E-commerce Cohort Analysis: Repeat Purchase, Revenue per Cohort & Payback

What you'll build

What you will be able to do

Each lab ends with a concrete result that you build yourself and the checker confirms, cell by cell.

• Tag every order with the customer's first purchase date, a cohort label like 2026-01 and whether it is a new or a repeat order.
• Count each month's new customers with COUNTIFS between two dates, their share of all new customers, their average first order and the growth from one cohort to the next.
• Give every order its cohort month, its order month and its month number (0 = month of the first purchase), and see how much revenue comes after the first month.
• Flag each customer's first order in every month, build the triangle of active customers per cohort and month, and see how far order counts would overstate it.
• Turn the active-customer triangle into retention percentages, colour it as a heat map, build the weighted retention curve and measure how weak the promotion cohort is.
• Total each cohort's revenue, turn the revenue triangle into cumulative revenue per original customer and chart the curves of the first three cohorts.
• Calculate each cohort's acquisition cost, its cumulative margin per customer and the month it paid back its CAC, or "not yet".
• Split the Q1 customers by acquisition channel and compare CAC, month-3 retention and 3-month margin per dollar of CAC to find the best paid channel.