Business Context

Lyft has launched a new rider reactivation campaign targeting users who have not taken a trip in the last 90 days. The Growth team sent a mix of push notifications, email, and in-app incentives to 1.2M dormant riders, and leadership wants to know whether the campaign actually drove incremental rider activity and revenue.

Metric Scenario

Before launch, the baseline 30-day reactivation rate for dormant riders was 8.5%. In the first month after launch, 132,000 targeted riders took at least one trip, implying an observed reactivation rate of 11.0%. Of those reactivated riders, 58% took a second trip within 30 days, average gross bookings were $24 per reactivated rider, and the campaign cost $1.8M in discounts and messaging spend. A holdout group of 200,000 similar dormant riders had a 9.2% reactivation rate over the same period.

The VP of Growth is asking which KPI should define success, how to separate true incremental impact from natural return behavior, and what guardrails should be monitored so the campaign does not simply buy back low-value riders.

Requirements

1. Define the primary success metric for the rider reactivation campaign and explain why it is the best KPI.
2. Calculate the campaign's observed and incremental performance using the data above.
3. Propose a metric decomposition to diagnose which channels, rider segments, and incentive types drove results.
4. Identify guardrail metrics to ensure the campaign is profitable and does not harm rider experience.
5. Recommend how often the team should review leading vs. lagging metrics.

Data Available

• campaign_exposures: rider_id, campaign_id, channel, send_time, incentive_amount, opened, clicked
• trip_facts: rider_id, trip_id, request_time, completed_flag, gross_bookings, promo_amount
• rider_dim: rider_id, city, tenure_months, lifetime_trips, last_trip_date
• payments_finance: rider_id, contribution_margin, refunds, promo_spend, payment_failures