Alexa
Attention System
Project Brief
Headless
Multimobile
Mobile
My Role
Lead UX Designer
Senior Motion Designer
Design POC for
Industrial Design
Product team
UX & Product Research
Privacy & Legal
Alexa’s Attention system is comprised of non-verbal audio and visual components that work together to communicate all of Alexa’s different states. Every Echo family device leverages the attention system to deliver Amazon-branded personality that is familiar to customers across all products.
Problem:
The original design of Alexa Voice Chrome (AVC) was not designed or produced with scale in mind. This led to many growing pains as Amazon created new hardware, features, modalities and end points where the AVC had to be adapted.
​
Solution:
The AVC needed to become a system that scaled across current and future hardware in headless devices (devices without a screen), different screen layouts for multimodal devices and Fire TV,. It also needed to be adaptive to emerging voice modalities as the capabilities of Alexa expanded.
CHALLENGES
1
As new features were created for Alexa, the LED system did not have an established visual or motion language. In addition, the creation of new LED animations was tedious and arcane.
2
As Alexa's device ecosystem expanded into devices with a screen, the headless system did not scale nor fit this new category of devices.
NECESSITY IS THE MOTHER OF INVENTION
When Alexa gained communication and messaging capabilities on headless devices, the Attention System (AttSys) collection of indicators did not have patterns that could scale. It also did not have a language that users could recognize.
This presented a large challenge when I was tasked with the delivery of new indicators on a very tight deadline. In addition, I was surprised to find out that a robust pipeline did not exist for production of the LED indicators creating a doomed to fail situation.
I created a new production pipeline that allowed me to create deliverables at 10 times the speed and meet my immediate deadline.
This production pipeline was crucial moving forward to produce a large quantity of deliverables that I anticipated would be requested in the near future.
Post deliverable, I worked with the design technologist team to create a refined version of the tool that I invented. Going forward, the pipeline I invented allowed me to create alternate versions of existing indicators for different arrays of LEDs more efficiently.
LEADING AND GUIDING
New Echo devices required all new animations as they had different LED arrays that varied in size, shape and number of LEDs. Although the production pipeline I created allowed me to produce assets at unprecedented speed, the UX of the indicators was not directly translatable to this new arrays. I partnered with the industrial design team to have better visibility of upcoming products. I was then able to create guidelines and requirements for LED arrays for all upcoming devices. This created a lasting relationship with the industrial design team.
User feedback regarding the LED indicators was mixed during BETA regarding color reproduction. After engaging with the research team we discovered that there were inconsistencies in color reproduction with previous devices. Diving deeper, I engaged with the optics and design technology team and we discovered that binning of the LED was not consistent at the manufacturer. After calibrating assets for the device as a stopgap measure, I engaged with the research and design technology team to find out how users perceive the separation of color. With this information I was able to approach the product team with guidelines for color reproduction targets and tolerances accepted for future devices to maintain the user experience consistent throughout the Echo family of products.
I created a self service page with references and a repository of all of the potential LED array configurations. This allowed product teams to understand what the system does, the expectations of performance and explicit guidance for next steps, how to use the device, and how and when to engage with design for QA and final asset delivery.
Foreseeing an ever increasing gamut of LED arrays in upcoming devices, I did a survey of the existing indicators. Without a budget for new research, I engaged with the research team to find as much anecdotal and tangential feedback from previous research studies regarding the LED attention system. This highlighted some of the strengths and weakness of the design that was originally handed over to me. I presented my findings along with a proposed outline and scope of work to redesign the Attention system. Unfortunately the proposal fell below the line during annual operational budget meetings.
ALEXA ON SCREENS
The launch of the Echo Show was a big change for the LED indicators. There were obvious challenges like the absence of an LED array and maintaining the Alexa branding. There were also less obvious ones like the UX patterns from headless not translating to a screened device because of very different requirements and customer needs. I also needed to reconciliate conflicts between existing paradigms on mobile devices and personal computing, and headless devices.
Working through the iterations of the design, scalability, branding, and a consistent customer experience were paramount. The final Attention System design that customers have enjoyed since 2017 has remained consistent with minor modifications. The design's flexibility has allowed it to endure through many different device sizes including many Echo Shows, FireTV, and Mobile devices and has remained consistent with devices without a screen.
OUTCOMES & LEARNINGS
01
The Attention system of Alexa devices remains one of the most recognizable, consistent and timeless designs in the Echo family of devices and services. For the last 7 years, it has been a reliable way for customers to interact with the devices and has done a good job at safeguarding customers' privacy.
03
Hard data is not only the best way to make design decisions, it is also crucial to drive agreement with stakeholders. The program was never a priority and resources were always hard to get in order to address "uncertain pain points".
02
Ownership of the Attention System awarded me with one of the highest achievements of my career, the patent for how Alexa looks, moves and functions. Patent WO2018053465A1 was recognized by the VP of Devices as one of the most significant patents the company has been awarded.
04
Although the bottom up effort was fruitful to maintain the system, it was insufficient to prioritize resources. In late 2022, I was able to engage directly with design leadership to drive prioritization from the top attached to a new device launch that required significant changes.