When software first came into the world, technological limitations made for a very low “experience” bar. Software either could do what you needed it to do, or it couldn’t. If it required a fifty page manual to explain how to calculate the product of two numbers, then that’s just how it was.
With the dawn of the Macintosh, “usability” slowly came into vogue and the experience bar was raised. Nowadays it’s expected that software actually be easy to use. Gone are the fifty page manuals, in are self-describing interfaces with strategically designed virtual affordances.
We are on the verge of an era where the software experience bar will be raised even higher. No longer should we only be concerned with how easy it is to do something, but with how a user should feel while they are doing it.
I recently worked at a company called Koko FitClub. Koko is a fitness center franchise, but one unlike any other. It was designed for people who hate the traditional gym experience. Part of that “traditional gym experience” is the aggressive personal trainer who pushes you out of your comfort zone, harasses you to show up on a fixed schedule, and berates you for imperfect performance. Koko’s philosophy is adamantly opposed to all this. Its attitude is that every step you take toward fitness – even if it’s only one workout a week – is a positive step toward better health, and should be celebrated.
This philosophy went beyond the in-club experience and marketing materials. It went right down into the functionality of the member web site. We had enormous volumes of member data to work with but what we didn’t present to the user was just as important as what we did. The site never commented on how long it had been since your last visit. It never challenged you to visit more often or lift more weight than your fellow members. And it never graphically displayed how much further you had to go on your predefined program before moving onto the next one. We had the technological capability to do all of those things, but we very intentionally chose not to.
Every visual element and interaction on the site was designed to give you a feeling of “I’m doing OK, and if I do more, it’s reason to celebrate”. This type of emotion-based design goes beyond usability. Usability is binary – something is either easy or it isn’t. Emotion is not – it’s very brand-specific. The ideal emotional design for Koko would not be the correct one for a CrossFit “box” or a Gold’s Gym.
When designing a website or an app, remember to consider how a feature will make someone feel before deciding if it belongs in your design. Your brand isn’t just reflected in your logo or visual design. It goes all the way down to the widgets on a screen.
When Apple first announced the iPad, I was skeptical. It seemed like an interesting enough proposition – a specialized, simple computer great for kids, seniors, and “people who walk around a lot and occasionally need to look at complicated stuff” (e.g. doctors). A great adjunct device, but certainly not something that would replace “real computers”.
Then, far more quickly than I ever expected, the iPad started becoming way more popular than “real computers”. “OK”, I thought, “We need to remember that there are a lot more ‘regular’ people than people like me —who eat, sleep, and breathe computing.”
Then something happened that made me feel… threatened? Apple had their “Back to the Mac” event, in which a lot of iPad-like functionality was migrated “back” to desktop and laptop computers. Then OS X Mountain Lion came along and took this trend even further. This bugged me, for reasons I couldn’t quite elucidate.
If iPads were for regular, every day people, and “traditional” computers were for “power users”, why is everything converging in only one direction? If iOS was the computing platform for “regular people” then why wasn’t OS X being optimized somehow for power users? It got me thinking about what, exactly, that meant. What differentiated a “regular person” from a “power user” and how should operating systems differentiate between the two?
As usually happens, the answer struck me while I was in the shower one morning.
It’s all about contexts.
In 2009, Google released a video entitled “What is a Browser?”. In it, 50 random people on the street were asked, well, “What is a browser?”
It’s easy to laugh at their responses — but for anyone paying attention, this was a huge wake-up call. The average person had a very hard time grasping what “computer people” consider one of the most basic, fundamental principles of modern computing — that computers run programs called browsers, to access services called search engines, which return results, which lead to web pages, which (hopefully) give you answers you need.
Hey, wait a minute. That doesn’t actually sound that simple.
Most of the “mom and dad tech support” that people like me suffer through center around problems with context. We tell people to “click the ‘OK'” button — but is that button part of a dialog, a document, an app window, or the operating system? This is madness, when you take off your Computer Wizard hat and really think about it.
Herein lies the “magic” of the iPad. With the iPad, you are either “doing a thing” (running an app), or “picking a thing to do” (using the home screen). Not a lot of room for confusion. But not a lot of flexibility, either. Power users want that flexibility. Regular users don’t.
Well, they do a little. But for “regular users”, cutting and pasting something between apps is an infrequent, exceptional situation.
Power users do that stuff all the time. It’s our bread and butter. In a given day, I’m reading requirements in Word docs, feeding information from them into a project management tool, exporting data into Excel, writing code in a text editor, and shuffling files around using various transfer tools.
That got me thinking of the following simple differentiator between “regular users” and “power users”.
Regular Users are people whose primary computing tasks involve working within one context at a time. Power Users are people whose primary computing tasks involve transferring information between contexts.
This is the key, irreconcilable difference between two types of computing that should be addressed by having two different styles of operating system. Both Apple and Microsoft are tripping over this dichotomy — Apple by having two ways of doing almost everything (Launchpad vs. Finder, full screen vs. windowed, iCloud vs. file system) in Mountain Lion, and Microsoft by almost comically slapping two radically different, and awkwardly isolated, modes together into one system, in Windows 8.
Some people think that tablets are for regular people and “real computers” are for pros. I don’t think that’s the case at all. Both form factors can work for both types of users – it’s the supported workflow that matters. Microsoft got this half right with Surface, and could actually nail it if they had the guts to release pure versions of “Metro” and “Desktop” operating systems for their RT and Pro models.
You can’t be a family sedan and a construction vehicle at the same time. Hopefully Apple and Microsoft (and Google) figure this out in the next round.
I just got back from An Event Apart Boston – an amazing experience every year. The presentation that wound up sticking in my head the most was Jared Spool’s – in which he talked about “intuitive experiences”. The intuitiveness of a system is not an absolute property of that system – it is variable based upon the experiences of the system’s user. And if you’re not careful, a redesign intended to make things better for everyone can be a disaster for your bottom line.
Complex systems can be intuitive to the right user. My mother certainly couldn’t develop a Ruby on Rails site – but talk to anyone who’s been doing it for a while, and they’ll say Rails is incredibly intuitive. It’s all a matter of experience.
The intuitiveness of a system is the inverse of the distance between the user’s “current knowledge” (what they already know how to do) and the “target knowledge” needed to use that system.
There are two ways to narrow that distance: You can increase the user’s current knowledge via training (never a good approach on the public web – no one ever reads instructions on how to use a website), or you can decrease the target knowledge required to use the system.
So, simplifying systems is a good thing, right? Well, maybe. Simplification changes the nature of the “target knowledge” needed to use a system, but doesn’t linearly move it closer to the user’s knowledge. Simplification invariably introduces change, and change invariably requires adaptation by people familiar with the system being simplified. A system can be redesigned to be more usable (and intuitive) to brand new users, but with a website, that is never the full picture.
Because here’s the kicker. A system is, by definition, intuitive to the people who are already using it effectively. This is why people get so up in arms when sites like Facebook and Twitter change their design. Outsiders may think the changes are good, or at least barely noticeable – but people who use those sites constantly have to change highly-ingrained behavior. The reason Facebook gets so much attention for “constantly changing for the worse” is because it has the most highly efficient users of any site in the world.
So, here’s where it gets really interesting. In Jared’s talk, he discussed a large eCommerce site with over $1,000,000,000 in annual revenue. Like all eCommerce sites, only a small fraction of people visiting the site actually make a purchase. In this company’s case, their “conversion rate” was 1.6% – out of every 1,000,000 visitors, 16,000 people made a purchase. But out of that 1.6 percent, 20% of those users (heavy, active users) accounted for 80% of all revenue.
So, if you’re this company, and you want to generate more revenue with your site, what would you do? Well, what most companies try to do is get more people into the “heavy” group (or at least the “actually made a purchase” group) – and they’ll do this by redesigning their system to be easier to use. But, again, any redesign makes a system less intuitive to its most active current users.
Let’s say frustration from these changes cause the heavy users to spend 13% less money on the site. That means a loss of over 10% overall revenue for the company – in this company’s case, over $100,000,000 per year! Now, if things went according to plan, this loss will be buffered a bit by new customers, or possibly “light” customers who buy a little bit more.
But you are taking a huge gamble that you’ll make up this difference. In one real world case Jared discusses, a major retailer made a $100,000,000 major site redesign and lost 20% of its site revenue – from which it took three and a half years to recover!
Ever wonder why online flight-booking systems never seem to get easier to use? Why they’re always so complex? Well, even moreso than most companies, airline profits are driven by “power users”. Any change will have a major ripple effect in their profits.
The harder your site is to use, the more invested your power users are in having it stay just the way it is.
It’s a weird, counterintuitive fact - the harder your site is to use, the more you risk by changing it!
What are the takeaways?
I’ve come to appreciate the important of a single person in the company owning the UX with this person being the arbiter of discussion around how to implement the UX. There’s nothing wrong with lots of different perspectives, but a single mind has to own it, synthesize it, and dictate the philosophy. But first, they have to understand the difference between UI and UX, and – more importantly – the product-oriented execs who approach things from an engineering perspective need to understand this.
Product Management (or Product Ownership, in Agile terms) has historically been the domain of Marketing. These days, more than ever, the Product often is the marketing and the stewardship for product development needs to be in the hands of someone completely obsessed with the experience of the user. Anyone who disagrees apparently missed about 15 years of Apple becoming the biggest company in the world.
I’ve been reading Al & Laura Ries’ “The Fall of Advertising and the Rise of PR“, and it got me thinking: Is advertising really dead? If so, why are so many people doing it? I’m utterly convinced that you cannot build a brand through advertising – at least not profitably. So, what’s the right way to view advertising?Well, first off, there are products that can be introduced and sold to someone in 30 seconds – cheap consumer products and food. You don’t need to be convinced to subscribe to a certain “world-view” to think “Hey, that cheeseburger looks tasty”. So, let’s skip that stuff and talk about things like cars and designer clothes.Branding is about building a imaginary world. A world where people behave a certain way, look a certain way, do certain things. Most of it is completely bogus. Drinking Budweiser will not make bikini girls flock to your backyard. Ralph Lauren Polo shirts do not come with free country club memberships, drinking Starbucks’ coffee does not make you part of the cosmopolitan elite. Convincing people (on a subconscious level) that these worlds are real cannot be done in a 30 seconds.What can be done in 30 seconds is convincing someone that now is the time to become part one of these worlds. One way is to offer a financial incentive (0% financing on that new Lexus). But another way is to simply tweak your message to appeal to someone you don’t ordinarily reach. Someone who knows all about your brand, is intrigued by it, but doesn’t quite feel like they belong. Starbucks can take 30 seconds to show average Joes that their brand isn’t just for rich snobs. Wal-Mart can take 30 seconds to, well, do the opposite.
The interaction design community is revisiting a decades-old debate regarding digital skeuomorphism. In simple terms, a skeuomorphic design is one in which features of an earlier version of an object are retained, although they are no longer functionally necessary. The skeuomorphic approach to interaction design was responsible for some of the worst UI abominations of the 90’s, and now it’s back. Are we failing to learn from our mistakes? I don’t believe so. (more…)
I’ve been thinking a lot recently about the trend toward minimalism in design – particularly in logo design. I’m wondering – is this trend simply correct? Is minimalism inherently better? Or is it just part of a cyclical phase?
A minimalist logo says “we are capable of doing anything, and don’t want to limit ourselves”. But is that what people want from brands? Could there be an impending movement toward rewarding brands who explicitly stand for something specific? And if so, would a move back toward more intricate logos reflect that value?
Constraints are a beautiful thing. They’re invaluable in the act of creation, but they also play a role in consumption. Without constraint, there is no sense of accomplishment. Reading a good book or magazine article gives you a feeling of accomplishment – that you walked away somehow wiser. Surfing the web does not. Sure it can be enjoyable, but who ever said “Wow, I feel so great after surfing the web for 3 hours!”
When trying to figure out how to structure your web-based service, it would be wise to consider what successful similar websites have in common. If you run an eCommerce site, you’re obviously going to analyze other eCommerce sites – a search engine, other search engines. But what if your site doesn’t fit neatly into any of the typical categories? What if it’s some new kind of service that’s never been offered before (think Twitter)? How do you know if you’re building something that will create a loyal customer base that will keep coming back without expensive prodding (i.e. advertising)?
I’ve neglected this blog for far too long. Today, I’m coming back with a hopefully potent post that has been sitting in my drafts folder for almost 2 years. Not sure what I’ve been waiting for, but it seems to holding up, so here goes…