UX Design Fail and How Not To
- By Lachlan Colquhoun
- September 17, 2018
For every additional piece of information which users are asked to put into an online form, there is a 10% drop off in the conversion rate.
It is one of the insights from Gartner's Research Director Magnus Revang at a Customer Experience and Technology in Sydney recently.
"If your alternative is for one email address, but you go for ten fields instead you have said no to 90 percent of your potential business," Revang said.
“And if you do that you are saying all the complexity we have in our systems, we are going to expose that to other users because it makes our job as developers easier.”
The Norway-based Revang’s presentation on “UX Disasters” highlighted some of the principals of good UX and outlined, as its title suggested, some disasters.
At the extreme end, there was the bad design of the warning button at a nuclear power plant which caused an accident, and then there was the faulty cursor on a missile system which created a civilian plane to be shot down.
There were also tragedies, such as the case of a warning button on a hospital case management system which was ignored by three nurses, with a total of more than ten years of experience.
The warning was to remind staff to give cancer patients hydration treatment after chemotherapy, but it was on a button positioned on the right-hand corner of the screen.
“Anybody would have missed that, I would have missed that, and you can say in this case that bad UX caused an unfortunate event,” Revang said.
"When the nurses looked at this they had so much information that they missed the most essential thing because the most essential thing was actually the smallest thing on the screen.
"But when people defend these applications they say ‘everything is available on the screen' or ‘these are super users' or ‘they have the training to use this, so there is no need for UX.'
“But if you are saying that, then you are saying that ‘I don’t need to do any design because I put all the responsibility onto the operator.’”
While most bad UX did not have such tragic consequences, it was often created out of the same flawed process: of experts and superusers creating designs without proper consultation with end users.
"You sit in committees and meetings, and every stakeholder has a different idea of what is important," Revang said.
"Lots of people from management are there, and they have said what the users need, but that is wrong because these people don't know what users need. The only way you can get that is by going out and shadowing these nurses and getting to the bottom of what it is they are doing,” he added.
Revang outlined several principles of “design thinking” in his presentation which would create the “clean path to success” through UX.
One was the “human rule” that “everything is ultimately human and social in nature.”
Then there was the “redesign rule,’ which is that you need to know what you are designing, and you need to be aware of precedents.
“There is no such thing as designing something new,” Revang said.
“Every single solution which you make for customers or internally is to support something that has already been done.
“So, we need to know what is being done otherwise we can’t design a better solution for it.”
Revang also described a “tangibility rule” which says that you cannot ask anybody about the solution based on “abstract thoughts or a textual description.”
“You need something you can discuss,” he said.
“How many times have you been in a meeting and people discuss something and say it's going to be ‘like that, that and that?’
“What happens is that everybody has a different view of what ‘that’ is going to be and you all go out and it starts being designed, but when you come back people say ‘this is not what we have discussed,' and this is because it wasn't tangible."
Revang was also critical of developers who, in his opinion, pass most of the work onto the user and in the process make it harder for them, as well as hinder conversion.
These developers “expose” users to the complexity of the systems which confuses them and, with a little more work, could be embedded.
"If I have the choice of doing a week on the extra development and that would save five minutes for my users, and I have 100,000 users, then that is almost a year of saved time," he said.
“So that is not having respect for your users if you don’t commit to making it easier, but that is contrary to what we learn in development when we are told we need a small and lean code base.”
Revang described a personal anecdote to illustrate another point on the hierarchy of functions.
All functions are not equal, he said, and developers need to ensure that the most likely and relevant function is given the greatest prominence.
In his home city of Oslo, there is a "genius" car parking app where users can set their parking time, pay remotely, and which even asks users if they'd like to cancel their parking when they start their cars and begin driving.
“But they made a big mistake on the UX, and it cost me 800 kroner,” Revang said.
Just as he was about to pay for his parking, a work colleague asked him a question across the office and distracted him, and he unknowingly pressed “regret purchase” rather than “ok.” It was a mistake which cost him a parking fine.
“You can say that all choices were clearly presented and I am an idiot,” Revang said.
“But the most likely action I am going to take is to press ‘ok,’ while only 0.1% of people will regret their purchase.
“Why are there the same buttons for all things even though there is a different percentage or likelihood of doing each thing.”
Finally, Revang urged designers to “design things inside out” and gave an example of a Norwegian bank, which had recently relaunched its website.
“These guys had their banner ads at the end of transactions, and all the marketing messages came up when people had finished what they were doing,” he said.
“These banner ads are highly contextual, and they have led to a 520% increase in people landing on marketing pages.
“It is like exiting through the gift shop because you never have the gift shop at the beginning of the tour, you go to the Eiffel Tower, and then you guy the statue."
"So, you front load the important things, and you don't make the mistake of creating detours, because it becomes an organic experience for a customer."