Executive readout
Executive sectionTwo prospecting campaigns were buying revenue below break even after shipping and returns.
Rebuilt order by order, the picture inverted. The two prospecting campaigns the brand was preparing to scale were buying revenue below break even once shipping, returns and payment fees were charged against each order. Gross ROAS looked healthy because it stopped measuring at the sale; it never followed the order through fulfillment or the high return rate on color cosmetics. On average, the spend on those two campaigns was losing margin: in the six month before window the loss was $28,735. The brief had named platform ROAS as the metric the finance lead trusted least; the campaigns were judged against a target of 2.0 and showed 2.49.
- Case 01 of 74DTC · Beauty & personal care · $8M to $20M revenue band · the published case
- Audit teamLead auditor Victoria Miller · reviewed by Cristian Bragau, practice principal
The finding in full
Finding section- Evidence
- Contribution margin on the two top campaigns: −7% in the 2023 window, +9% in the 2024 window. The margin improvement / yr was compared like for like over two matched six month windows, reconciled to a cohort P&L built from the order, return, payment, fulfillment and cost records. Client identity redacted to protect their commercial position.
- The fix
- Rebuild both campaigns around contribution margin measured after shipping, returns and payment fees, so success is defined by profit per order rather than gross ROAS.
- Calculated from
- Ad platform daily spend, Order and return records, Payment rate card, Fulfillment rate card, Prepared cohort P&L
- Review
- Lead auditor Victoria Miller · reviewed by Cristian Bragau, practice principal
This is the “anatomy of a finding” format. See how it is produced, or score your own six areas first with the free marketing audit template.
Working paper · finding 01 · Both windows Redacted
Working papers are the calculations behind a finding. The figure traces back to a sheet like this one.
| Campaign | Contribution margin |
|---|---|
| Retargeting | +12% |
| Google Shopping | +8% |
| Prospecting camp. A & B | −7% |
| margin improvement / yr | +$134,240 |
Working paper: the rows are the before window margins, from the client’s order and return records, with payment fees, fulfillment and product cost applied from its rate cards and cost sheet, campaign names redacted and figures scaled down by one factor for anonymity, then the six month totals rounded; margins are campaign averages, not every order. The total is the annualized improvement on Prospecting A and B across both windows, twice the six month difference. Annualized: twice the six month improvement on the two campaigns, an annual run rate.
What was proposed
Plan section- 01Rebuild both campaigns around contribution margin measured after shipping, returns and payment fees, so success is defined by profit per order rather than gross ROAS.
- 02Cut or rework the audience and creative combinations that converted below break even on average, while protecting the pockets that paid back.
- 03Install a margin after returns guardrail in the media plan, so no campaign scales past the point at which each extra order starts losing money.
- 04Give finance and marketing one shared view of contribution after returns, closing the gap between the ad dashboard and the P&L.
From kickoff to the readout: 6 days.
$134k a year of improved contribution, read a year after the audit, an annual figure equal to 23× the $5,950 Audit + Sprint fee. It is read on the two campaigns, over matched windows.
Against the benchmark
Benchmark sectionThe benchmark counts one highlight figure per published case across 74 audits in 9 industries; its lower median is $58k a year and the DTC / Ecommerce median $53k. The 10 DTC / Ecommerce cases average $59k a year, from $31k to $134k, and this audit sits among them by its loss at the readout, $57,470 a year. Its $134k a year is an improvement read a year after the audit, a different measure, so it is not ranked against them. Read the benchmark.
