Bonnom

Bonnom

UX Research Lead & Designer

During my PM internship at amo, I led user research and designed core UX improvements. Combining insights from over 50 user interviews and product analytics, we identified key friction points in the friend invitation flow. Based on these findings, I designed improvements to the onboarding experience that significantly increased viral growth and conversion rates.

Additionally, I designed UX/UI that leveraged new iOS functionalities, such as App Clips and widgets, to improve new user engagement. I also established and managed a global localization system, coordinating with 30 localizers and engineers to launch and maintain 3 apps across 15 languages with daily releases.

SPENCER'S SCOPE

Product Management

UX Research

UX/UI Design

Localization Lead

TIMELINE

Post-Grad Internship (Sep. 23-Mar. 24)

TECH COLLABORATORS

Kevin Bessiere, iOS Lead

Maxime Gerardin, PM

Jose Goncalves, UI Designer

LOCALIZATION COLLABORATORS

Jason Akakpo, iOS Lead

Cyril Mottier, Android Lead

Finding Friends on amo
A UX Research Case Study

The Problem

In social apps, the k-factor (viral coefficient) measures how well each user brings in new users, and when a k-factor is greater than 1, on average each user brings in more than one new user. At amo, we measured our path to a K factor > 1 by one clear early indicator: how many users invited at least one friend within their first day.

Part of my UX research mandate was to optimize the friend invite flow and increase invites sent for new users. The core problem I identified was that users struggled to find relevant contacts within our invitation flow. What made this a compelling problem to solve is that users had already showed high intent—they'd already granted contact permissions and were ready to find friends to share the app with.

Research Methodology

For all user tests at amo, we prioritized in-person sessions with participants in our target demographic (18-30) to capture detailed interactions and identify potential bugs.

Our dual-camera setup enabled comprehensive feedback:

  • One camera recording facial expressions to capture emotional responses

  • Second camera focused on screen interactions to track usability issues


After each session, we compiled the footage and shared findings with the team across three main categories:

  • Wins: Successful interactions and positive responses

  • Improvements: Areas needing UX refinement

  • Bugs: Technical issues with documented reproduction steps for engineering in Linear tickets

Example User Session

From an illuminating session with "Jane," a 23-year-old freelancer based in Paris, there were four key qualitative observations:

  1. Unlikely Friend Suggestions: When presented with a preselected group of contacts, Jane felt forced to send an unnecessary invite to her mother. This created friction in a flow where users primarily should want to connect with peers.

  2. Core UX Discovery Issues: Jane couldn't find the search function initially, limiting her ability to invite relevant contacts from her network.

  3. Eventual Success: After discovering the "more contacts" button, Jane found the full contact list. Once there, she noted it was "easy" when she could finally invite her boyfriend. (This is the purple flow in the next section.)

  4. Willingness to Abandon Flow: Jane expressed relief upon finding the "I don't want to invite more friends" option to abandon the invite flow - a behavior that worked directly against our goals for user growth.

From an illuminating session with "Jane," a 23-year-old freelancer based in Paris, there were four key qualitative observations:

  1. Unlikely Friend Suggestions: When presented with a preselected group of contacts, Jane felt forced to send an unnecessary invite to her mother. This created friction in a flow where users primarily should want to connect with peers.

  2. Core UX Discovery Issues: Jane couldn't find the search function initially, limiting her ability to invite relevant contacts from her network.

  3. Eventual Success: After discovering the "more contacts" button, Jane found the full contact list. Once there, she noted it was "easy" when she could finally invite her boyfriend. (This is the purple flow in the next section.)

  4. Willingness to Abandon Flow: Jane expressed relief upon finding the "I don't want to invite more friends" option to abandon the invite flow - a behavior that worked directly against our goals for user growth.

Product Data Analysis

Using Jane's session as a starting point, I dove into Amplitude to validate if her behavior patterns were common among users. This quantitative analysis confirmed our qualitative findings:

  1. Most users who granted contact permissions were following the "Unlikely Friend Suggestions" path, skipping the subsequent invitation flow steps (shown in green).

  2. The path Jane eventually found—leading to her "easy" moment—had our highest conversion rate (~23.5%) but was our least traveled path (in purple).


This mismatch between our most successful path and actual user behavior presented a clear opportunity for optimization via design.

The Fewest Users Go Down Our Ideal Path

Those who go down ideal path convert at ~25%

Those who go down ideal path convert at ~25%

The Fewest Users Go Down Our Ideal Path

The Fewest Users Go Down Our Ideal Path

Those who go down ideal path convert at ~23.5%

Design Solutions

Original Experience

The initial design had a few key points of friction:
  1. Users expected an upfront search experience; instead, search functionality was hidden behind a button

  2. A prominent negative option ("I don't want to invite more friends") that discouraged future engagement

  3. Unclear invitation goals despite showing three empty circles

    1. Originally these were non-interactive UI elements that confused users

Implemented Changes

  1. Gamified Copy Introduced copy indicating the goal to invite three friends while maintaining the ability to proceed with fewer invites.
  2. Intuitive Search Added a search bar at the top, recognizing that users often had specific friends in mind when entering the flow. When activated, this automatically expanded to the full contact list.
  3. Interactive Progress Made the three circles interactive and tied them to user actions, providing clear feedback as users added friends to their invitation list.
  4. Positive Language Replaced negative opt-out copy with "Find more friends" to encourage exploration of our highest converting path.

Impact & KPIs

Impact & KPIs

Because I don't know the app well I usually hesitate when inviting friends. In the onboarding, I could easily find friends who I'd want to share silly photos with

"Scott", 23 year old amo user after redesign

Scott, 24 year old amo user after redesign

These design changes led to significant improvements in our onboarding metrics:

  1. Increased Engagement Average invites sent per user increased from 1.2 to 2.5 within a week of launching, representing a 108% increase in invitation rate

  2. Improved Flow More users discovered and used the full contact list, our highest converting path

  3. Better User Experience Users were more likely to engage with subsequent invitation prompts, showing our copy changes successfully reduced resistance


Timeline: Changes were put into the app on February 1st, experiment data was analyzed daily through February 9th before changes were kept in the app.

Other Product Impact

Global Localization System


During my time at amo, I built and managed a comprehensive localization program:

  • Scaled to 15 international markets

  • Coordinated 30 localizers and 3 engineers

  • Maintained daily release cadence

  • Ensured cultural relevance and technical accuracy

Exploring iOS's Boundaries (UX Design)


Leveraging iOS's latest functionality, I designed the core UX for an app clip as a new entry point to amo. App clips are lightweight versions of apps that don't require an install.


For Tilt (previously Capture), one of amo's apps, I focused on allowing users to experience their friends' content before committing to a download. Users would receive an app clip when friends shared content with them, leveraging our core sharing mechanics. Within the app clip, we strategically prompted app installation so users could create their own content.

From top left to bottom right

  • User receives text invite to view a friend's capture

  • User sees instructions on how to "Check out Spencer's Capture"

  • After consuming content, user is prompted by a banner to download Capture

  • App store page opens

Other Product Impact

Global Localization System


During my time at amo, I built and managed a comprehensive localization program:

  • Scaled to 15 international markets

  • Coordinated 30 localizers and 3 engineers

  • Maintained daily release cadence

  • Ensured cultural relevance and technical accuracy

Exploring iOS's Boundaries (UX Design)


Leveraging iOS's latest functionality, I designed the core UX for an app clip as a new entry point to amo. App clips are lightweight versions of apps that don't require an install.


For Tilt (previously Capture), one of amo's apps, I focused on allowing users to experience their friends' content before committing to a download. Users would receive an app clip when friends shared content with them, leveraging our core sharing mechanics. Within the app clip, we strategically prompted app installation so users could create their own content.

From top to bottom:

  • User receives text invite to view a friend's capture

  • User sees instructions on how to "Check out Spencer's Capture"

  • After consuming content, user is prompted by a banner to download Capture

  • App store page opens