Cyanogen
Themer.
Role.
Design Director
​
Company.
Cyanogen
​
Team Size.
10
​
Year.
2014
Unlike iOS, Android is, at its core, open source. Over the course of a decade, the Android Open Source Project (AOSP) has provided roots to hundreds of upstart operating systems with a wide range of ambitions. Possibly the most prolific of these forks was Cyanogenmod. Which promised to democratize stock android, and open up the platform for customization. With enough knowhow, more than a little patience, and a willingness to risk turning top of the line smartphones into paperweights, one could create a mobile experience completely personal to them.
​
A huge aspect of this customization was the visual aesthetic of the user interface. Cyanogen leveraged a powerful theme engine to give users the ability to customize, on top of background choices, taskbars, icons, buttons, typography, and even sounds.
That level of tweaking was complicated.
​
The Short
Version.
What did we build?
A deeply integrated, OS level theming engine that was powerful and straightforward, allowing for both wholesale and mix-and-match device theming, complete with a marketplace and creation tools.
​
How did we validate?
We used the Lean methodology to define our early assumptions, and small batch testing to validate them. ​
​
My Experience
Two things. One, it was incredibly rewarding to make such a complex process straightforward and enjoyable to use. Two, It was my first time managing not just designers, but an entire cross functional team, and thanks to that experience, having creative and engineering collaborating real time will always be my first choice.
​
Deliverables.
One shipped android theme store.
One OS level theming experience
UX Concepts for theme creation tools.
​
 |  |
|---|---|
 |  |
 |  |
The Long
Version.
The Problem.
To create, distribute and consume themes required a technical acumen that only a small subset of our users had patience for, and because we wanted to add more non-enthusiast users to the platform, we knew this hurdle would be a non starter for many potential users.
The Fix.
Redesigning theming was ambitious, but we had two major advantages you don't normally find.
​
First, being part open source project, Cyanogen had an enthusiastic and organized online community as a fanbase, and that had it's benefits when it came to user testing.
​
Second, because of AOSP, we controlled the OS down to the framework. we could make very ambitious design choices.
Creator
Create.
Sell / Purchase.
Apply.
Consumer
Our Approach?
Get Lean.
Given our advantages, and the complexity of the challenge, we decided Lean prototyping was the best way to ensure we built a user forward solution. We broke all of our early theories into assumptions that we could test.
We assumed we had two users who used themes in different ways.
Some of our user base were creators. they made themes for themselves, and to share with the community. Others wanted to sell their work. Whether they created themes themselves or not, however, everyone was ultimately a consumer of the final product.
​
We figured both consumers and creators were highly technical, given the difficult UX of the current theme engine. We also expected them to be very curative because of that effort spent, wanting more theme variety, as well as the ability to mix and match theme elements.
​
Finally, we assumed that to grow a themes marketplace, we would need to lower the bar to entry, and make theming simple and engaging for less particular users.
Technical
Current
Creators
Current
Consumers
Indifferent
involved
Desired
Creators
Desired Consumers
Ludite
We assumed current user technical acumen was high, but we needed to attract a broader range.
Solution
As easy as a wallpaper.
Applying a theme should feel completely integrated with an OS. Instead of an app to launch, we developed a method for the user to access themes on a long press from the home screen, the same way they would select a wall paper, piggybacking off their built in muscle memory for changing their devices appearance.
Once the user has accessed the theme functionality, they are greeted with a visual representation of their theme options, giving them more capacity to see their themes than a traditional list format.
Solution
Power to the Picky.
Tapping on an icon associated with each theme would split the theme into individual compartments for system UI, Icons, wallpapers, typography, lock screen and sound. Each compartment could be swapped out for any other like compartment from another theme.
This gave power themers the capacity to mix and match multiple themes until they had something that was completely unique to them.
Solution
A familiar marketplace.
Theming is a unique feature for an operating system, it requires a unique approach in it's design to succeed. Some elements of the system, however, have tried and true patterns that should be adhered to. Because of this, discovery and consumption of themes needed to take place in a store like experience, separate from the theme application.
Solution
Straightforward creation tools.
A complete theme consists of icon packs, system UI changes, wallpapers, sound packs, Navigation and Notification drawer changes. The only way to get this into a package the theme engine could use was by compiling them into an APK. This had a high technical barrier. We figured if we could create visual theme compiler, it would save our current users time and attract new ones that were previously put off by the large hurdle of making a theme.
Cyanogen Themes:
Assumption success rates.
The envelope, Please.
Throughout the development of these solutions, we continuously referenced back to our original assumptions, and at critical intervals, tested them with real users.
​
The results validated many of our theories, but more importantly, challenged some as well, helping us priorities the features we knew would benefit our users the most.