Meta & Children's Online SafetyIG Mental Well-being: Behavior Reduction Strategy: Late Night Use
Raw text
Case 4:22-md-03047-YGR Document 2758-15 Filed 02/20/26 Page 1 of 24 AMENDED Exhibit 305 PLAINTIFFS' OMNIBUS OPPOSITION TO DEFENDANTS' MOTIONS FOR SUMMARY JUDGMENT Case No.: 4:22-md-03047-YGR MDL No. 3047 In Re: Social Media Adolescent Addiction/Personal Injury Products Liability Litigation Case 4:22-md-03047-YGR Document 2758-15 Filed 02/20/26 Page 2 of 24 Document Provided in Native Format HIGHLY CONFIDENTIAL (COMPETITOR) META3047MDL-096-00000003 Case 4:22-md-03047-YGR Document 2758-15 Filed 02/20/26 Page 3 of 24 IG Mental Well-being Behavior Reduction Strategy: Late Night Use [A/C Priv] WB Leads Review 02 Feb 2023 Case 4:22-md-03047-YGR Document 2758-15 Filed 02/20/26 Objectives Page 4 of 24 Goals Align on plan to achieve late night use reduction goal Non goals Design and content feedback (directional mocks only) Case 4:22-md-03047-YGR Document 2758-15 Filed 02/20/26 Page 5 of 24 Executive summary Asbart of our multi-half strategy for Problematic Use, IG Mental Wellbeing team has an IG level goal to reduce teen late night time spent/DAU by in H1'23. For our target teens, drivers of night time use are notifications, long sessions, and Reels consumption. Based on this, potential levers to meet this goal are: : Through a continuous test and learn approach, we plan to ship levers that are most impactful for our goal to reduce late night time spent. Our metric is still being operationalized, but rough estimations suggest that currently we are at (not stat-sig). Based on current eng bandwidth we expect to ship #1 and prioritize either #2a OR #2b. Our rough end of half estimate is between and Please note: We will need to fine tune our approach and revisit hypothesis as we get test results. *PS: This value was when pre-read went out, assuming we ship all the levers. Case 4:22-md-03047-YGR Document 2758-15 Filed 02/20/26 Page 6 of 24 Guiding principles Empower through controls We empower teens to be more intentional about their time by creating contextual behavioral awareness and offering relevant controls. Identify and target teens in need We use research and expert feedback to generate conviction in our products and target adoption with teens who can benefit most from them. Drive long-term engagement through well-being Despite potential near term ecosystem impact, our products aim to create long term retention and engagement by encouraging healthier social media use. Foster healthy norms Our products and amplification tools enable teens to set norms around usage habits and influence their communities. Case 4:22-md-03047-YGR Document 2758-15 Filed 02/20/26 Page 7 of 24 H1'23 Goal The MWB goal for H1 2023 is to reduce late night (10PM - 4AM) time spent by Users included in goal What must be done to meet this goal Currently we estimate we are at compared to goal of - (note: this is only an estimate, goal metric is still being operationalized). Case 4:22-md-03047-YGR Document 2758-15 Filed 02/20/26 Quiet Mode Post Launch Status Wils • • Page 8 of 24 users total as of 1/28 of whom are stated age teens) Our test showed a reduction in topline late night time spent (12 AM - 4 AM) of- among stated age teens with 5 minutes of late night time spent reduction in the past week. At a strong adoption is equivalent to per adopter. We also saw a topline lift in core foreground sessions time spent as well as directional lift in VPVs that we hypothesize is driven by the two digests (activity and direct) and Needs Attention • Retention at 60 days is and has not yet flattened (compared to retention for Take a Break.) • • Additional effort will be needed to retain auto reply and digest functionality as E2EE rolls out on IGD in 2023. Will follow up on any trade offs against our roadmap. We continue to maintain a high bar on product quality, most recently - IGD notes tray presence suppression (ref), hold out leak (ref) and QM disable issue (ref). Product dashboard here Retention definition: Number of users with the feature still enabled divided by number of users who could have the feature enabled (i.e., enabled long enough ago to be included in the specified "days since enable" figure) Case 4:22-md-03047-YGR Document 2758-15 Filed 02/20/26 Page 9 of 24 Existing coverage and opportunities Quiet Mode addresses late night use primarily through intentionality, with teens opting in to use it. Additional opportunities exist for awareness products (like insights) and in the moment attentiveness products (like contextual nudges). To maximize impact (1, 2, 3), we'll prioritize improving Quiet Mode and contextual attentiveness products. PRIORITIZED PU-RELATED BEHAVIORS Frequent checking Overall amount of time spent Late night use (passive and overall) Content consumption patterns (includes long scrolls) LEVERS FOR SENSE OF CONTROL Awareness Products that give people feedback about their own behavior and experiences on platform. Attentiveness Products that give information in context - tied to specific behavior. Intentionality and choice Products that enable behavior change or make change easier. Quiet mode Your Activity Daily Limit Take a break Daily Limit Take a break Quiet mode Alternative topic nudge Multi-hide in Explore, Sensitive content controls, Persistent controls (ie WAIST) Users don't expect usage insights alone to change behavior. Expert input supports this: https://www.google.com/url?q=https://docs.google.com/document/d/16BMDeel xli_QSkRhxiiHJ5csXf6xYftITFD9fle5998/edit?usp%3Dsharing&sa=D&source= editors&ust=1675145072381267&usg=AOvVaw0z0QooMAjxRjlvg6m8cxUX https://fb.workplace.com/notes/537148410448191/ Case 4:22-md-03047-YGR Document 2758-15 Filed 02/20/26 Page 10 of 24 Late Night Use 4 Thip90 threshold for late night (10 PM - 4 AM) time spent over the last 28 days corresponds to 25 minutes per day on IG. • • • DRIVERS OF LATE NIGHT USE (ref) Notifications: 35% of late night time spent for p90+ teens comes from sessions that had a notification delivered within 5 minutes of starting a session on the app. Long sessions: 64% of late night time spent for p90+ teens comes from sessions over 5 minutes long, and 35% comes from sessions over 15 minutes long Reels Consumption: Time spent on Direct and consuming Reels accounts for 56% of late night time spent for p90+ teens (30% and 26%, respectively) With Quiet Mode, we've already built table stakes features that mute notifications and set social norms in direct. Based on the above, we have additional opportunities to target long sessions and Reels consumption. Source: https://docs.google.com/spreadsheets/d/1h9F9CimDR_RZQ46xy- xenv-GE6vne8WYVdso_qC9Qo/edit#gid=0 I aggregated all late night (10-4) time spent for recommended age teens over a 28 day period. I then calculated the p90 value of time spent over that 28 day period and divided it by 28 days to get the daily average time spent required for a user to qualify for p90 over that 28 day period Research on late-night use is a bit limited, but we have the following: US teen WAU report that notifications (21%), ephemeral content (21%), autoplay videos (23%), recommended content (23%), and content by followed users (21%) make it harder to manage time spent on IG (Hanko, PU foundational survey). In interviews, teen WAU in US and IN attribute difficulty managing late-night use to similar sources, including perceiving more new content at night, actively messaging with friends, notifications, and engaging content, especially on surfaces with unlimited content (e.g., Reels, Explore) (Hanko, teen interviews). Case 4:22-md-03047-YGR Document 2758-15 Filed 02/20/26 Page 11 of 24 Case 4:22-md-03047-YGR Document 2758-15 Filed 02/20/26 Page 12 of 24 Case 4:22-md-03047-YGR Document 2758-15 Filed 02/20/26 Page 13 of 24 Case 4:22-md-03047-YGR Document 2758-15 Filed 02/20/26 Page 14 of 24 Case 4:22-md-03047-YGR Document 2758-15 Filed 02/20/26 Page 15 of 24 Case 4:22-md-03047-YGR Document 2758-15 Filed 02/20/26 Page 16 of 24 Case 4:22-md-03047-YGR Document 2758-15 Filed 02/20/26 Page 17 of 24 Case 4:22-md-03047-YGR Document 2758-15 Filed 02/20/26 Page 18 of 24 Case 4:22-md-03047-YGR Document 2758-15 Filed 02/20/26 Page 19 of 24 Case 4:22-md-03047-YGR Document 2758-15 Filed 02/20/26 Page 20 of 24 Case 4:22-md-03047-YGR Document 2758-15 Filed 02/20/26 Page 21 of 24 Case 4:22-md-03047-YGR Document 2758-15 Filed 02/20/26 Page 22 of 24 Case 4:22-md-03047-YGR Document 2758-15 Filed 02/20/26 Page 23 of 24 Case 4:22-md-03047-YGR Document 2758-15 Filed 02/20/26 Page 24 of 24