I'm a big fan of About Face 3 by Alan Cooper and similar books about interaction design and how the software must be done correctly.

At the same time, working as a freelancer, I observe daily my customers, i.e. in general people with no technical background, asking to create products which violate every best practice and every rule of UX to finally obtain a mess I would never use myself. They are happy with it, and some will be angry and lost if they were forced to use a product designed according to the UX rules, scientific research¹, etc.

The difference between, on the one hand the rules and guidelines advocated by extremely competent and smart people, and on the other hand the users which often reject the products which follow those rules and guidelines, surprises me.

I imagine that this difference is too complicated to be explained in a short answer. Therefore, do you know any books or research papers on the subject, helping me to understand both why there is a such difference, and how to make products which will at the same time follow the best practices and not be rejected by people with no technical background?

¹ An example is a ribbon implemented in more and more Microsoft products. While created based on studies about the usage of software, productivity, etc., ribbon is perceived extremely badly by many users.

  • 2
    A frequent problem with changes to extremely common software like Microsoft Office is that users have come to learn to interact with a program a certain way, and suddenly you've changed that, perhaps drastically. Lots of users reacted negatively to "learning" GUI based office apps because they had spent so much time learning awkward keyboard shortcuts for command line text editors, even though the same functionality was often available in the GUI application, often also via keyboard commands. Don't have a link to research though so no answer. – Ben Brocka Oct 2 '11 at 4:57

There is no solution to clients who do not want to listen to UX advice. Just like, as a developer, there is no solution to clients who know how I should do my job.

I would make two suggestions: Firstly, can you get some certification in UX, which means that you cna wave that at clients and explain that you do know what you are talking about ( you don't need certification to know what you are talking about, just to prove it to clients ). Secondly, when they suggest something wrong, take some time to find web site indicators as to why it should not be done that way, and send them these - on a case by case basis.

It all takes time and effort, and it may not work, but it means that you have done what you can. At the end of the day, some clients are stupid, and there is nothing you can do.


If it's not accepted by users than it's not a good UX. Besides "rules", research and best practice studies you'll always need to attach your audience emotionally. Maybe Donald Norman's "Emotional Design" would be a good start for that topic. Even though it's not about interaction design paradigms. But the world of design is transboundry in my opinion.

A quick example with a Jesus style metaphor to point out what I'm trying to tell here. :)

Consider designing a car only by functionality. For example you could throw a piece of clay in a wind tunnel. And let the wind shape the car body. I'm pretty sure it will look kind of shitty in the end but has perfect aerodynamics. Now think about looking at a row of Porsche, Ferrari, Lamborghini, Nissan GTR and that windshaped thingy. So just looking at the cars is the first user experience. And it will feel fast, fast, fast, fast and weird.

So in the end user experience is not only created through (sure relevant) research facts but also mainly through well thought and emotional design.

  • If it's not accepted by users it's not good UX for them. Your real world users may have biases that didn't show in research samples. What may be amazing UX to novel users may frustrate the crap out of previous users of a product. – Ben Brocka Oct 2 '11 at 19:03
  • True. Your statement doesn't conflict with mine. Products should definitely be awesome for new and previous users. – erikrojo Oct 3 '11 at 10:14
  • The difference is new users don't always have to "suffer" because old users prefer things a different way. It can be more complex to implement but a degree of customization can be allowed to keep a "classic" interface or use a new one, similar to how Windows 7 lets you use the old Win 95-esque GUI stylings. – Ben Brocka Oct 3 '11 at 13:54

We often use UX and "user interaction" interchangeably, but I'd like to rephrase your question this way: "Why software with good user interaction is not always well accepted by users?" Now the problem is easier to spot: a user's experience involves more than interaction. The software has to do the right thing (purpose), it has to do it right (correctness), and should not require from the user any effort which users are not prepared to give to it (interaction, ease of learning, past experience, signing off your arm and leg in costs, all other barriers to entry). As soon as one of those elements does not work, the whole can fall apart.

Bottom line: You can do all changes you want as long as users are willing to climb the barrier. If they are not, then usability alone will not save you.


Can't give you a response to your question re books, but in response to your footnote I've looked at the research conducted for the Microsoft Office Ribbon and it's pretty dubious - so just because something was based on research doesn't mean it followed correct and rigorous research synthesis practices.

Not the answer you're looking for? Browse other questions tagged or ask your own question.