E-commerce Revenue Drop Diagnosis Case

Context

Week T net revenue is down 20% versus the prior 4-week moving-average baseline. You have weekly e-commerce data and the following signals:

• Sessions: flat (+0.5% WoW)
• CVR fell; AOV changed (magnitude not provided)
• Returns increased from 7% to 10%; cancellations from 1% to 2%
• Stockouts on top-50 SKUs rose from 3% to 9%
• A price increase (+8%) began 3 weeks prior
• New free shipping threshold ($75) began 2 weeks prior
• Paid media mix shifted toward upper-funnel
• A competitor ran a 25%-off sitewide sale in Week T

Assume "revenue" refers to net recognized revenue after post-purchase deductions unless clarified otherwise.

Tasks

1. Decompose the revenue delta into traffic × CVR × AOV, and further into:
   • AOV subcomponents: price, discount/promo depth, and mix/units per order
   • Post-purchase deductions: returns and cancellations
   • Stockout-driven lost sales Provide formulas and a stepwise waterfall with approximate percentage-point contributions for each factor.
2. List the quickest falsifiable checks you would run within 24 hours to validate root causes (e.g., device-specific CVR drop, OOS by category, promo overlap, latency, checkout errors, funnel step breakage, channel mix, cohorts).
3. Causal assessment: Propose a geo- or cohort-level difference-in-differences (DiD) or interruption model to isolate the impact of the free-shipping change from the competitor sale and seasonality. Specify treatment/control groups, pre-trends checks, and robustness tests.
4. Experiments and fixes: Outline targeted tests (pricing, shipping-threshold A/B, back-in-stock alerts, assortment substitution) and the success metrics you’d monitor (lift, profit, LTV impact).