5

Are there any studies showing that a design that replicates users's expressed wishes is worse than one based on conventional UX wisdom?

Like, users saying they want as much information on the screen as possible, but when asked to find information or read the cramped page, they have difficulties to find the next row due to too little leading (row distance).

I've seen quantifying field study ROI and case studies to help sell UCD, but they do not deal with the user wishes.

If you're interested, this is the background of my question: While migrating an existing product to a new design, I often hear that "the customer wants this design", where I do not believe "this design" to be good. It's difficult to get acceptance for User Research, because the customer has already told us what they want. Oftentimes, my suspicion is that "this design" is requested because that's the way it is done in the old version of the product, and thus, we do not get a chance to dig down to the real pain points to come up with a good User Experience.

  • Not an answer, but we often get clients that ask for some poor design on a system they want building. We recommend a "better" way for it to be designed and they decline. 6 months later they come back asking for an update to change the design to our original recommendation... I'd be interested if there's any studies or solid data on this so it can be presented to the client first time round! – crazyloonybin Sep 7 '17 at 15:49
  • There are some, but I don't remember any specific study right now. However, most research you have done should have shown this behavior. In your user case, the user should not be named as user (well, not end user at least), but stakeholder. And we all know stakeholders have biases, and pre-research and client's brief is not a substitute for usability tests. – Devin Sep 7 '17 at 17:29
1

a design that replicates users's expressed wishes is worse than one based on conventional UX wisdom

Yes, it is a known fact that it is difficult to get reliable feedbacks from users when it comes to preferences. A metaanalysis has been conducted, see Measuring usability: preference vs. performance by Jakob Nielsen and Jonathan Levy.

The most probable issue when asking about preferences is that, there is a change biais involved and users will prefer the old way because that is the way they know.

  • Old, but still the only data-driven result. Interesting is the observation that expert users seem to be even more attached to an existing solution than novices. Thanks! – virtualnobi Jul 9 '18 at 17:43
2

Oftentimes, my suspicion is that "this design" is requested because that's the way it is done in the old version of the product, and thus, we do not get a chance to dig down to the real pain points to come up with a good User Experience.

That, plus it's expensive as hell to start over with new designs and new data models and code base and front end, and then coach internal users and customers on how to use the new thing, and then recover the lost business from the customers who may decide to leave because they liked the old version of the product the way it was.

It's a fact of life that,m people resist change and software users prove it by taking their business elsewhere when the furniture is rearranged on them arbitrarily. (Basecamp’s innovative way of handling this is by supporting whatever version you signed up for - for eternity.)

Another big problem with user research is that it often fails to account for a breadth of possibilities constrained within practical reality. It's too often all or nothing, designers want something made to spec with no deviations. But that’s as rigid as the most egregious waterfall process.

Split the difference. Find something in between starting completely over $$$$ vs the "legacy," and provide that option as part of a range of possibilities that differ in their risks/cost but come from the same research conclusions.

And read Steve Portigal's interview about going from research to results, too. He does a better job answering this question.

  • Nice rant, Luke, but I'm in a big company and sometimes we have the choice (or should I say, luxury) to redesign an existing process (with an existing product) on a completely new platform. So all the "new data models and code base and front end" is needed anyway, but we can do it differently from last time. At that time - do I follow the customers' "we want this design" or do I create one myself? – virtualnobi Jun 27 '18 at 14:29
  • It's been a while since my rant so I'm having trouble relating what I wrote then to this. Now with fresh eyes, I think I'm hearing you say you have an opportunity to start over and redesign, and you're conflicted about letting what is characterized as "what the customer wants" dictate how that is approached. If that is accurate then I think it's your job to do some research and contrast what the customers (or their proxies) say they want with what careful observations suggest they may actually need. Disentangling the solution from the problem is especially critical in a big company. – Luke Smith Jun 27 '18 at 17:25
0

I think this also can be a helpful link. In "First Rule of Usability? Don't Listen to Users" isn't many numerical data, it is kind of summary of observations through years that telling us how badly users are in predicting what they need.

  • Thanks, Ada, for the additional reference. Interestingly, the bottom cites the "Measuring Usability" that @asiegfried already mentioned. – virtualnobi Aug 19 '18 at 20:55
0

I think you'd be hard-pressed to find any evidence that implementing things your users ask for is a bad thing. It's a mediocre way of doing business, but it's far from ineffective. If what you're looking to do is take your product beyond mediocre, you'll need to innovate. There is plenty of evidence that big risks don't always pan out...However, in the long-term, that way of developing products can pay off big time. (Apple, Amazon, Netflix, Airbnb, Facebook...etc.)

Our job as UX professionals is to clearly illustrate the benefits, risks, and trade-offs of ideas that we want our clients, bosses, and stakeholders to invest in.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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