Good enough design
8 minutes read
Albert Einstein famously said that you should make things as simple as possible and no simpler. Now he was no doubt referring to something scientific but the premise holds equally true in the world of design and in a nutshell embodies the principle of good enough design. A design should be as simple as possible and no simpler. It should be simple, but good enough to meet the demands of users, good enough to get the job done. In effect this is also the approach that nature takes to design problems. Organisms evolve until environmental challenges are met; until the organism is good enough to survive. OK so you get some design anomalies, like the appendix (obviously the medical not literacy kind) and men’s nipples, but by and large nature is pretty good at good enough design. So how might you go about replicating this miracle of nature? Well first it probably makes sense to look at a few examples, this time from the world of design.
Some examples of good enough design
My first example of good enough design is the Flip video camcorder. For those of you that haven’t encountered this little gadget before it’s basically a simple point and shoot video camcorder (it’s the one in the picture to the right). You point it at what you want to shoot, press the big red record button and then upload what you’ve shot using the inbuilt USB key to your computer. There’s no zoom to worry about (its fixed focus), no complex settings to fiddle with and all the complicated stuff like editing and organising videos is done on your computer once the videos have been uploaded. The design is streamlined and ingeniously simple, but good enough for the vast majority of people who just want to shoot some video without having to worry about the sea of buttons and options you find on most camcorders.
Now I can guarantee that you will be familiar with my second example of good enough design because it’s the most used page on the Internet – namely the Google search page. The page is basically a big fat search box, with a few links to other useful Google stuff, like Google maps and Gmail. It’s simple but good enough to get the job (of allowing people to search for stuff on the web) done. Apart from a few extra links here and there the page hasn’t changed in donkey’s years but then why should it? It’s simple, but good enough.
My third example of good enough design might be slightly more controversial. It’s a site that has taken quite a lot of flack for its design over the years, and has spawned more suggested makeovers than I’ve had hot dinners – namely Craigslist. Designers hate the design so much that they’re forever suggesting alternatives, here are just a few from Wired, Smashing magazine and Design Eye. All good stuff but there’s a reason why Craigslist is still fantastically popular (its ranked 32nd on the Alexa Top 500 global sites list). Sure the site isn’t the best looking out there and it’s not exactly feature rich but the design is simple and good enough for the job.
Creating a design that’s good enough
Creating a design that’s as streamlined and as simple as possible, whilst still being good enough to meet the demands of users isn’t as easy as you might think. It’s not just a case of reducing the design timeframe, or making the design as minimal as possible. There’s a lot more to it than that. Here are some of the steps that you can take to help create a good enough design.
1. Find out what constitutes good enough
Before you begin you need to determine what constitutes good enough. What sort of user experience does the design need to deliver? What does the design need to allow users to do? How well must the design perform? These can be tricky questions to answer because if you simply go out and ask users they’ll invariably tell you that they need something that has all the bells and whistles imaginable, and does pretty much everything they could think of. Rather than investigating what users want it’s better to investigate what users need. What’s going to be most useful to them? For example, I doubt that going out and asking people what they would like from a camcorder would have resulted in the Flip video camcorder.
Of course what constitutes good enough very much depends on what you’re designing, who you’re designing for and what competition you’re up against. In some instances you might even decide that good enough just isn’t good enough. For example, if you’re releasing a new version of a product you probably don’t want your customers complaining that lots of features have been removed because you deemed them unnecessary. Equally a good enough design doesn’t have to mean a design that has a very rudimentary user interface. You might determine that good enough requires a design that delivers lots of features and a very simple but rich user experience. After all, if you were designing a mobile phone to compete with the iPhone something that just allowed users to make telephone calls and save their contacts certainly couldn’t be called good enough.
2. Set targets
Once you’ve determined what constitutes good enough for your design you need to try to quantify this into tangible targets. How usable does the design need to be? How desirable does it need to be? How does it need to perform against the competition, or even against the current design?
If you can’t set targets then how will you know whether the design is good enough? For example, you might decide that good enough means that 90% of participants can complete key tasks during usability testing. Alternatively you might decide that good enough means that having used your design and a competitors, a majority of people prefer yours. The key thing is that you have something to aim towards. You have something to tell you when your design is good enough to be launched.
3. Determine the key goals and tasks
Keeping things as simple as possible starts with determining what the key user goals and tasks for the product are. What must users be able to do? Why will they be using the product in the first place? For example, two of the key user goals for a camcorder are to record stuff and to share videos with friends and family. By distilling your product down to its essence you can not only determine where the design really needs to perform, but more importantly determine where the design doesn’t need to perform. For example, there might be features that you can throw out, or parts of the design that can be further simplified and streamlined, whilst still retaining an overall design that’s perfectly good enough.
4. Determine the key users
One man’s trash is another mans treasure and what is good enough to some users might not be good enough to others. This is why it’s so important to determine the key users for your product. Who really matters to you? Who must be able to use the product?
In his new book, Simple and Usable Giles Colborne divides users into three general types:
“Experts are happy to explore your product or service and to push the limits of what it can do. I call the next group willing adopters. They’re tempted to use something more sophisticated, but they’re not comfortable playing with something entirely new— they need to be given easy ways to adopt new features. The vast majority of people are mainstreamers. They don’t use technology for its own sake; they use it to get a job done. These are the people who say, “I just want my mobile phone to work.” Most people fall into this group.” Simple and Usable by Giles Colborne
It’s important to determine where your key users fall so that you can gauge user needs and expectations. For instance, what is likely to be good enough for a mainstreamer is unlikely to be good enough for an expert.
5. Strip out any unnecessary noise
I’ve found that one of the key parts of creating a good enough design is stripping out any unnecessary noise. Noise is basically stuff that isn’t really needed, and ultimately just gets in the users way. It might be features that aren’t really needed; text that doesn’t need to be on a page; or even aesthetic styling that ultimately doesn’t enhance the design. For each part of a design you should ask the same basic question – would the design still be good enough if this was gone? If the answer is yes then it’s likely to be a sign that this part of the design is merely adding unnecessary noise and can be stripped out.
6. Test your design
OK, so you’ve set targets for your design. You’ve determined your key goals, tasks and users and you’ve used these to create an ingeniously simple and streamlined design. But how do you know whether the design actually meets those targets? How do you know whether it’s good enough? Simple, you test it. Testing is the most important part of creating a good enough design because it tells you whether the design is good enough or not. How you test your design somewhat depends on what targets you’ve set, but will typically involve getting feedback from one or more of the following:
- Usability testing – There’s nothing like observing users as they use your design.
- A/B testing – For example, you could test the conversion rates for the current and new good enough design.
- Beta testing – Get the design out into the (semi) open and see whether users find it good enough.
7. Stop (when your design is good enough)
Leonardo Da Vinci supposedly said that art is never finished, only abandoned and the same can be said of design – there is always something that could be added, an area that could be tweaked and a part of a design that might be enhanced. The temptation is to carry on tweaking, adding and enhancing until a design is as good as it can be (given the design constraints of course), but you must resist. Once you’ve tested your design and it’s been deemed good enough stop. It’s as simple as that.
Going agile
Given its focus on frequent, smaller iterative releases you’d have thought that a good way to drive a good enough design is to go Agile. And you’d be right. Agile and good enough design are a great combination, as is Agile and UX in general. Be warned however that there are a number of pitfalls you should look out for. Firstly whilst Agile certainly encourages you to focus on the key user goals and tasks (usually in the form of user stories), often these are the key goals and tasks from a product owner perspective, not necessarily a user perspective. Furthermore, Agile can often result in a design that’s claimed to be good enough, but in actual fact isn’t. The design ticks all the boxes (i.e. the acceptance criteria have been met) but when users start using it they find it wanting. This is often because the design hasn’t been adequately tested with users, or insufficient user research has been carried out to determine exactly what constitutes a good enough design.
