Thoughts on Design
— Create the right context for your product
Design is not graphic art. This has been well documented. I think product design is more closely aligned to engineering and psychology than the visual arts, because building a product entails so much more than having it look nice. It sounds obvious, but few product builders truly deliver on making something that just works.
Paul Graham of Y Combinator noted in 2012 that, without Steve Jobs, Apple would have “no more great new stuff beyond whatever’s currently in the pipeline”. Indeed, we see that playing out. Arguably Apple hasn’t released anything ground breaking since then.
Steve Jobs was first and foremost a designer, although he never claimed so.
Design is a really loaded word … we actually just talk about how things work. Most people think it’s how they look, but it’s not how they look, it’s how they work.
I feel this sentiment. Design is more about making sure the product works right, in that the product simply gets you. In essence, a well designed product empowers you to do more with less [1].
Understanding the potential impact for what you’re building is the starting point. Be very deliberate about: Who will use this? What are their values and goals? How can this product make their lives better? They will help you uncover what’s truly important.
Brian Chesky of Airbnb uses a Disney style storyboard (they hired an animator from Pixar), to bring to life a person’s ideal vacation experience. Every touch point of that ideal experience is sketched out in detail. This process of visualizing is a journey of discovery. It will help you uncover the right manifestation of your mission in your product.
People prefer Airbnb as opposed to Craigslist when it comes to renting vacation homes, because compared to Airbnb, Craigslist is just such a bad platform. On Craigslist there is no way of finding out what kind of person the host is or what reputation she/he has; there is often a lack of (decent) photos and (clear) pricing about the place. These are just some of the reasons as to why the world needed a product like Airbnb.
However, this is not to say that Craigslist is a bad product, for the purpose of buying and selling used things. Sure, it doesn’t have a great visual design, but we must be careful about not confusing visuals with the product. Greats visuals can be part of a great product, but it’s not a prerequisite. Craigslist’s lack of visual styling might be an example of “bad design” turning out to the be the best design, as seen in Snapchat. Perhaps, similar to how Snapchat’s whimsical colours and lack of polish reduce users’ mental friction when sharing private moments, a visually stunning version of Craigslist might take people out of the mentality of looking for a good deal [2].
The argument here is not that visual design is unimportant. The opposite is true. Visual design is very important, not because products need to be inherently beautiful, but rather because how we choose to visually present our product has a huge implication on people’s perception and usage of it. As product builders, we must be keenly aware of the intricacies and nuances of this cause and effect relationship.
In the book Thinking, Fast and Slow, the author Daniel Kahneman highlights many examples of how presentation affects human behaviour. One example in particular stood out to me.
Princeton students were recruited to undertake in a study that measured their cognitive ability. They were asked a series of questions, like this one below:
In a lake, there is a patch of lily pads. Every day, the patch doubles in size. If it takes 48 days for the patch to cover the entire lake, how long would it take for the patch to cover half of the lake?
Many students responded intuitively, thinking that half the pond would be covered in half the time, so 48/2 = 24 days. This is of course wrong. The correct answer is 47 days. If everyday the lily pads doubled in size, on the day it covers the entire pond, we can be sure that on the day before, it only covered half of the pond.
Half of the students saw this in a clear legible typeface. The other half saw it in a small font in washed-out gray print. 90% of those who saw the normal font, made at least one mistake in the evaluation. When font size was barely legible, the proportion of those who made mistakes, dropped to an astounding 35%.
This is explained in detail in the book, but the gist of it, is that seeing the bad font induces cognitive strain, which tends to make us think in a slower, more deliberative, and more logical way.
In this case, if our goal was to help the students answer these questions correctly, then we must consider what is influencing their mode of thinking [3]. This is not to necessarily suggest that we should make the questions hard to read, but rather that we create an environment or context in which the students’ deliberative and logical mode of thinking can be invoked.
This example gets to the heart of what design is really about, in my opinion.
To design is to do three things:
- Question why things are the way they are.
- Understand people and the implications of your decisions on their perception and behaviour.
- Push the boundaries of constraints and deliberately create the world you want see.
FOOTNOTES:
[1]: This can be less UI, less clicking, and less time spent trying to figure out how to use something, so that you’ll have more time for what you love and more time to focus on what’s important.
[2]: There was a related research conducted by Microsoft on why Nigerian scammers (or at least those who claim to be Nigerian) send emails with bad spelling. The reason is simple: they use it as a screening mechanism. Chances are if you’re smart enough to recognize the bad spelling, you’re not their target audience. These scammers are designers in that they’ve thoughtfully considered the approach that maximizes the impact of their product. Of course, designing for evil is a topic for another day.
[3]: There is a popular example of this. In a not-so-scientific experiment, it has been shown that college essays written in Georgia typeface scored on average one and half letter grades higher than ones in written in Comic Sans. Interesting, but not surprising.
