Huawei (Eames Project) Part 1
In December 2017, I joined Huawei's R&D group in San Francisco as a senior UX designer. Our primary focus was on developing a US version of Huawei's mobile software, EMUI, which is a customized version of Android that is pre-installed on all Huawei phones. As part of the Eames project, which included three phases (discovery, design/prototyping, and testing), our team worked on launcher explorations for EMUI. In my role as a senior designer, I contributed a launcher concept that I prototyped and tested as part of the Eames project.
Eames Project Objectives
Developed and tested a launcher concept customized for the US market through the design, prototyping, and testing process.
Identified potential opportunities for new features or unique design patterns that could enhance the overall EMUI user experience.
Illustrated key design principles that informed the EMUI user experience, as requested by the HQ team.
Discovery Phase
Heuristic evaluation key findings
To begin, I conducted a heuristic evaluation of the existing EMUI software, and drew some key insights from my analysis. Furthermore, I audited a beta release of Android P to evaluate its gesture-based navigation system. As gesture-based navigation was not yet a standard convention at the time, I replicated this behavior in the launcher concept I created in order to test it thoroughly.
Mix of iOS, Android elements & pattern
The EMUI experience blended design patterns from both iOS and Android, but with iOS visual elements. This mixture created a number of interaction inconsistencies. Additionally, due to the implementation of iOS design patterns, performance was an issue, making the UI feel inconsistent. As a result, there was a lack of expression of the Huawei brand within the user interface.
Convoluted gesture patterns
The screen captures above illustrate some examples of custom gestures that were included in the "Smart Assistance" feature. However, these unorthodox gestures were not intuitive within their intended context, and required a high degree of recall on the part of the user.
Convoluted interactions & inconsistent native patterns
During my evaluation, I observed some unusual interaction patterns, such as long-pressing to launch the digital assistant, and inconsistent behavior of the action bar. I also noticed that certain design elements, such as arrows in list views, were borrowed from iOS, which led to inconsistencies in the user experience. This mixing of iOS patterns in an Android-based system contributed to various usability issues.
Gesture based navigation issues
During the audit, I evaluated the beta release of Android P which featured a new gesture-based navigation system. However, the "pill" button was problematic due to its multifunctionality, making it difficult to determine whether it required a tap or a swipe gesture. Additionally, the long-press gesture to access the digital assistant often misfired, either when swiping or tapping the pill for too long.
Initial Interviews
Discovery challenges
As Huawei had no North American presence, it was not feasible to leverage their data. Additionally, features that were popular in the Chinese market did not make sense in the North American market. Therefore, in the discovery phase, qualitative data had to be used to inform the rationale behind the launcher designs and to make arguments against existing conventions within EMUI. As it was challenging to access Huawei's customers, the subjects interviewed were a mix of Huawei, Samsung, Pixel, and HTC customers.
The initial methods of discovery where broken down by initial phone interviews, followed by an in-person experience audit.
Phone interview objective
The goal of the phone interviews was to gain a comprehensive understanding of users' attitudes and comprehension of their current launcher experience, covering all features such as the lock screen, notifications shade, settings shade, app tray, minus one screen, and primary navigation. A total of five phone interviews were conducted with participants aged between 32 and 55, all of whom were Android users with varying phone brands such as Samsung, HTC, Pixel, and Huawei.
Phone interview insights
The lock screen was perceived by most interviewees as a security measure to protect sensitive information. Only one participant understood its original intent to prevent unintentional opening of the phone.
When asked about notification organization and prioritization, most interviewees could not recall. Two participants found the notification shade too cluttered.
One interviewee suggested that the Google Assistant would be an ideal way to engage quick settings when asked how the settings shade could be improved.
All interviewees found the order of the settings shade acceptable but were not aware that it could be personalized.
Interviewees found the minus one screen of little value due to the lack of interesting and relevant content, and they struggled to explain its concept.
Most interviewees did not utilize the digital assistant and did not understand its functions. They preferred to conduct manual searches instead.
Phone interview testimonials
“I leave it open {Lock screen}. No pin, no finger and no face id. I don’t store information on my phone that could be stolen. One last thing I don’t need to use”
– Michael, Construction Worker“I preselect that the only notification that I want to get are text, emails. Social media is more for leisure. Work takes more of precedent. If I get a lot of notifications, it’s like being bombarded constantly and not only is a distraction it takes up room.”
– Aasim, Home Finance“It’s interesting. I used it for a little bit but I defaulted back to Yahoo News. I’m impatient. It has to be something that catches my eye.”
– Aasim, Home Finance“No. Just by accident. The reason why I don’t use it [digital assistant] is primarily because I’m ignorant to it and I don’t have the time to learn it. I can’t say I don’t like it because I don’t use it.”
– Ernesto, Corrections Officer“If the assistant had more character, I’d engage it. Articulate what your trying to ask. Suggestion would be helpful”
– Maggie, Office Manager
Samsung has this feature called the “always on display”. I don’t really care for it and I haven’t got around to taking it off. It’s like a double edge sword. I want the information there but I don’t want the information when I don’t need it. I didn’t jump into a smartphone for the bells and whistles.”
– Ernesto, Corrections Officer“mmm. You know and I haven’t tried this yet on every feature but there is the “Ok Google” feature. It’s like the “Hey Siri” feature. I’ve tried it with calling someone and I tried it when asking for directions while I’m driving. That would be a good way to access a quick setting. I haven’t tried it for all the settings.”
– Aasim, Home Finance“I don’t like it and I never use it [Minus one screen]. I find the feature useless. I never found a use for it. It just seemed like product placement. It seemed more like bloat. If it had some relevant information, then maybe I would use it. But then I could just place that on my home screen. The feature has to prove to me that it’s not bothersome. If it’s interesting, then it’s not bothersome.”
– Ernesto, Corrections Officer
Experience audit (in-person interviews)
To conduct a comprehensive audit, I primarily focused on the Android system since most interviewees had not used Huawei's EMUI software. Over several weeks, I conducted in-person user tests to observe and record user behavior.
During the audit, I concentrated on the essential features of the Android system, including the Lock screen, launcher, notifications, settings, app tray, and minus 1 screen. I instructed each test subject to complete specific tasks across all these features while narrating their thought process.
The experience audit consisted of:
6 Android users
Participants where a mix of Huawei, Samsung, HTC and Pixel users.
Ages of subjects ranged between 24 and 55.
Key findings from experience audit:
Notifications caused increased habituation due to constant triggering.
The majority of users found no utility for the minus-one screen.
I observed an overall inconsistency of behavior as users completed certain tasks per feature. Some test subjects were unaware of certain features and/or design patterns, possibly due to inconsistency set by OEMs.
The heuristic eval., telephone and in person interviews produces good insights, uncovered existing pain point with native Android. From the feedback captured certain themes came to light that informed the launcher designs.
Continue to part 2…