Embedded Files
(Multiple launches 2023–2025)
*I assure you, I did not pay the Verge to say this. 😂
But the tech press's organic recognition of the Settings tab is a small source of pride.
About settings for the Google Home app: So...what's tricky about settings for the connected home?
About settings for the Google Home app: So...what's tricky about settings for the connected home?
I'd say imagine the settings you'd typically need for a user account, add the settings for a smart device, the settings for a home (a collection of those devices), then throw on an app for managing and staying connected to your device. Examples of these settings types are:
I'd say imagine the settings you'd typically need for a user account, add the settings for a smart device, the settings for a home (a collection of those devices), then throw on an app for managing and staying connected to your device. Examples of these settings types are:
Device: What language does my thermostat use?
Account: What voice does the Assistant use when it recognizes me?
Home: What's the home address? What household automations are in this home (that members can add/edit/delete)?
App: What Wi-Fi credentials are saved with this app instance (used to make device setup faster)?
Now consider the combinations of settings you also need:
Now consider the combinations of settings you also need:
Account + device: Do my phone calls ring on this speaker?
Account + home: Who are members of the home?
Account + Account: What music accounts do I have?
Home + device: Is this speaker used for presence sensing (detecting occupancy)?
Home + account + device: Is my phone used for presence sensing for this home?
Now multiply all that by dozens of devices, in 1 to 2 homes, each home shared with up to 6 people who have their own settings...using multiple phones, tablets, and wearables that are sometimes personal, sometime communal. Sometimes setting A in this screen conflicts with setting B in another screen...
The Google Home settings framework aims to make it possible for users to use ever-more powerful features and devices, whether they're new to tech, or an early adopter.
The Google Home settings framework aims to make it possible for users to use ever-more powerful features and devices, whether they're new to tech, or an early adopter.
The problem
The problem
Settings had FOUR entrypoints across the app
Settings had FOUR entrypoints across the app
Additional challenges
Additional challenges
Simplifying the complexities of:
Device settings (What language your thermostat uses)
Home settings (What automations are in this home?)
Account settings (What voice the Assistant uses when I'm recognized)
Account + device settings (Do my calls ring on the guest room speaker?)
Account + home settings (Who are members of the home?)
Account + account settings (What music accounts have I linked?)
Device + home settings (Is this device used for presence sensing?)
Account + device + home settings (Is my phone used for presence sensing?)
...now multiply all this by 43 devices, in a home with up to 6 people, and up to 5 homes!
Conflicting needs: users want more simplicity, while stakeholders or revenue needs call for more entrypoints
We had some periods of no product manager
My contributions
My contributions
UX strategy
UX/UI design & prototyping
UX co-writer
The team
The team
1 Lead UX designer (myself)
1 UX writer
1 UX researcher
1 product manager
5 engineers
Partners
Partners
7 device verticals
3 cross-Google product areas
Impact
Impact
(2022) A consolidated settings tab for the redesign of the Google Home app, uniting four settings entrypoints
(Ongoing) Google Material 3 updates
The settings changes did not get promotion, but the tech press took notice organically:
9to5Google: "Perhaps the biggest improvement that comes with these pages is a revamped Settings page."
9to5Google: "Perhaps the biggest improvement that comes with these pages is a revamped Settings page."
Launch dates
Launch dates
2023 The new tab launched to the general public
2022 The new tab launched in Public Preview
The timeline/process
The timeline/process
Initial redesign proposals
Initial redesign proposals
In Fall 2019, we made an initial proposal on logically separating device settings vs. home settings vs. account/app settings.
This was eventually revised to re-think the complexity of that approach
App-wide design sprints + a "Home profile" concept
App-wide design sprints + a "Home profile" concept
In 2020, an app-wide redesign took priority and a "profile" concept emerged, making the last tab of the app a destination with metrics and insights about the home.
I suspected this was information better saved for the everyday-use tabs of the app.
UXR eventually bore this out
Refinements on a "Settings" tab
Refinements on a "Settings" tab
With the team aligned on a settings and not an insights tab, we explored many visual approaches
The final design balanced the goals of having some warmth while also getting users to the most-tapped items in home settings
Implementation
Implementation
With the Google Home Systems Team, we defined new component variants
With engineering, we implemented the new variants for the settings tab
Navigating challenges:
A "home profile" vs. a setting tab
Navigating challenges:
A "home profile" vs. a setting tab
As part of the 2020 app-wide redesign sprints, a "home profile" hypothesis arose:
Make the last tab of the app an everyday destination with metrics, insights, and tips about the home...and of course settings too.
As part of the 2020 app-wide redesign sprints, a "home profile" hypothesis arose:
Make the last tab of the app an everyday destination with metrics, insights, and tips about the home...and of course settings too.
My hypothesis was:
Our #1 need is to consolidate all those scattered settings into one place...
My hypothesis was:
Our #1 need is to consolidate all those scattered settings into one place...
...and to quickly get people to the most important settings items.
...and to quickly get people to the most important settings items.
Because the meme here sums it up:
Because the meme here sums it up:
Our foundational UXR found that the vast majority of participants rarely visit settings, but when they do, it’s to address something important.
But settings was part of a bigger design-led redesign of the entire app. And a hypothesis that came out of the brainstorming workshops was:
But settings was part of a bigger design-led redesign of the entire app. And a hypothesis that came out of the brainstorming workshops was:
What if this tab was a destination, a dashboard of insights and suggestions…how to make your home run better?
What if this tab was a destination, a dashboard of insights and suggestions…how to make your home run better?
To explore the "home profile" hypothesis, a designer on the home intelligence team and I explored multiple explorations.
To explore the "home profile" hypothesis, a designer on the home intelligence team and I explored multiple explorations.
The idea was that participants would find metrics and insights to be helpful snapshots of their home.
The idea was that participants would find metrics and insights to be helpful snapshots of their home.
My take on this after a while though was:
"I think insights are better suited for other parts of the app..."
My take on this after a while though was:
"I think insights are better suited for other parts of the app..."
...because this tab was still expected to house critical settings entry points. And while I liked the info and tips, that's not data people want mixed in with the settings items they need to get to when things are broken.
...because this tab was still expected to house critical settings entry points. And while I liked the info and tips, that's not data people want mixed in with the settings items they need to get to when things are broken.
And this showed up in the research, where users didn't grasp the value prop of a "home profile".
And this showed up in the research, where users didn't grasp the value prop of a "home profile".
A quote from a research participant:
A quote from a research participant:
"I'm equating [Profile] to settings and I'm not sure why we don't just call it Settings."
"I'm equating [Profile] to settings and I'm not sure why we don't just call it Settings."
The term “profile” also wasn’t clear:
The term “profile” also wasn’t clear:
“Is this my digital-wellbeing profile, or the whole house?"
“Is this my digital-wellbeing profile, or the whole house?"
With that, our efforts recentered on my hypothesis that we needed a settings tab*.
With that, our efforts recentered on my hypothesis that we needed a settings tab*.
There was still a lot of interest in more warmth and larger, more colorful visuals.
There was still a lot of interest in more warmth and larger, more colorful visuals.
However, I noticed we were making it more difficult for people to get to the most-visited settings items—in the following image, see the most-visited items (highlighted in pink) in relation to the fold.
However, I noticed we were making it more difficult for people to get to the most-visited settings items—in the following image, see the most-visited items (highlighted in pink) in relation to the fold.
Research confirmed the need for reliable settings destinations and the need to get to important items quickly.
Research confirmed the need for reliable settings destinations and the need to get to important items quickly.
Users were split on the home photo and functionality vs. aesthetic and “fun”.
Users were split on the home photo and functionality vs. aesthetic and “fun”.
They were aware that the photo pushed down the items they wanted to get to.
They were aware that the photo pushed down the items they wanted to get to.
A recurring ask was to make the home member avatars larger, and even include email addresses.
A recurring ask was to make the home member avatars larger, and even include email addresses.
But we found that people rarely edit home members, people usually know who they moved in with, and a separate screen is better for privacy needs like seeing members' email addresses.
But we found that people rarely edit home members, people usually know who they moved in with, and a separate screen is better for privacy needs like seeing members' email addresses.
“I would like the ability to rearrange the list because I don't know how often I would want to interact with ‘household’. I would prefer devices and services up top.”
“I would like the ability to rearrange the list because I don't know how often I would want to interact with ‘household’. I would prefer devices and services up top.”
-UX research participant
-UX research participant
And eventually, it turned out that we didn't have the resources for the new imagery that a lot of the warmth relied on.
And eventually, it turned out that we didn't have the resources for the new imagery that a lot of the warmth relied on.
We didn't have the photography or technical infrastructure to provide device photography or to show which services were linked to people in the home (an account + account + home settings relationship).
We didn't have the photography or technical infrastructure to provide device photography or to show which services were linked to people in the home (an account + account + home settings relationship).
😞
No budget or time to build an infrastructure for device photography
😞
No budget or time to build an infrastructure for device photography
😞
No Assistant-team APIs to extract which services members had linked
😞
No Assistant-team APIs to extract which services members had linked
Final decision: list items or carousels?
Final decision: list items or carousels?
One systems designer and I were still favoring a very utilitarian list-item-only approach.
One systems designer and I were still favoring a very utilitarian list-item-only approach.
In the last usability studies, my research partner shared that we could go either way. The user preference wasn't strongly for either direction.
In the last usability studies, my research partner shared that we could go either way. The user preference wasn't strongly for either direction.
In the end, I chose to keep the device and services carousels because these UI mechanisms put more touchpoints at people's fingertips (the pink highlights are items from the top-ten most-tapped settings items from our metrics before the redesign)
In the end, I chose to keep the device and services carousels because these UI mechanisms put more touchpoints at people's fingertips (the pink highlights are items from the top-ten most-tapped settings items from our metrics before the redesign)
❌
List-only layout + text summaries
❌
List-only layout + text summaries
(with most-tapped items highlighted)
✅
Final UX: Avatar strip + two carousels
✅
Final UX: Avatar strip + two carousels
(with most-tapped items highlighted)
Navigating challenges:
Break from the system when it makes sense
Navigating challenges:
Break from the system when it makes sense
The final UX includes a "+ Add" floating-action-button in the "Devices" tab, even though adding devices is not a "day-to-day control" need.
The final UX includes a "+ Add" floating-action-button in the "Devices" tab, even though adding devices is not a "day-to-day control" need.
I prefer opinionated stances on design systems, but I also believe in breaking from a system when it makes sense. I often see that users simply expect to manage things (devices) in the place where they see those things (in the "Devices" tab).
I prefer opinionated stances on design systems, but I also believe in breaking from a system when it makes sense. I often see that users simply expect to manage things (devices) in the place where they see those things (in the "Devices" tab).
But not everybody was on board.
But not everybody was on board.
The initial premise was that the "Devices" tab would be for everyday control, and the "Settings" tab would be for management
The initial premise was that the "Devices" tab would be for everyday control, and the "Settings" tab would be for management
Initially, I also thought this could work. Why put an add-device button in an day-to-day-use screen when you rarely add devices?
Initially, I also thought this could work. Why put an add-device button in an day-to-day-use screen when you rarely add devices?
But I had a small suspicion that users would not have the same mental model that we had.
But I had a small suspicion that users would not have the same mental model that we had.
I requested a UXR study dedicated to critical settings journeys*. The results show about half of users went to the "wrong" tab.
I requested a UXR study dedicated to critical settings journeys*. The results show about half of users went to the "wrong" tab.
*These tasks included moving devices, getting to device settings, removing devices, and adding devices.
Research and I thought the results were clear...
Research and I thought the results were clear...
I put forward that the "Devices" tab actually needs an always-visible floating action button, not just a scroll-to-the-end-to-see-it "Add devices" button.
I put forward that the "Devices" tab actually needs an always-visible floating action button, not just a scroll-to-the-end-to-see-it "Add devices" button.
...but partners for other areas of the app still prefered the less-visible UI.
...but partners for other areas of the app still prefered the less-visible UI.
Settings' team's ask
Settings' team's ask
(always-visible floating action button)
Partner team's preference
Partner team's preference
(scroll-to-the-end-to-see-it button)
As a backup measure, I prepared an add-devices-in-the-new-settings-tab instruction in the new user education that would accompany the app redesign
As a backup measure, I prepared an add-devices-in-the-new-settings-tab instruction in the new user education that would accompany the app redesign
Normally, you don't want to have to teach users much about settings, but the research team agreed this action was critical enough to warrant a user-education feature.
Normally, you don't want to have to teach users much about settings, but the research team agreed this action was critical enough to warrant a user-education feature.
When possible, I also asked to include the "add device" user journey in research studies
When possible, I also asked to include the "add device" user journey in research studies
Even after 6 months, and even with the user education*, the data showed that still about half of people went to the "wrong" tab.
Even after 6 months, and even with the user education*, the data showed that still about half of people went to the "wrong" tab.
*We had a user-education learning: Our research found that people often don't read wizard-like user education sequences. It's better to use tool-tips if your product needs users to learn what you're pointing out.
Many participants clicked around the “Favorites” and “Devices” tab.
Many participants clicked around the “Favorites” and “Devices” tab.
[UXR, July 2021]
5/9 participants had some difficulty
completing the “add a new device” task.
5/9 participants had some difficulty
completing the “add a new device” task.
[UXR, Dec 2021]
Observation:
Majority of participants did not think to go to Settings to add/link devices
Observation:
Majority of participants did not think to go to Settings to add/link devices
Recommendation:
Add an additional “+ Add” entrypoint in the “Devices” tab
Recommendation:
Add an additional “+ Add” entrypoint in the “Devices” tab
[UXR, July 2021]
Eventually, we had agreement:
Eventually, we had agreement:
The data was clear and the partner teams agreed: it made sense to break from the system
The data was clear and the partner teams agreed: it made sense to break from the system
Outcomes for the tab
Outcomes for the tab
We addressed the original problems & challenges:
We addressed the original problems & challenges:
Settings were consolidated into a new tab—any setting affecting the home can be reached via the tab
Met the needs of stakeholders who wanted to retain items in the tab, but kept the list curated and digestible for users
9to5Google: "Both [the Home features and Nest services & support sections] are concise and easy to navigate…
9to5Google: "Both [the Home features and Nest services & support sections] are concise and easy to navigate…
I genuinely can’t express how big of an improvement this is.”
I genuinely can’t express how big of an improvement this is.”
A tidier "Add" menu still has what users need:
A tidier "Add" menu still has what users need:
9to5Google: "There’s also an 'add' button that shows every action that you might expect to fit that description."
9to5Google: "There’s also an 'add' button that shows every action that you might expect to fit that description."
Problems that were addressed
Problems that were addressed
The old menu included settings, not just items to add
It included a low-usage, paid customer-service line
Unconsolidated media services
Lack of a Works with Google entrypoint to link compatible devices
(Ongoing)
Google Material 3 updates
(Ongoing)
Google Material 3 updates
The project
The project
Applying accessibility standards, responsive layout behaviors, and Google Material 3 dynamic color to the high-priority settings screens in Google Home app
As we finish this work, it will increase the speed of design and engineering as more of the app is on an infrastructure that scales
My contributions
My contributions
UX strategy
UX/UI design
Mentor & hiring manager
The team
The team
1 Lead UXD (myself)
1 UXD
1 product manager
3 engineers
Partners
Partners
1 accessibility UXD
10 device verticals (PM, UX, Eng)
Launch dates
Launch dates
[Ongoing]
Current challenges
Current challenges
Competing features and new product launches
Tight engineering resources
The cost of undoing technical debt is high at the moment
Pre- and post-Google Material 3 design
Pre- and post-Google Material 3 design
Examples of the visual and interaction design before and after the Google Material 3 updates
Examples of the visual and interaction design before and after the Google Material 3 updates
Page updated
Google Sites
Report abuse