This is a teaching example using simulated data; it describes no real company.
This is a teaching example using simulated data; it describes no real company. It applies channel-entry method end to end: commissions, logistics, returns, support, fixed cost, created demand and transfers from a more profitable channel. Entry threshold concerns incremental volume and full contribution, not channel-reported revenue.
Decision and method.
Scale only if expected volume exceeds the calculated threshold after internal transfers, fixed costs and prudent future value. Compute per-order contribution using distributions of basket, returns and logistics, not one average. Estimate created demand, competitive acquisition and transfers through zone test, survey, identifiers or cohorts, and publish a range. Deduct lost margin, integration, content, stock, team and launch cost. Add only demonstrable future value from identifiable, consenting and truly new customers. Solve profitability volume under several transfer rates; restrict pilot to a zone where service and data enable learning.
Worked example.
A marketplace brings 700 orders at €31 unit contribution. 30% transfer from an internal channel at €48 contribution. 40% are new customers yielding €18 future contribution. Fixed costs are €30,000. Incremental contribution per order: 31 − 30% × 48 + 40% × 18 = €23.80. Threshold: 30,000 ÷ 23.80 = 1,261 orders. At 700: 700 × 23.80 − 30,000 = −€13,340. Refuse full opening; continue pilot only to lower costs or prove more new demand, not because apparent margin is €21,700.
Checks and limits.
Per-order contribution includes every variable channel cost; transfers use source-channel margin; future value concerns only genuinely new customers; expected volume exceeds threshold in prudent case. Order origin changes after novelty. Logistics cost varies with density and returns. Partner can alter commission, visibility or access rules.
Resources and sources.
Download channel-entry model. Related: compare channels, calculate full cost, value transfers. Sources: Yoo and Lee (2011); Balasubramanian (1998); Fisher, Gallino and Xu (2019).