A marketplace brings demand, but it also takes control away.
Key distinctions
Three conditions before calculating.
New demand
Distinguish genuinely new customers, customers who would have bought elsewhere and volume merely brought forward. Use a control area, an identifier cohort or a calibrated survey. A new marketplace account does not prove that the customer is new to the brand.
Full take rate
The visible commission is only one layer. Add payment fees, storage, picking, returns, penalties, internal advertising, discounts and compliance cost. The effective rate must be computed on the net revenue retained.
Value of control
Price, presentation, customer data, service and future access have an option value. High dependency reduces bargaining power and makes withdrawal expensive. That exposure must appear separately from the annual account.
Method
Building a complete marketplace business case
- 01Measure the revenue retained
Start from the sales collected, subtract VAT, cancellations, returns, funded promotions and refunds; reconcile order, payment and stock to avoid counting revenue that never becomes income.
- 02Compute the cost to serve
Add product cost, logistics, commission, internal media, content, support, fraud and operational overhead; segment by category, basket size and fulfilment mode.
- 03Estimate incrementality
Assess the share of new orders and the margin sacrificed on the source channels; add the future value of new customers only if a credible mechanism allows them to be retained or recognised.
- 04Value the risks
Run scenarios for a commission increase, a promotional obligation, delisting and price conflict; set a dependency ceiling, a minimum margin and contractual exit conditions.
The economic waterfall of a marketplace
From the displayed revenue to the genuinely new contribution, every deduction and transfer is made visible.
€900k of marketplace sales produce €29.8k of incremental contribution
A brand generates €900k in sales on a marketplace. After returns, revenue is €820k. Product cost is €410k, commissions and fulfilment €205k, advertising €72k and service €44k. Forty-five percent of the orders are genuinely new.
Own contribution: 820 − 410 − 205 − 72 − 44 = €89k. The transferred sales would have generated €74k of contribution on the direct site. The future value of the new customers is estimated at €28k, but only 53% is capturable, giving €14.8k. Incremental contribution: 89 − 74 + 14.8 = €29.8k, or 3.3% of the displayed sales.
The channel stays open as a capped acquisition tool. Extension requires an incremental contribution above 5%, internal advertising below 9% of revenue and no erosion of the direct price.
Acceptance conditions
What must be true to act.
- 01
Reconciled net revenue replaces displayed sales.
- 02
Commission, fulfilment, media, returns and service are included.
- 03
The genuinely new share and the cannibalisation are estimated.
- 04
A dependency ceiling and exit conditions are documented.
Limits
What this analysis does not prove.
- Incomplete customer identification limits the measurement of transfers between channels.
- The platform's algorithms and rules can change without notice.
- The visibility gained can produce brand effects that are hard to isolate.
References
Works cited.
- Hagiu, A. & Wright, J. (2015), “Multi-Sided Platforms”, International Journal of Industrial Organization.
- Ryan, J. K., Sun, D. & Zhao, X. (2012), “Competition and Coordination in Online Marketplaces”, Production and Operations Management.
- Neslin, S. A. et al. (2006), “Challenges and Opportunities in Multichannel Customer Management”, Journal of Service Research.