Diagnostic Plan: Account Switching Up, Active Users Down

Context

You observed a sudden pattern: the number of users switching accounts increased, while overall active users decreased. Assume:

• Active user = distinct user_id with any qualifying activity in a day (DAU) unless otherwise noted.
• person_id is a cross-account identity (e.g., same human across multiple user_ids via SSO/identity graph).
• An "account switch" is an explicit event (e.g., switch_account, add_account) or inferred by consecutive session logins to different user_ids from the same device/session.

Provide a concrete diagnostic plan covering (a)–(f).

Requirements

(a) List 5–7 plausible hypotheses spanning measurement artifacts, product changes, and behavior shifts.

(b) Specify the exact metrics and breakdowns you will pull (e.g., by client, geo, new vs returning, person_id vs user_id, cohorts, time-of-day).

(c) Describe at least three targeted slice analyses to confirm or refute cross-contamination between accounts (i.e., multiple accounts used by the same person/device).

(d) Outline event/experiment diffing steps to identify recent launches correlated with the shift.

(e) Provide minimal SQL or pseudocode to quantify how many active user_ids map to the same person_id compared with last week.

(f) List immediate mitigations if the root cause is: (i) a logging bug, (ii) authentication friction, or (iii) users gaming policy limits by rotating accounts.