E-commerce marketing mix: deciding on contribution per order.
Key distinctions
Three conditions before calculating.
Net collected basket value
The basket value must be reduced by discounts, cancellations, refunds and taxes that are not retained. Using the displayed basket value overstates the base on which the margin is actually earned.
Order-driven costs
Payment, picking, packaging, subsidised delivery, returns, fraud and customer-service contacts vary with the order. They must be allocated to the cohort that generates them.
Incremental acquisition
Average advertising cost often blends organic orders with customers already acquired. A growth decision must compare the cost and the additional orders with a no-spend counterfactual.
Method
Building a cohort contribution statement
- 01Freeze the cohort
Define acquisition date, source, offer, device, area and new-or-existing status; follow the orders until the end of the return window.
- 02Reconcile the flows
Link order, payment, discount, shipment, return, refund and service contact through a stable identifier; unreconciled differences are published.
- 03Compute the contribution
Contribution = net revenue × product margin − payment − logistics − returns − service − incremental acquisition; fixed costs are added only if they change with the decision.
- 04Compare with the threshold
Compute the maximum CAC and the minimum basket value compatible with the target contribution; test new customers, reactivation and repeat purchase separately.
The economic waterfall of 1,000 additional orders
Basket value and gross margin are adjusted for the costs that advertising dashboards often ignore: acquisition, payment, picking, returns and service.
A ROAS of 3.18 that produces a loss of €3,700
A campaign generates 1,000 incremental orders, a net basket value of €70, a product margin of 45% and €22,000 of media spend. Payment, picking, returns and service cost an additional €1,500, €7,500, €4,200 and €0 respectively.
Product margin: 70,000 × 45% = €31,500. Contribution after costs: 31,500 − 1,500 − 7,500 − 4,200 − 22,000 = −€3,700. The ROAS of 70,000 ÷ 22,000 = 3.18 does not signal that loss.
The campaign must be corrected. With unchanged unit economics, its maximum acquisition cost is €18,300, that is €18.30 per order.
Acceptance conditions
What must be true to act.
- 01
Orders are tracked until their return window closes.
- 02
The CAC used is incremental, or explicitly presented as an approximation.
- 03
Payment, logistics and service costs are allocated to the cohort.
- 04
The CAC threshold and the minimum basket value are computed before the budget is increased.
Limits
What this analysis does not prove.
- The first order underestimates the value of customers who actually buy again.
- Shared logistics costs can change in steps rather than per order.
- Ad attribution is no substitute for an incrementality test.
References
Works cited.
- Gupta, S., Lehmann, D. R. & Stuart, J. A. (2004), “Valuing Customers”, Journal of Marketing Research.
- Fader, P. S. & Hardie, B. G. S. (2009), “Probability Models for Customer-Base Analysis”, Journal of Interactive Marketing.
- Gordon, B. R. et al. (2019), “A Comparison of Approaches to Advertising Measurement”, Marketing Science.