive.
Posted by Dan Russell on June 13, 2006 | Permalink
TrackBack URL for this entry:
Listed below are links to weblogs that reference Intuition:
» Blogger´s Digest IV - Intuition, Intuition und nochmal Intuition from marketing-blog.biz Marketing für Besserwerber
Einfach mal meiner Intuition folgend sei hier auf 2 - so wie ich meine - herausragende Artikel zum Thema hingewiesen: streuverluste: Oft ist die Intuition der richtige Weg! Wenn es darum geht richtige Entscheidungen für das eigene Unternehmen zu tref [Read More]
Tracked on Jun 22, 2006 6:49:56 AM
Good post, Dan.
Looking back at some of the interfaces I've used, or even those I've written is fascinating.
The technology has a strong stamp on what is considered "intuitive" at any time. For example, at the risk of dating myself, I loved Borland's Turbo C environment, even though it was character-based. These days, it would be considered totally unacceptable because of more advanced technologies now available. But it was really intuitive at the time.
Now, the widespread use of Microsoft Office products, Web browsers, and sites like Amazon have conditioned users to expect certain behaviors in their applications.
So "intuitive" often means do it like that site or product that I already know how to use. But to improve the state of user interfaces often means pushing the envelope past what previous applications have done. It's a tricky balance.
Posted by: Dave Churchville | Jun 13, 2006 12:19:59 PM
Some of my favorite moments during tests are when the customer is asked to show how she would perform a specific task, she does so rapidly and without pause, and then reacts to the facilitator as if to say "Well, duh, is that all you wanted me to do?"
Good interface design is like good lighting for a play; it is at its best when it goes unnoticed.
Posted by: Jon | Jun 13, 2006 9:49:25 PM
Nice post and thanks for the 'positive transfer' termin. But I think there is something lacking in this post - beside intuition and user tests there is also the science of user interfaces with laws and everything (see for example http://en.wikipedia.org/wiki/GOMS, http://en.wikipedia.org/wiki/Hick%27s_law).
Posted by: Zbigniew Lukasiak | Jun 14, 2006 4:20:30 AM
Z - I didn't mean to imply that this is ALL there is to UI design. Heaven forbid! I worked for years at PARC with Stu Card and Tom Moran, so I think I'm pretty up on GOMS, Fitts, Hicks and the rest of those models. I really wanted to make the point that intuition isn't enough, tempting though it is to use as the primary motivation for a UI design. There are many other things to consider, but more on those in another post...
Posted by: Dan Russell | Jun 14, 2006 4:33:26 AM
To come at this from the other end, this relates to one of the "Rules of Technology" that I've developed (and often teach) over the years.
Namely, when you're trying to figure out anything--hardware, software, UI, whatever--and it doesn't make sense, remind yourself:
"Somebody, somewhere, thought this was intuitive."
By putting myself in their shoes, even if (as often happens) it doesn't really make sense to me, I can often figure out how something works.
Makes for a nice rep as a tech whiz, too.
Posted by: Gray T. Miller | Jun 14, 2006 8:31:14 AM
I couldn't agree less with you Dan. While you talk about UIs for the computer, the cell phone UI (which is seldom talked about) needs even more attention. I'm talking from my own experience. As a startup we launched an alpha version of LinknSurf (www.linknsurf.com) on Jan 19th for people to be able to track and share blogs and also stay updated with each other. We thought (and still think) that it's the most elegant solution of its kind. However, over the last couple of months if we realized one thing, it was how poor the usability was. We have been working on it ever since and have been releasing quick versions and all I can say that right now it's much better than what we had earlier. Whether it is really good or not, I can't say. Maybe the readers here could provide some feedback.
The one thing I have indeed realized is that it is one thing to understand and realize a user's requirements, but another to understand and provide for preferences. A good UI is about preferences.
Posted by: Harsh Dhundia | Jun 14, 2006 11:19:03 AM
Excellent post. To me creating something intuitive is two-fold. I agree that an interface has to 'disappear' and allow the functionality of the tool that is being used become the focus.
But 'intuitiveness' is also achieved by following design patterns. If a user has seen it before, it can become intuitive. As you mentioned, if you've spent hours upon hours using an application, it becomes intuitive. That's natural. Real intuitiveness comes at the point your first user jumps in and starts to play.
If you adhere to design standards and patterns, chances are that adaptation of users will be quicker and easier because they have seen it before, although in a different application.
Posted by: Andy Brudtkuhl | Jun 14, 2006 1:30:32 PM
Excellent post! However, how does one judge intuition then? I remember working with a gentleman once who said he scored high on an intuition test - I think it was one of those leadership tests. Are some people gifted with the ability to gauge intuition? I wonder???
Posted by: Walker Thompson | Jun 15, 2006 7:30:33 AM
Nice post; First time I've been to the site, will add it to my reader :) Thanks
Posted by: Ken | Jun 15, 2006 7:27:20 PM
Isn't it time to say that the right word is not "intuitive" but "conventional"?
The "transfer" that is obtained, drawing upon the experiences that millions have had with leading software is a transfer of user interface conventions. Convention is maleable, changeable, learned. And intuition is conventional too, in a rather personal way, just as preferences are.
This "intuition" is intuitional in the way that a language is. Internally reasonable (conventional), but foreign to the users of some other language-convention, as in "it's all Greek to me."
Posted by: MJ | Jun 15, 2006 8:44:15 PM
I am lucky enough to regularly take part in usability tests with 6-10 year old kids while developing video games. We see a wide diversity of ‘ability’ with our software, from almost perfect control to near zero unaided interaction.
Kids tend to have more limited experiences and therefore less conceptual models to which they can compare a product, behaviour or interface. I’d suggest that intuition is guided by the conceptual models you have built up and that convention is itself a set of re-enforced models.
Posted by: 'gene cloud | Jun 16, 2006 2:51:25 PM
Thanks for the fun read!
I must admit, although I enjoyed your previous posts, I was missing Kathy, just a wee bit. ;) But, I am now very much looking forward to future entries from you. I intuitively feel that you have finally let your intuition guide you in writing, and your personality is starting to shine.
I agree, it is sad to realize that the pinnacle of one's career is achieved when no-one really notices your work, because it is so seamless. But, that is also the misfortune of great actors, great writers, and great craftmen.
I think you addressed the most difficult issue for designing an interface. You need some-one who has the extensive knowlege of what is possible to create, but at the same time must remember how they used to think about things, rather than how they think about things now, to create an interface that is more intuitive to more people. In a very real sense you must live in the past, present, and future.
Intuition is very personal. I find I get very frustrated when my husband tries to explain 'computer stuff' to me. I end up figuring it out on my own and then saying to him, "Why did you not tell me I had to do such and such". He will then give me *that look* that says, 'I thought you knew you had to do that.' Gah! As Bridget Jones would say! Intuitive only works if you both are drawing on the same pool of experiences, otherwise, it is like trying to make water flow uphill. ;)
P.S. For those of us not in the biz, a glossary would be helpful. For example, GUI. I figured out the UI must be user interface, but the G has me completely stumped. And please do not tell me, 'I thought you knew what that meant. ;)
Posted by: Mary-Anne | Jun 16, 2006 3:30:46 PM
The results are sometimes ambiguous.
Posted by: Sahna | Jun 16, 2006 5:33:54 PM
"Intuition: it’s not what you think."
Anyway, another great post from you guys, I'm glad to see the bench has some depth. There are a few memes in here that will inevitably pop into my head when they are least welcome - and thus most needed.
Posted by: Kyle Bennett | Jun 18, 2006 2:12:16 PM
You know, the "grandma" test works for me... she hasn't seen sqwat! So, she's about as unpolluted a testbed as I get for truly novel stuff (and she forgets half of what she saw earlier... even better!).
So, if she can use it... I know I've got something reasonably simple to get the head around AND use. (sometimes too simple really)
Posted by: Gerald Buckley | Jun 19, 2006 12:17:41 PM
I like the term one of the posters used, "unaided interaction." That's when I can figure out how to accomplish what I need on my own without having to look at the documentation or call customer support! I also like the cell phone reference - most of those interfaces only a 10-year-old can immediately use! Several of the cell phones I've had put the ringer volume control hidden somewhere underneath some obscure menu option like, well, the icon that has little arrows going at a diagonal that mean increase/decrease volume only to the person who designed it - and the poor end-users who eventually figure it out... and can never remember the next time we need to use it, where it actually lives!
As a software writer and trainer, I personally judge a "good interface" as one that allows me to perform my workflow and accomplish my task without making me hunt all over a page (or multiple pages) for the options I need to do my (the user's) job - or have to interpret and guess and even try (with error results) what the developers/product marketing people decided to call various interface components - as I write the documentation.
One of my soapboxes concerns all the creative terms developers and product folks come up with for buttons to "commit" to the database. Whatever happened to industry-standard terminology for those widgets??? Where are the UI standards guides in these development environments? What's wrong with Apply (commits and you stay where you are) and OK (commits and closes the dialog box)? Those are standard UI terms and behaviors. Why can't we use them for Web apps, too? Why has the invention of Web UIs caused so much furor in terminology? If the action does the same thing as it does in a Windows app, what's wrong with consistency? Then, you don't have to rely on "intuition" - you can rely on "unaided interaction" because it does what you are used to experiencing.
Believe me, I have encountered many, many, many more nightmares than easy-to-use UIs.
I'm also always between the rocks and the hard places - as a user advocate - trying to concisely and directly guide a user through a successfully-repeatable task topic (what really works) - and the development team (what they think works) - and the product management folks (who have no idea what the user is trying to accomplish in the first place because they never bothered to ask).
It has NEVER ceased to amaze me how little actual user interviews and usability testing occurs in the real world (particularly related to large corporations' internal apps).
Hey, I'm not complaining - it's job security for me, for sure!
Posted by: Angelina | Jun 24, 2006 10:10:47 AM
Excellent post! However, how does one judge intuition then? I remember working with a gentleman once who said he scored high on an intuition test - I think it was one of those leadership tests. Are some people gifted with the ability to gauge intuition?
Look her --> Bayerischer Wald
Posted by: Bayerischer Wald | Jul 29, 2007 8:38:51 AM
The comments to this entry are closed.