Qualitative Trajectory Arrays (Sentiment Drift Forecasting)

Description

Convert diary or ESM entries into per-participant trajectories of clarity, trust and frustration, cluster them into patterns such as 'fragile success', and use them to qualify quantitative forecasts.

Monte Carlo and SHAP extrapolate numbers; diaries and experience sampling extrapolate feelings over time. The playbook argues that even if completion rises short-term, negative sentiment drift predicts future drop-off ('Even when I succeeded, I felt drained'; 'After three tries, I stopped trusting the feature'). Diary entries are arrayed with columns for session index, perceived clarity (−2..+2), trust (1..5), frustration (0..100), success (Y/N) and a key quote, then sorted by outcome and clustered into patterns such as 'fast trust gain', 'fragile success' and 'churn risk'. Clusters are triangulated with top SHAP drivers (e.g. high mistake rate clusters with fragile success) and turned into what-if notes. The deliverable is a panel of two or three curated trajectories shown beside forecast curves to argue for emotional sustainability.

Attributes

Loss type
Community-proposed
Effort
Weeks

Steps

  1. Run a diary or ESM study across multiple sessions with structured prompts for clarity, trust and frustration plus an open quote.
  2. Array entries per participant: session index, clarity (−2..+2), trust (1..5), frustration (0..100), success (Y/N), key quote.
  3. Sort trajectories by outcome and cluster them into patterns (e.g. fast trust gain, fragile success, churn risk).
  4. Link clusters to the top quantitative drivers from SHAP or survival models.
  5. Write what-if notes, e.g. 'If SNR uplift holds and the rules wizard ships, expect the fragile cluster to shrink.'
  6. Present two or three curated trajectories alongside Monte Carlo curves and re-check them at 1, 3 and 6 months after launch.

Sources

  • Vivek Gopinath (2025), Forward-Thinking UX Research Playbook: A Mixed-Methods, Decision-Ready Approach from Prediction to Validation

Also known as

  • Sentiment drift trajectories
  • Diary trajectory arrays

Categories

  • Community-Proposed Frameworks