Darren Wilson
How is it that some products and digital interfaces can end up so poorly designed? The world is littered with such examples that are amusing and frustrating in equal measure!
Microwaves are notorious for this. This model below has 31 buttons, which is a little over the top when all I really need to do is adjust the power level and time required.
Kitchen appliances are clearly a theme. My experience with this slow cooker was made so much worse when i couldn’t work out whether it was on or not. Just put a light on please! I can’t leave the house till it slowly warms up. Not helping my morning much...
It's not just electronic devices that this applies to of course. Check out these new scissors. How should I open that vacuum sealed packaging? With another pair of scissors maybe? The mind boggles.
Designers can become so engrossed with preconceived ideas of the look and feel of a product, they forget to solve some fundamental questions. Crucially these questions revolve around the user and not the designer's view of the world.
What do they need?
What do they want to achieve?
How will they use it?
What are their cognitive and physical limitations?
This cartoon illustrates the problem perfectly! [1]
Ultimately products will be used by real people, so the design process should also start with real people. User centred design (UCD) is the philosophy that positions the user at the centre of the creative process.
Throughout all of the development phases , there is a focus on a deep understanding of who will be using the product in order to optimise the design for those users.
The international standard 13407 is the basis for many UCD methodologies [2]. It can be embedded into any form of project management process such as AGILE, Waterfall or otherwise, so there really is no excuse for not applying it.
In the 1980s, Donald Norman advocated the practice of UCD, promoting that such a philosophy was needed to make products usable and understandable.
As a result, users will effortlessly and instinctively know what to do, and will be less likely to become frustrated with the product [3].
If that is the case, then the product has a much greater chance of being a success. It's as true today as it was then.
“Good design is actually a lot harder to notice than poor design, in part because good designs fit our needs so well that the design is invisible.” (Donald Norman).
Whether designing a physical product or digital interface, the process is the same. UCD starts with the user, by establishing their needs and requirements, and where and when the product will be operated.
Techniques including focus groups, questionnaires, interviews, physical measurements and modelling, observation and ethnography may all be used.
UCD also ends with the user, by field testing the product to validate those user needs and requirements and to ensure the product is desirable, attractive, usable and accessible.
I don't believe it needs a rocket scientist to work out whether this piece of genius design is meeting user requirements... the loo for two, discovered at MIT no less.
Where was the UCD? What does the user need? What do they want to do? How will they use the product? There are no words!
How to follow a user-centred design process
Variations of specific details of the UCD process exist, but a common thread persists where the user is involved in all design and evaluation phases [4].
UCD is iterative, allowing product design to evolve and improve until reaching an optimal state. It is typically categorised into four phases:
Context of a product is established first. For example, when designing an in-vehicle interface, factors such as reach, readability, vision angle, cognitive workload and potential distraction from the driving task are to be considered. Design of mobile apps demand a very different scrutiny.
User requirements are then created and agreed by stakeholders for clear definition to design teams. A number of techniques can be utilised, such as focus groups, interviews or user observations. All too often these critical requirements are poorly defined introducing costly changes later.
Design can now actively begin and will be continuous throughout the product development lifecycle. This phase may also be subdivided as the design evolves from concept to completion.
Evaluation follows the design phase and a number of iterative design / evaluation cycles may be needed to improve the product, pending feedback. Testing realistic prototypes is preferred, subject to time and resource. Lo-fidelity pen and paper prototypes still have a role, although hi-fidelity rapid prototyping is often required later in development to simulate representative designs and complex system interactions.
As product design reaches completion, the need to be test in situ becomes more significant as some issues may not be prevail in short desktop based studies. They may require users to ‘live with’ a product in real world situations for those issues to surface.
Acting on recommendations from evaluation is crucial for evolving design. Test protocols must be designed appropriately to answer open design queries with reliable and unbiased data.
Good moderators also unearth issues by observing and listening to participants, following up on interesting comments without ‘leading’ them or introducing bias.
Verbatims provide valuable information and clues to problems that numerical data might not (i.e. subjective ratings or task times).
What can happen if the user is ignored?
I wonder how much evaluation and UCD went on at Microsoft for Windows 8? Upon removing the 'Start' button there was a social media storm and a consumer backlash.
After some red faces and backtracking it was re-introduced for Windows 8.1. The shut down process is not much better.
This happened when Microsoft wanted to develop a platform that worked on all its devices from smartphone through to PC.
Unfortunately the metro interface which was designed to be touch friendly was in effect the same underlying system (more or less). The fusion created chaos.
Microsoft isn't alone in UX design calamities. Apple has been at the forefront of great design, yet iTunes remains terrible. Originally a mechanism to quickly download music, it became a monster.
Quality of the music experience has been diluted in adding masses of functionality.
Major issues include:
An unwieldy interface that is unnecessarily difficult to use and infuriatingly slow
Apple tried to crowbar way too much functionality into a single application
Lack of integration with social media
iTunes has for a long time been the great faux pas, perhaps only challenged by Apple Maps. It is designed around a business model rather than the needs of the user.
Perhaps in a time where streaming content is the preferred experience, the days of iTunes could well be numbered. As a user of Apple products I do hope so.
Measuring success of UCD
When evaluating a product, it’s important to consider if user requirements are met, the product is intuitive and usable, and evokes positive emotive responses such as user satisfaction and delight.
In some cases, it might also be an opportunity to observe if there are any potential adverse effects on the user's safety or well-being in a controlled environment.
For example, simulations are often used to assess whether in-vehicle displays are distracting while driving.
Criteria for success will vary depending on the product and its intended context for use. Core to whether a product will be successful is it's ‘usability'.
Major factors that contribute to this include being: Learnable, memorable, effective and efficient. Desirability, usefulness and delightfulness may also be used as measures for how a product resonates with users [5].
Errors will naturally happen when interacting with a complex product. That is symptomatic of human behaviour.
However, practising sound design principles and making good use of evaluation feedback will alleviate this by minimising the proliferation and the impact of such errors.
Yes, it’s worth the effort!
It is often clear when interacting with a product or interface whether UCD has been applied. Remote controls are a bug bear of mine.
As users we take them for granted, but it's easy to get the design very wrong and it seems like I am not the only one who thinks so [6].
Take the Sky+ model, which sits wonderfully in either hand, allowing for comfortable, single-handed, often blind usage with minimal frustration. Frequently used functions are easily accessed and reliably operated.
Buttons for non-critical, often mind boggling features are thankfully sacrificed and accessed via on screen-menus.
Apple have a completely minimalist approach, albeit with a different functionality. The product is beautiful in hand and a joy to use.
Compare to some standard models which suffer from ‘feature-itis’ with buttons all over the place and little thought to ergonomics. Then there is the Sony remote for Google TV: Designed for a brain surgeon?
Enough said. The difference is startling. As technology advances, the challenge is to retain design simplicity despite increases in functionality.
Search mechanisms for on-demand content are a prime example, and although voice-control is a supportive interaction for that task it is not desirable for everyone.
Conclusion
There are undoubted benefits of working to a robust UCD:
Products will be easier to operate for users who will encounter with fewer errors and have an easier learning curve
Users will experience greater satisfaction and more trust in the product
Development costs will decrease with late change reduced
Customer retention and loyalty will strengthen
Brand credibility will improve and sales will increase
Working directly with users will aid design of the best possible customer experiences but it’s not always easy. Interpretation of test data can be an arduous process.
Participants can be notoriously ambiguous and are affected by fatigue and general emotional state. Furthermore, individual differences can produce conflicting results.
Nonetheless, it is worth the journey. As Faulkner stated, working with users can give you problems, but the alternative of ignoring them would be foolhardy and should never be an option [7].
References
[3] Norman, D.A. (2002). The Design of Everyday Things. USA: Basic Books.
[5] Krug, S. (2014). Don’t make me think. Revisited. A common sense approach to Web and Mobile Usability. USA: Pearson.
[6] Rogers, Y., Sharp, H. and Preece, J. (2012). Interaction Design: Beyond human computer interaction. 3rd Ed. Great Britain: Palsgrave.
[7] Faulkner, X (2000). Usability Engineering. Great Britain: John Wiley and Sons Ltd.
Get in touch with the author
Darren Wilson
Managing Director at UXcentric
07854 781 908
コメント