OVERVIEW

Our previous research revealed that users typically shared their AI creations privately through DMs or Stories. However, market research revealed a growing trend of small creators leveraging GenAI tools to engage broader online communities in new ways.

To understand this shift, I conducted a qualitative diary study with 13 Gen Z creative hobbyists who shared their AI creations publicly beyond friends and family.

IMPACT

This research prioritized the launch of Imagine in Threads by identifying new user archetypes and uncovering three product opportunities for Meta AI. These insights informed JTBD statements that guided a design strategy sprint involving leadership, product management, and design.
‍
I have omitted or edited confidential information in this case study. All information in this case study is my own and does not reflect the views of Meta.

TYPE

Foundational Research

ROLE

UX Researcher
Team: UXR Lead, Research Director, Design Director

METHODS

Diary Study

TIMEFRAME

6 weeks

Goals

1) To understand the motivations and conditions for GenAI users who widely broadcast their AI creations online.
2) To uncover product opportunities to drive engagement with Meta AI across Meta platforms.

Research Questions

1) What GenAI tools are creative hobbyists leveraging to create their AI content?
2) What motivates creative hobbyists to broadcast their GenAI creations to their online communities?
3) What unique value does GenAI bring to these online communities?

Research Plan

Recruit

N = 13 Gen Z creative hobbyists (teens and young adults). We identified "creative hobbyists" based on behavioral trends that our quantitative teams pulled. *The goal was to recruit 25 Gen Z creative hobbyists (goal: 20 responses), but the criteria was hard to meet due to DScout's panel availability and our timeline.
Must create AI content and share this content beyond close family and friends
Must share AI content at least 1x per week
Must be actively involved in their online communities, either by sharing or commenting on posts at least 1x per week

Methodology

Initial Generative Screener: I initially sent out a screener that 1) recruited participants based on our criteria and 2) asked screened participants to elaborate on their AI creations and online community involvement. Based on their responses, we identified key niches such as gaming and fan fiction communities.
Diary Study (3 parts): The 18-day diary study was completed on DScout and consisted of three parts:
1) Get to Know: Favorite AI tools, what they create, and their online communities
2) Active Creation: Motivations, goals, creation experiences (delights/challenges)
3) Community: Where they share AI images/content, how people react, how they engage with AI content online

Timeline

This study took a total of 6 weeks, including recruit and synthesis.

Learnings & Reflections

Challenge: Advocating for the right method. Our design director initially suggested conducting in-depth interviews to allow designers to observe participants in real time. However, I recommended a diary study instead, as it would enable us to capture authentic moments from participants' creation processes. To maintain stakeholder engagement, I developed a "Get to Know" slide deck with participant profiles and provided weekly updates, incorporating 2-minute snippets from participants’ video responses.
Challenge: Time. This research needed to be done before the design strategy sprint. Given more time, I would have invited participants to in-depth interviews to co-design innovative ways to connect with their online interest communities through GenAI tools.
Learning: Involve leadership early on. I initially focused on telling a narrative-driven story of the research findings. However, after feedback, I realized I needed to better align my recommendations with Meta's larger GenAI goals. I worked with my UXR lead and directors to revise the deck, making sure to integrate those organizational objectives throughout, so our findings were more strategically relevant.

Strategic Opportunities for GenAI in Interest Communities

Background