The question
Suppose two product groups report the same ACOS. Should they receive the same advertising target?
The exercise uses one assumed 30-day reporting window in US dollars. It does not compare a period before an intervention with a period after one. The purpose is to see what a blended advertising ratio leaves out.
The assumed inputs
Both product groups use the same reporting window, currency and attribution definition. For the calculation only, assume the sales bases are consistent with the contribution percentages. A real review must reconcile those definitions first.
Product group A
- Ad spend: $6,000
- Ad-attributed sales: $20,000
- Contribution before advertising: 35% of the assumed sales value
Product group B
- Ad spend: $6,000
- Ad-attributed sales: $20,000
- Contribution before advertising: 25% of the assumed sales value
Account-wide assumption
- Combined ad spend: $12,000
- Combined ad-attributed sales: $40,000
- Total account sales: $75,000
The contribution assumptions allow for relevant product costs, marketplace fees, fulfillment and expected returns or promotional costs. They exclude advertising, fixed overhead and taxes. The percentages are deliberately simplified; real costs can vary by product, order and reporting period.
Work through the ratios
ACOS is ad spend divided by ad-attributed sales.
- Product group A: $6,000 ÷ $20,000 = 30.0% ACOS
- Product group B: $6,000 ÷ $20,000 = 30.0% ACOS
- Combined: $12,000 ÷ $40,000 = 30.0% ACOS
TACOS is ad spend divided by total sales in the chosen account scope.
- Account-wide: $12,000 ÷ $75,000 = 16.0% TACOS
These ratios describe the assumed inputs. Neither establishes profitability or the sales that would disappear if advertising stopped.
Add the product economics
Apply the assumed contribution percentages to ad-attributed sales as a simplified planning calculation, then subtract the assumed ad spend.
- Product group A: $20,000 × 35% = $7,000 before advertising; $7,000 − $6,000 = $1,000 remaining in this calculation
- Product group B: $20,000 × 25% = $5,000 before advertising; $5,000 − $6,000 = −$1,000 remaining in this calculation
The same 30% ACOS sits five percentage points below A's assumed pre-ad contribution rate and five points above B's. A single blended target hides that distinction.
The remaining amounts are not measured profit or incremental advertising contribution. They apply simplified cost assumptions to attributed revenue. Attribution assigns credit; it does not establish whether advertising caused a sale. Fixed overhead and taxes are also excluded.
Turn the calculation into a decision
The model raises a question about group B. It does not establish that every B campaign should be cut or that A should automatically receive more budget.
- Validate the product-level costs and reconcile discounts, returns, sales definitions and reporting windows.
- Separate campaign purposes, such as branded demand capture, non-branded acquisition and a deliberately bounded launch test.
- Check search-term relevance, conversion, price, offer quality and available stock before deciding which lever to test.
- Choose one controlled change with an explicit spending boundary and review window. Record what would support continuing, changing or stopping it.
For example, a mature acquisition campaign with reliably weak economics raises a different decision from a small launch test approved to answer a specific learning question. The reason for spending must be stated rather than hidden in the blended ratio.
What could change the answer?
A different product mix, understated return costs, a temporary discount or changes in reporting attribution could change the interpretation. A campaign could also capture sales that would have happened anyway.
Evidence about repeat purchasing may matter to a longer-term acquisition decision, but it would require a defined measurement window and actual customer data. It should not be inserted as an assumed benefit to rescue an otherwise weak calculation.
This exercise cannot establish the right budget, incremental sales, organic ranking effects or business profit. Its useful output is a more specific question about product economics and campaign purpose.