Design is not a Super Power
A designer is not a lone hero with ready-made solutions.
If you put a cart before a horse, the horse wouldn’t be able to pull the cart. That’s what I imagine whenever I see ready-made design ideas being asked for when pitching for new business.
It’s a good metaphor for user experience design (or lack thereof). That cart might be the most beautiful carriage ever magicked out by an artistic fairy godmother, but if it’s not pulled by the horse that is proper UX work, then it ain’t getting anywhere. And proper UX work requires both preparation and iteration.
Design can be art, sometimes. And by “art” I mean the idealised notion of it — a spontaneous act of genius arising from pure inspiration. But the rest of the time, it’s industry.
Teamwork and preparation are key
“Wait, your title promised something about heroes and super powers, but carts, carriages and horses belong to a different kind of Disney story.”
Right, let’s rectify that.
So let’s picture the idealised designer as a hero, cape and all. That’s the kind of image popularised most famously by the cult of Apple. As Jeff Gothelf and Josh Seiden wrote in Lean UX, the expectation there is that the “heroic” designer simply ingests some requirements and then spits out work that makes people go “ooh” and “aah”.
Unfortunately, that’s more fantasy than reality. For one, something beautiful might not actually function well. And what is beauty anyway? It might depend on context as well as the beholder’s eyes. This is where I bring in the notion of super powers — unlike those convenient things, there is no surefire solution that can be pulled out in a clutch when designing. A designer needs preparation and, probably, help to create a solution.
Now, this isn’t to say that many designers don’t prepare. They do typically start by doing research, at least by looking at examples of similar designs. But this phase should be extensive and, ideally, include primary research like talking to users. And that makes it less likely that a designer can work alone. Collaborating with researchers or UX professionals who are trained in research could help designers navigate this phase in a more thorough manner.
And research is important because it helps ensure that the design will create a better UX, which is good for conversion and retention. Also, if you want to create a habit for users to use the product or platform you’re designing, it would be helpful to understand the internal triggers that can spur them to use it. Sufficient research gets you much closer to understanding what’s needed to meet users’ needs and create good experiences, instead of you simply trying to guess.
This turns design from something arcane into something grounded. As the philosopher Jean Hyppolite remarked, “the only secret is that there is no secret”. It might seem less sexy, but it’s a much more reliable way of doing it.
The work is always in progress
Remember the horse and cart? Now, if we have the horse before the cart, ready to pull, we still won’t arrive at the destination immediately. We’ll need to go through a journey to get there. It’s the same with design.
Moving away from the idea of a hero designer also means moving away from Big Design Up Front (BDUF). If we can’t bet on conjuring surefire designs out of pure genius, then we can’t expect to get designs right on the first try. Chances are, a design needs multiple iterations before it can be good. So it’s best to find out how well it’s working as early as possible.
Usability testing involves getting users to try a prototype of the product or platform with minimal guidance and getting insights on their experience through observation and follow up interviews. It allows us to make improvements that we can, ideally, test again until we know the design really works. It’s a staple of UX design, but it’s still prone to BDUF if you try to design everything before testing.
There’s always the risk that our design approach has problems we’re not aware of, and if we only test after everything is done, we might have to do extensive redesigns. Hence, increasingly, experts like the authors of Lean UX recommend that designers test early and often. Break the product up into chunks to test as soon as something is ready, and don’t rush into creating high-fidelity prototypes. It helps that you don’t need a lot of users to test at once, so you can just test with four or five of them each time. This enables multiple iterations to be done more quickly, getting us closer to a good solution in a more economical timeframe.
Such a process is, not coincidentally, part and parcel of the agile methodology, which is in opposition to BDUF.
But that’s not the end of it. Even after the product has launched, it should continue to be improved based on real-world insights derived from data sources such as analytics and user feedback. After all, a test environment might not perfectly replicate real situations.
Sometimes you gotta meet users halfway
We’ve talked about how good solutions are not created overnight or through genius alone. Now I want to debunk the idea that “good” means perfect for the user. The funny thing about this is the solution alone is unlikely to solve all users’ problems.
Let’s say we’re designing a product and, after multiple iterations, we’ve arrived at a design that’s as usable as it can be. Does that mean all users will find it easy to use? Probably not. And there could be cases where, no matter how good the design is, especially given technological limitations, many users may still face challenges in using the product.
A good example is chatbots. I’ve worked on some and reviewed their accuracy based on real inputs from users (i.e. what users actually say to the bots). Some users tell stories or combine multiple questions into one query, expecting the bots to understand like a human would. But, given the capability of the AI powering them, these bots just can’t digest those. And I believe this is a common problem for commercial chatbots today, no matter how well-designed they might be.
In a way, to solve this, we need to look beyond the solution itself–beyond design. We need help and documentation to play a role, and that might go against our instincts as designers. Ideally, if a product is intuitive enough, its users wouldn’t need any extra hand holding. Of course, we don’t want to return to the days of long user manuals. But even the creator of persuasive system design, Harri Oinas-Kukkonen, thinks that user documentation is still needed. That means, even if our design is sophisticated enough to speak to users on a psychological level, it’s still best to provide another layer of guidance to ensure that they can successfully do what they need to do.
Fortunately, guidance can be integrated well with the design, such as through a quick interactive tutorial that appears when the user sees an interface for the first time. And in the example of chatbots, it could take the form of a friendly note at the start of the conversation letting users know to be concise and to ask one question at a time.
This way, we’re not relying on the design to be perfect and to be intuitive to every user right off the bat. And, once again, it means we stop thinking of design as a silver bullet that immediately solves all problems.
At this point, you might ask. If design is a journey, how can we be sure that we’ll arrive at the destination, that we’ll come up with something good? And how can clients tell we’re the right partner if we don’t present our solution to them when pitching? As with individuals in the job market, the answer is past experience. Having experience means you’ve had practice and a track record, which gives you both the confidence and the proof that you can deliver. Showcase that experience.
Of course, the reality is the idea of the hero designer and BDUF are not going to suddenly vanish. They will continue to influence people’s expectations out there. But the change process is also a journey. It won’t happen overnight, and until we arrive at the destination, we’ll just have to continue to push on and do our best.