Skip to main contentUse this guide to analyze lifetime orders (LTO) by cohort (typically first order month) and understand which acquisition cohorts generate the most repeat purchasing.
Recommended tables
Define the cohort
Most commonly:
- Cohort grain: first order month
- Cohort key: customer’s first valid purchase date (using SourceMedium-valid orders)
Tip: Keep cohort logic consistent with any LTV reporting you already use.
Build the analysis (high level)
- Filter to SourceMedium-valid orders (
is_order_sm_valid = TRUE).
- Determine each customer’s cohort (first valid order date).
- For each cohort, compute:
- Total orders
- Distinct customers
- Average lifetime orders =
total orders / distinct customers
- Add breakdowns (channel, product category, geography) only after the cohort math looks right.
Common pitfalls
- Mixing invalid/test/refunded orders with valid orders (always start with
is_order_sm_valid = TRUE).
- Comparing cohorts with different “age” (newer cohorts haven’t had time to accumulate repeat purchases).
- Blending channels (e.g., Amazon + Online DTC) when the question is channel-specific.
Next steps
