From Platform Data to Personal Insight: How Users Make Sense of and Reflect on Personalized Social Media Annual Recaps

Paper Title

From Platform Data to Personal Insight: How Users Make Sense of and Reflect on Personalized Social Media Annual Recaps

Publication Info

• Topic area: User sensemaking and reflection on platform-curated personal informatics data.
• Keywords: Personal informatics, sensemaking, reflection, social media, annual recaps, data visualization, user agency, algorithmic narratives, privacy, HCI.

Background and Problem

• Problem / challenge: Existing research on personal informatics (PI) focuses on user-initiated data tracking, leaving a gap in understanding how users make sense of unsolicited, platform-curated data presentations like annual recaps.
• Significance: Personalized annual recaps are widely popular and have the potential to support self-reflection, memory, and identity presentation. However, their unsolicited nature and opaque algorithmic curation challenge user agency and meaningful engagement.
• Motivation and related work: Prior work in PI and lived informatics has explored self-tracked data for reflection but has not addressed platform-curated data. This study builds on sensemaking and reflection theories to investigate how users derive insights from such data presentations.

Solution

• Proposed approach: A qualitative investigation into how users make sense of and reflect on personalized annual recaps, combining semi-structured interviews with artifact analysis of recap screenshots.
• Novelty:
  1. Identification of seven data presentation types in annual recaps and their associations with reflection levels.
  2. Mapping of five sensemaking activities to four reflection levels, revealing distinct sensemaking paths.
  3. Extension of personal informatics research to platform-curated data, highlighting challenges in user agency and algorithmic narratives.
• Procedure and key techniques:
  • Conducted semi-structured interviews with 20 participants who viewed annual recaps from seven platforms.
  • Analyzed 66 screenshots to develop a taxonomy of data presentation types.
  • Applied codebook thematic analysis to map sensemaking activities and reflection levels.
  • Identified four sensemaking paths that connect data presentations to reflection outcomes.

Results

• Concrete findings:
  • Seven data presentation types identified: extreme detail, keywords, data summary, social interaction, comparison, trend, and persona.
  • Five sensemaking activities: scanning, data annotation, verifying, seeking plausible explanations, and negotiating with data.
  • Four sensemaking paths lead to reflection levels: descriptive (R1), dialogic (R2), transformative (R3), and critical (R4).
  • Persona presentations were most associated with critical reflection (33%), while data summaries were prominent in transformative reflection (27%).
• Advantage over baselines: Provides a detailed framework linking sensemaking activities to reflection levels, addressing gaps in understanding unsolicited data presentations in PI.
• Experiments / evaluation:
  • Data collected from 20 participants (aged 18–25) across seven platforms (e.g., Xiaohongshu, QQ Music, Bilibili).
  • Analysis of 66 screenshots and interview transcripts using thematic coding and co-occurrence analysis.
• Limitations and future work:
  • Cultural and platform-specific focus on Chinese social media limits generalizability.
  • Non-linear sensemaking paths and variations in prior exposure to recaps were not fully captured.
  • Future research should explore longitudinal impacts, cross-cultural contexts, and multimodal data collection (e.g., eye-tracking).

Summary

This study investigates how users make sense of and reflect on personalized annual recaps from social media platforms. By analyzing seven data presentation types and mapping five sensemaking activities to four reflection levels, the research reveals distinct sensemaking paths that facilitate progression from descriptive to critical reflection. The findings extend personal informatics research to platform-curated data, highlighting challenges in user agency and algorithmic narratives. Design implications include strategic sequencing of data presentations, scaffolding sensemaking activities, and supporting social sensemaking to enhance reflection. These insights contribute to transforming annual recaps from passive consumption into active reflection tools.