Hardware

10 things every UI designer should know about end users

An interface that fails to consider the user perspective isn't likely to win acceptance. Jack Wallen looks at user habits and attitudes that UI designers should keep in mind.

With Ubuntu Unity having its first anniversary recently and Windows 8 on the brink of release, it's becoming quite clear that not all user interfaces are created equal. In fact, it is possible to have a brilliant design that could revolutionize the way computers are used -- but if that design doesn't take into account the end users, that design will fail before it has a chance to prove its brilliance.

After years of using nearly every Linux desktop and each Windows desktop, hearing the broad spectrum of complaints, and speaking to both designers and end users, I believe I have a solid handle on what UI designers need to know about end users to create fantastic, user-friendly interfaces that can easily avoid the adoption hurdle. Note: This all applies to the AVERAGE end user.

1: End users do not like change

This is a most unfortunate reality with end users. Change is something that does nothing but keep them from getting their work done. The truth is, most end users live in a constant state of fear that they are going to break their computers or that if something changes, they'll never figure out how to do what it is they need to do. This means that any drastic change you make in the UI is going to cause backlash from users. Does that mean you can't change your design? No. It just means it must be done with care.

2: End users do not typically use keyboard shortcuts

I live and die by keyboard shortcuts. Most power users I know do as well. In fact, I do everything I can to not move my hand away from the keyboard (unless working with graphics). But end users are not built that way. They like the mouse and will always like the mouse -- until they are forced to use nothing but a touch screen. To that end, user interfaces should not be created that rely on keyboard shortcuts. Yes, it's good (and encouraged) to include them, but don't design your UI so that keyboard shortcuts must be used.

3: End users separate their mobile platforms from the desktop platforms

Unless you're Apple, you must understand this. Users separate their mobile interfaces from the desktop interfaces and do not expect them to be one and the same. It's clear that Apple is working its way toward making the IOS and the OSX interfaces the same. That's fine. Users have been dealing with the IOS interface for years. That translates into user familiarity. The Windows 7 mobile interface? Not only is it one of the least used interfaces, it's also one of the least mature interfaces. Users are not familiar enough with Windows 7 Mobile to all of a sudden have it become the UI for their desktops.

4: End users do not think like developers

This one is a challenge for developers. It's hard to think without the use of your own personal filters, which makes it difficult to put on the cap of a different level of user. It should be obvious that end users do not think like developers. They simply don't have the skill sets or the fundamental understanding of how computers and their interfaces work. What does that mean in the end? Keep it simple and keep it focused somewhere between the lowest and middle common denominator.

5: End users want their apps easy to access

This is simple: End users want launchers. This can mean icons, quick start launchers, start menu buttons, or panel launchers. In whatever form, users must have near-immediate access to their applications. I have noticed (when trying to support end users) that even having them navigate through a start menu, they quickly become confused. To those of us in this industry, the idea of Start | All Programs | Mozilla | Firefox is second nature. We navigate menus all day. To the end user, this can quickly become a complex maze of confusion.

6: End users adhere to "If it ain't broke..."

I can't tell you how many times I have heard this very statement from end users having to experience a new interface for the first time. They don't get the idea of evolving the interface. To them, everything on the PC is a separate entity and the idea that a program must make user of widgets and elements already in place is absolutely foreign. Only those in the know understand that when the foundation of a program evolves, so too must the program evolve.

7: End users needs slow, steady acclimation

Apple is doing this perfectly. Slowly but surely it is adapting its users to its design scheme. Eventually, OSX and IOS will be one and the same and the users will hardly notice the change. If you want to make a grand sweeping change, do it gradually and you won't have users losing their marbles.

8: End users do not practice safe computing like you'd want them to

To the end user, the computer is a vast playground where anything can be installed and everything is safe. We all know that is a fallacy and one that could land the end user in deep water with the IT department. You can't count on companies having policies that prevent end users from doing anything and everything they desire, so maybe it's time for designers to understand that end users aren't always to be trusted and that fact should be reflected in the design of the UI.

9: End users don't care about eye candy

I'll confess I love me some eye candy. I always have. I am part graphic designer, so it's in my blood to want things to look nice and have a bit of added eye candy value. The average user? Not so much. In fact, the average user wouldn't care if the interface still looked like Windows 95 -- so long as it was easy to use.

10: End users do have a valid opinion on how your interface should be designed

Let's face it: User interfaces are meant for end users. In a perfect world everyone would have the same level of knowledge and designers could design the most lovely, complicated interfaces they wanted. But the truth of the matter is, end users are the target and their opinions should be of high value to the designer. I know many desktop interfaces (such as KDE and GNOME) have design summits, where designers and developers meet to work together to design the most viable interfaces possible. Those summits should have numerous end users in attendance to help guide the designers down the perfect path.

Design success

UI designers face a hefty challenge. Not only are they charged with designing elements that work well for the program (or platform), they must do so while not making things too complicated or abstract for the end user. Ubuntu has faced this challenge over the last year and Microsoft is about to see just what Canonical went through. I hope Microsoft listened well to end users' opinions during the final phases of the design. Users are an important group that designers must pay attention to -- or else their designs fail.

About

Jack Wallen is an award-winning writer for TechRepublic and Linux.com. He’s an avid promoter of open source and the voice of The Android Expert. For more news about Jack Wallen, visit his website getjackd.net.

21 comments
bjmoore
bjmoore

The article and many comments made excellent points about the organization and structure of a UI but omitted discussion of a common UI flaw - readability. "User interfaces are meant for end users" should be expanded to also cover the physical needs of the end user and the intended environment. Many of our UI designers are hooked on the look of tiny size, lightweight sans serif font, in a pale grey or pale blue on a pale grey, blue or white background (Other than font size - does this sound familiar Tech Republic?). Sure it looks attractive on a large bright monitor - but how does it work for your target user base in the real world? For end users in their upper 30's and beyond - pale on pale is hard to read. Especially if the font is small and densely packed. The use of higher contrast and/or larger font size eases eye strain and allows the user to focus on the using the interface rather than struggling to read the interface. For applications designed for the rapidly growing senior citizen market - consider adding additional vertical space between links. Fine motor control diminishes as people age and tightly packed links leads to frustration and misclicks. Even if your target audience is younger - highly readable text keeps the user's focus on the content - not squinting to read low contrast headings and buttons. The end use environment needs to be considered as well in UI design. I often see pale on pale design on applications intended for outdoor use where the interface all but disappears in sunlight. Many medical applications are used in slightly darkened rooms and have their own set of requirements for good visibility. Useability in the intended end environments, whether office, sunlight or darkened rooms should impact the design choices made.

j_powell
j_powell

Speaking as a power user, in my workplace it's been an uphill battle ( but improving over the years ) to have user input inform design issues. This is more an awareness issue than actual desire to make things difficult. I would suggest that a similar article about what power users want and need would be useful. I don't think they want / need the same things as the average user. I can tell you, one thing power users want are simple ways to automate and schedule processes. They don't want to click, click, click ( dance, whatever ) their time away repetively to accomplish their end results. Living at the MS boundary where Access is too small, SSIS is restricted to the priesthood, and distributed queries is turned off is untenable. Ouch. Once an Oracle person visited with a user automation tool that looked like the promised land. I guess it lives somewhere else, unfortunately. I guess it was too expensive to retool our shop. I had simpler and easier to learn batch automation processes available to me in the 1970's than I have today. Personally, I think MS wants sustain the IS priesthood ( I guess that would translate as "Developer Community" ) for profit reasons. This involves ensuring that users cannot automate too much. Part of the engineering mentality is to keep it complicated, or rather, not see why it's important to make it simpler. Just want to give a suggestion for further discussion and exploration.

notreal
notreal

Sometimes, it dawns on you that the basic principles of a particular field are either not known or not established.l. I firmly believe this is the case with UI design. I've been an end user for decades, a software trainer from time to time too, and am presently working on a UI for a form with well over 100 input fields. I'd like to put forward the following as fundamentals: 1. An interface should be intuitive. That means that someone who has never used it before can do so. Measure the success of your UI by the number of times that a new user is halted by their lack of previous knowledge. How often do you 'just have to know' where something is located in order to be able to use it straightaway? When a user doesn't know, how long does it take for them to find it? Measure that. Test it. 2. I don't believe that users are best served by a tailorable interface. If an interface varies according to set ups and preferences it soon becomes impossible for users to be given simple advice on how to use it. With an unvarying interace, support from co-workers will be more prevalent. Long and complex explanations soon destroy the users confidence ("How come such a simple thing is so complicated?") in both their abilities to use the software and the ability of those teaching them ("Ah... now that doesn't work because... hmm... let me see... maybe you've got..."). 3. For tasks that end users use a lot, UI designers should flow chart the minute processes that a user will go through to achieve the task. As long ago as Apple's System 7 it had become very easy to organise files, navigate directories and find anything. That's now a much tougher process in almost all modern OS. 4. If there is always going to be at least some learning for an end user to do, produce a simple and learnable system rather than a complex one that tries to compensate for the failings of the lowest 10%. 5. Keyboard shortcuts? Handy if it is quick to find out what they are. Not handy if you need to pick up a manual or google for twenty minutes. Show them alongside their equivalent menu commands and everybody's happy. And if you think I might only believe the above because I'm too dumb to use a computer, be assured that I have done more than enough to prove otherwise. Really. One thing above all: Time spent learning a new system is unproductive, and people aren't silly for being averse to that. Aircraft designers used to have a saying, 'Give me enough power and I'll fly a barn door' . The equivalent for UI designers is asking users to spend time learning their product. Ask too much and there's a chance your product isn't worth the time that has to be lost in learning how to use it.

alamedaod
alamedaod

I am an Optometrist and use a vertical market package to run my practice. It's my fourth attempt at implementing electronic health records and management software my current attempt burned $150000 and 800 personal man hours (I do know what I'm doing). If I hire a high school graduate to make appointments it takes 2 months for them to get comfortable using the software. Designers give you the swiss army knife, a bunch of flexibility much of which is not supported by the documentation or help screens. While its nice to have a corkscrew, my user does the main blade, again and again, all day long. That Process, appointment making, should be honed to the extent that it operates like the single blade in switchblade. click...Done. Instead my user has to remember a long script and sequence of clicks and drop downs to get the job done. The task is not brain surgery but has as many steps as a lobotomy. Watching users do the same easy thing the same stupid and complicated way, again and again makes the observer want a lobotomy. Even a guy donig more variable tasks. programming or photo editing does the same thing again and again.....more switchblade less swiss army. Enough ranting for now.

Seotop
Seotop

after 11th: end users neglect details, but it's the goal to success for geek users )

alfred
alfred

The most important element of any development of a UI is that the user should have easy access to help. About 20 years ago help files were useful since they informed the user about how to do what they wanted. Now the help files have three serious failures. Firstly, especially with MS, they do not cover the item the user is seeking. Secondly they hide useful information under headings no user would think applied to their problem. Finally and worst of all they tell the user what can be done without a clear statement of how to do it.

vuurklip
vuurklip

The worst sin of the designer is to change the UI in order to achieve exactly the same result. Microsoft loves doing this. We therefore spend months learning repeatedly different ways perform the same tasks. Absolute hell for those who have to support end-users!

SirWizard
SirWizard

The interface should be adaptable to individual user needs. Not all users think or work the same way. What may seem efficient or logical, even to a great many users, may be nearly unusable to others, who thrive with different goals. Consider Microsoft Office and the ribbon interface, which for many is a pleasing combination, while for many others (I, especially) it's like an unfortunate combination of explosive diarrhea and locked pay toilets. Similarly, what seems to be a clever idea based on one design goal may be a constant irritant for some (too many) users who would prefer to trade off a different design goal. For another odd Microsoft example, consider toolbar buttons that appear in grayscale until the pointing device hovers over a button, which then appears in color. While that makes it easy to locate the focus of the pointing device, it loses the easy differentiation of buttons that a continuous color presence would provide, making it difficult to identify the desired UI element. Beyond this, for me, a user interface that cannot be customized is severely limited. I can always come up with a better layout and features FOR ME than what the UI designer (or far less clever management) provides.

JohnOfStony
JohnOfStony

Although I am by no means an Apple fanboy, I have noticed that Apple has never made a radical change to its user interface so I particularly like point 7. I don't know what it is about Microsoft that makes it feel a need for change when there is no need. Two UIs from Microsoft that I consider real steps backward in terms of usability are the Ribbon Interface in Office 2007 onwards and the Metro Interface. I know it's said that you can't learn from others' mistakes but Microsoft should take a look at Coca Cola's major mistake when they changed the formula for the standard drink. They asked lots of drinkers whether they liked the taste of the new formula; what they didn't ask was whether the drinkers preferred the new to the original. This was only discovered when, after changing over completely to the new formula, millions of Cocal Cola drinkers rejected it and asked why they couldn't buy the original formula. Coca Cola learned from its mistake and reverted to the original formula. Remember, Microsoft, making a U turn isn't a sign of weakness. It's having the courage to admit that a mistake has been made and that you're going to correct it.

bobc4012
bobc4012

Great article. It essentially puts in one spot what many people have either been saying or implying ever since Unity and, now, Metro "hit the fan". BTW, this is especially true for "seasoned citizens", most who are "technically challenged". It also holds true for many of those who need to get productive work done rather than learning "new tricks" every couple of years. My youngest son works in the medical field. He watched as a lot of doctors jumped on the I-pad bandwagon and were going to use them to record their patient histories - after much frustration, nearly all went back to their laptops. Nothing like a real, usable keyboard for data entry.

Deadly Ernest
Deadly Ernest

given for dumping a particular OS for another - and the big one is the familiarity. I lost count of the people I talked in to using Kubuntu to keep their very old machines running, only to have them call me back a couple of years later to explain why their nice Kubuntu has gone crazy because they version updated to a version with KDE 4 and they HATED it with a vengeance. A quick wipe of the system and an install of an earlier version of Kubuntu and disabled the update manager. I recently found Zorin OS 5 and have been busy bringing them up to date with that. making them all happy to have a newer version of Linux that's what they like. Heck, a few have had me install the Win 2000 GUI in Zorin OS 5 for them as it's SOOO much like the Win 95 and Win 98 they learned on and they learned to love.

the_tech_mule
the_tech_mule

I don't think there will be fully convergence between iOS and OS X as is suggested in the article. I think there are going to be a lot of common design elements between the two OSes, but I have to think that Apple realizes that Macs are not just giant tablets. Now, Apple may move to LaunchPad as the default starting point, but I think they'll continue to make the dock and desktop available for a long time to come. I do recall that Steve Jobs was against anything on the desktop (he wanted it completely clean), but he lost that battle and I think that will continue to be the case for a long time to come.

TsarNikky
TsarNikky

Excellent observations, based on a lot of actual experience. Why did Microsoft chose to ignore so many, if not all of them? No wonder Windows-8 is not well received by non smartphone and tablet users.

wizard57m-cnet
wizard57m-cnet

when my kids were young, I spent COUNTLESS hours pleading, begging, threatening, literally whatever I could, to try and keep them from touching the screen on my monitors! Didn't matter which computer it was...the old Amstrad PC, the IBM PS2, my home built 386 right up to present day..."Please, kids! Don't touch the screen!!!"...never worked, I could always find little smudges all over the monitors. So, to say that users forego "touch screen for the mouse"...nah, not without a lot of training!

Deadly Ernest
Deadly Ernest

fonts or easy to work with voice command software.

alfred
alfred

While I agree with most of what you say, as an engineer for 60 years I think I know a bit about engineering mentality. The many I have met had a logical method of thinking which led to clear conclusions. I also met designers, most of whom needed to express their personality in their design, which led to clashes of understanding. I feel that it is designers not engineers who create the UIs being discussed.

SirWizard
SirWizard

Your item 2 states, "I don't believe that users are best served by a tailorable interface." The most extreme example, ala Microsoft, is the menu that changes every time the user selects a new command so as to favor a "most recently used" metaphor. It's horrible trying to find features in a constantly changing UI where the user isn't the one doing the tailoring. But I disagree with the statement for less radical implementations. Certainly, a large fraction of typical users would find tailoring a reasonably well-designed UI to be beyond either their interest or capability. Software support remains straightforward. However, for the power user, an immutable UI can be utterly crippling. But they (include me in that group) understand the tradeoff. I'm happy to trade away availability of easy support for access to hyper-productive usability courtesy of my custom-tailored UI. The disadvantages of a tailorable interface are mostly self-limiting either way. But what does tailorable mean? Turning a few toolbars on and off? Adding a few new elements while eliminating a few unused ones? Overhauling the whole thing to run custom designed add-on applications? Finally, consider Microsoft???s Ribbon UI travesties (see Office 2007 or 2010 for examples). They virtually require users to tailor the so-called Quick Access Toolbar (really Less-Slow Access Toolbar) to avoid clicking to death or suffering carpal tunnel syndrome.

Ernesto.Guiterman
Ernesto.Guiterman

Excellent comment (improvement?) to an also very good article. Thank you for sharing your experience. I agree 100%

Deadly Ernest
Deadly Ernest

All the UI designers claim their UI is intuitive, however, they never say WHO it is intuitive for. What is intuitive for one is counter intuitive for another. How we perceive things is as much decided by our cultural and social background as anything else, even our sub-cultural background. I once was involved with a bunch of people working out how to do menus for an app we were putting together. Some were IT software people, some were IT hardware people, some were teachers and some were writers. What was interesting is that the layouts and menu organisations done by each group had a few things the same, but there were many more that were very different between the groups while very similar within the groups. This was because all the groups were used to looking at things differently and thus what was intuitive for one was not for another. Windows 8 Metro is a very good example of this problem. Those who use touch screen devices and apps all day find it very intuitive, while those who rarely or never use a touch screen find nothing about it intuitive. When designing a menu or UI for any group, you need to involve some people from the intended end user group so you can design it to be intuitive for the intended end users. Or design it so that it can be heavily customised post sale so the end user can make it more fit their own personal needs.