A campaign showing a strong ROAS at $1,000 a day is not evidence that the next $1,000 will perform the same way. The only useful scaling question is what the additional money produces after incremental costs. Black Friday makes that question urgent because promotions alter margin just as competition and shopping intent change.
Adobe reports $44.2 billion in U.S. online Cyber Week spending in 2025, with $11.8 billion on Black Friday alone 1. Shopify reports $14.6 billion in global BFCM merchant sales in the same season 2. These figures confirm substantial seasonal volume, but neither predicts the incremental return from raising your brand's campaign budget.
Average CPA hides the next order's cost
Assume a brand spends $5,000 and acquires 200 new customers, for an average CPA of $25. It increases spending to $7,000 and acquires 250 customers in total. The blended CPA becomes $28, which still looks acceptable. But the additional $2,000 bought only 50 more customers: marginal CPA is $40.
| Metric | First level | Expanded level | Increment |
|---|---|---|---|
| Advertising spend | $5,000 | $7,000 | $2,000 |
| Acquired customers | 200 | 250 | 50 |
| Blended CPA | $25 | $28 | not applicable |
| Incremental CPA | not applicable | not applicable | $40 |
These figures are an illustrative calculation, not a measured FENIX account.
If first-order pre-ad contribution is $32, the incremental acquired customer loses $8 before fixed operating expenses. Total revenue can increase while the last layer of advertising destroys contribution.
Three ways the dashboard can overstate the benefit
Attribution overlap: multiple platforms may take credit for the same orders. Do not sum attributed purchases across Meta and Google to estimate the number of unique incremental customers.
Returning-customer credit: existing demand may convert during the campaign period. Separate new customers from existing customers whenever possible and judge the discount given to buyers who might have purchased anyway.
Lag: recent campaigns may still have conversions to receive attribution, and platform reporting dates can shift. Compare consistent mature time windows rather than pausing after a few incomplete hours.
The spend-increase gate
Before raising budget, answer four questions: (1) What contribution does one additional first-order customer bring at holiday prices? (2) What is the marginal CPA in recent comparable spend increments? (3) Can inventory and fulfillment support additional orders? (4) How much of the apparent return is likely incremental rather than demand capture?
A strong response is not always 'scale'. If the best next action is repairing checkout, switching promoted SKUs or using existing customer email, the prudent decision may be to reallocate rather than increase spending.
How to measure a better answer
Use controlled spend ramps where practical, monitor orders and contribution, and exclude periods with radically different promotions when interpreting increments. In higher-spend accounts, geographically structured tests or holdouts may improve incrementality estimates, but their design and limitations need to be disclosed. In smaller accounts, admit that noisy samples limit causal conclusions.
Calculator output: allowable CPA, current blended CPA, estimated marginal CPA, and incremental contribution at the proposed budget. State whether the calculation uses measured cohorts or a user-entered scenario.
Run the numbers for your store
The values below are an example. Replace them with your own. Nothing you type leaves your browser, no sign-up is needed, and taxes are excluded.
Sources
- Adobe, 2025 actual Cyber Week results news.adobe.com
- Shopify, global 2025 BFCM merchant sales shopify.com
- Google Analytics, ecommerce measurement support.google.com
Nenad Vasic
Founder and CEO of Fenix Network Media. His work focuses on the economics behind customer acquisition, conversion, measurement and profitable growth. Calculations in this article are original; third-party figures belong to their publishers and are credited in the sources.
Calculate the cost of the next 100 orders
Use your own numbers, then bring the result to a conversation if you want a second set of eyes on the plan.
