Design offline segments for Meta Portal retail

Meta Portal is a plug‑in home video‑calling device sold via offline retail. Target segments are not finalized. You have historical, anonymized Facebook video‑calling telemetry from app and web (no Portal usage yet). Today is 2025-09-01. Design a segmentation and go‑to‑market plan:

1. Propose 4–6 mutually exclusive, data‑driven segments (e.g., heavy vs light vs non video‑callers; app‑first vs web‑first; family distance; diaspora/expat; caregivers; SMB remote‑workers). For each, state a falsifiable hypothesis for why a plug‑in, stationary device increases utility vs phone/PC.
2. Opportunity sizing: outline how you would estimate SAM/SOM per segment in the US given tables daily_users(date, user_id, country, dau_flag) and video_calls(date, caller_id, recipient_id, duration_sec). Specify required joins/filters, key assumptions (e.g., minimum weekly active calling threshold), and how you will avoid double‑counting users who qualify for multiple segments.
3. Prioritization: define a scoring formula that combines estimated SOM, expected adoption uplift, margin/CAC constraints, and operational feasibility (retailer coverage, demoability, return rates). Explain how you’d set cutoffs and perform sensitivity analysis.
4. Offline validation: design a 4‑week in‑store/geo experiment to test segment‑targeted displays/offers. Pick the unit of randomization (store or DMA), define primary KPIs (demo→purchase conversion, 14‑day activation rate, incremental weekly call minutes), guardrails, instrumentation, and a power analysis with assumed baselines. Address spillovers, geo imbalance, and stock‑outs.
5. Incorporate the plug‑in constraint: explain how a stationary, always‑powered device changes which segments to target and which retailers/aisles to use. Propose privacy‑respecting proxy signals that suggest a user has a stable home calling context (without needing exact addresses).
6. Before large spend, describe an uplift‑modeling or lightweight geo‑targeting pilot to verify that your segments predict adoption.