Investigate ride declines and test free trials

Shared rides in Los Angeles are declining 10% month over month. Put together a diagnostic and action plan: (a) Break the drop into acquisition (new riders), activation, cohort retention, and supply constraints using cohort tables plus time-of-day and geographic cuts; specify the exact plots/queries you would run and what each would confirm or rule out. (b) List plausible root causes (seasonality, price/ETA shifts, operations outages, competitor entry) and the smallest amount of data needed to eliminate each. (c) For a '$100/month unlimited after a 1-week free trial' offer, decide which users should get the trial (new users, churn-risk users, high-propensity payers) and why; define targeting rules. (d) Give the break-even formula for the trial: $\text{incremental LTV} \times \text{conversion} - \text{CAC} - \text{variable costs} - \text{fraud loss} \ge 0$; solve for the minimum conversion rate given inputs. (e) Sketch the A/B test design (unit, duration, sample size, stratification), guardrails (support load, refund rate, fraud), and success criteria for a 4-week run.

Overview: This question tests product analytics, cohort and funnel analysis, retention and supply diagnostics, pricing and subscription economics, and experiment design.

Community answers

Answer by SS

How can we run an A/B test when customers choose whether they want this offer or not?