Stealing design is an exercise in futility unless you do it right

When I began designing websites at the age of 17, I had lofty aspirations. I was going to one-up the status quo. But, as many do, I quickly discovered how difficult it can be to build a truly inspiring user experience. I found myself instead relying on the oft-repeated adage, which has become something of a cornerstone of user experience design for novices (and even some experts, like Picasso): “Imitation is the sincerest form of flattery.”

Seems like a simple concept, right? Don’t reinvent the wheel. Don’t make an unfamiliar user experience just because it has the potential of being cooler or sleeker. Build something that is already acceptable and to which users are already accustomed.

While this basic strategy safeguards against over-zealous design and can help overcome designer’s block (it’s like writer’s block, but prettier), I don’t think it works all that well in practice. I was continuously frustrated with the performances of my sites’ user experiences during user testing, despite merely imitating popular user experiences from widely-visited sites.

For example, when we were designing the website for my last startup, Hublished, we sat in front of a screen with windows open for Facebook and LinkedIn. One of the core use cases of the platform we were building was that a user could have a single account to manage both their own consumer experience as well as a business page that they managed. We figured that people are already used to the Facebook experience where you can enter ‘page manager mode’ so we’ll build out the user experience in the same way.

So while in consumer mode, our users would see one set of icons:

And while in publisher mode (which would be made clear to them with a message similar to the one displayed on Facebook), they’d have a different set of icons:

To our utter disbelief, no one without access to our design meetings could figure out how to use the platform. Unless all the users we tested the platform with didn’t use Facebook, which I believe may actually have been statistically impossible, we needed a redesign. So we looked to LinkedIn:

LinkedIn doesn’t take the user out of their own consumer mode, but rather adds a toolbar of sorts when the user is simultaneously managing a business page. So we did the same thing:

There was certainly an improvement in performance of our key performance indicators (KPIs) like the number of seconds (or, more often, minutes) to switch from a consumer profile page to a business page. However, users still expressed significant confusion navigating the platform.

How could this be? We were simply imitating already popular sites. Why did users find this so confusing? It wasn’t until months later that it hit me while reading Designing with the Mind in Mind: Simple Guide to Understanding User Interface Design Rules.

Jeff Johnson’s core premise throughout the book is the concept of ‘learning curves.’ While designing virtually anything, we must be aware that there are biologically innate perception mechanisms we use to evaluate interfaces. At the core of these is a theory known as The Gestalt Principles. They are the most basic of user experience laws that describe how people perceive things like similarity and proximity. For users to become comfortable with any design that relies on an intuition above and beyond these principles, there is an inevitable learning curve (you didn’t think using a mouse and keyboard was natural, right?) that can only be achieved via gradual releases.

Interesting, but still no light bulbs went off in my head. Until Johnson began illustrating examples, and that’s when I stopped reading altogether and just stared blankly at the page. Published in 2010, Johnson’s examples were mostly screenshots of websites from the mid to early 2000s. Holy GUI, my eyes burned! The websites depicted were hideous.

I instantly recalled a fascinating Mashable slideshow with screenshots of the most popular websites from back when they launched. The slideshow made me realize the countless iterations major websites have gone through to get to where they are now.

While web design standards certainly change and widespread adoption (unless you’re designing, or intentionally not designing, Reddit) leads to new design elements becoming innate to modern users, replicating the user experiences unique to Facebook and LinkedIn does not guarantee the same results.

How many native Facebook users do you know that create an account on Twitter and are immediately befuddled? The learning curve is tremendous despite both sites using many of the same design elements.

When we seek to imitate, we must remember that the proverb only guarantees achieving a level of flattery – not a level of user understanding. Consider first that other websites have gone through hundreds and thousands of iterations that slowly nudge users over a learning curve. No site could simply launch today with the numerous features and nuances that Facebook has and expect anyone to become comfortable with them in any reasonable amount of time.

When we simplified the Hublished design and relied on more basic and universal design principles, we found that KPIs drastically improved. Through iterative design, we were able to slowly build toward a user experience that was more flexible and powerful, but we could not have started there.

Instead of imitating what your competitor’s site looks like today, study how it got there. Analyze the iterations and the learning curve users have overcome to appreciate its current experience. You can’t just start from the nth iteration – you, and your users, need to build up to it. After all, if imitation worked so easily, we’d all be Picasso.