Calculating SKU-Level Return On Ad Spend: Integrating Fulfillment Costs And Ads

Return on Ad Spend is Revenue attributed to advertising divided by advertising spend. Calculating ROAS at SKU level requires attributing revenue from ads to individual SKUs and dividing that revenue by the portion of ad spend that drove those SKU sales.

For merchants and warehouses that combine fulfillment analytics with advertising performance, SKU-level ROAS is a powerful way to see which products truly justify promotional investment after accounting for picking, packing, shipping, storage, and returns. This article shows a practical method to compute SKU-level ROAS, data inputs required, and common adjustments warehouses must make for accuracy.

Core Inputs Needed

• Attributed Revenue By SKU: Revenue assigned to ad-driven conversions—use your ad platform’s attribution model and confirm it maps to SKU-level transaction data in your order system.
• Ad Spend Assigned To SKUs: The portion of campaign spend that can be reasonably allocated to each SKU (via campaign-level budgets, keyword bids, or conversion weights).
• Fulfillment Cost Per Order/SKU: True per-unit costs including pick-pack labor, packaging, outbound shipping, storage, and average return handling.
• Return Rates And Refund Amounts: Historical return data by SKU to adjust net revenue used in the numerator.

Step-By-Step Calculation

1) Export conversion-level data from your ad platform including order IDs and conversion values for the chosen attribution window.

2) Join conversion data to your order management system (OMS) or WMS by order ID to allocate revenue to SKUs. If an order contains multiple SKUs, apportion conversion value by SKU price or by a pre-defined margin share.

3) Allocate ad spend to SKUs. Methods include direct spend mapping (campaigns that target single SKUs), proportional allocation (distributing campaign spend across SKUs based on revenue shares), or modelled attribution (using multi-touch attribution to assign fractional spend).

4) Adjust revenue for returns and refunds during the attribution period to produce net attributed revenue per SKU.

5) Compute SKU-level ROAS = Net Attributed Revenue For SKU ÷ Attributed Ad Spend For SKU. Optionally subtract fulfillment costs from revenue first to produce a contribution-based ROAS.

How To Include Fulfillment Costs Correctly

• Direct Per-Unit Costs: Include packaging, labeling, and specific materials tied to the SKU.
• Shared Order Costs: For multi-SKU orders, apportion pick-and-pack labor and outbound shipping across SKUs by weight, volume, or price share.
• Storage Allocation: Convert storage fees to per-SKU per-day costs based on cube, then apply the appropriate holding period for the sale.
• Returns And Processing: Estimate average return handling per SKU and reduce net attributed revenue accordingly.

Data Integration And Tools

A robust SKU-level ROAS calculation needs joined datasets. Common architectures include:

• Direct Join: Connect ad platform exports to OMS/WMS using order IDs in a BI tool (Looker, Power BI, Tableau).
• CDP Or Data Warehouse: Ingest ad impressions, clicks, transactions, and fulfillment events into a central warehouse (BigQuery, Snowflake) and run SQL joins for attribution.
• Attribution Platforms: Use specialized attribution tools that can output SKU-level spend shares when fed order and product data.

Practical Example

SKU A had $12,000 in ad-attributed gross revenue over 30 days and $3,000 in ad spend assigned (proportional allocation). Fulfillment costs were $2,000 (packaging, shipping, handling), returns reduced revenue by $1,000. Net attributed revenue = $11,000. SKU-level ROAS (gross) = 12,000 ÷ 3,000 = 4.0. Contribution-based ROAS = (11,000 − 2,000) ÷ 3,000 = 3.0. The contribution-based number shows true post-fulfillment efficiency.

Common Calculation Errors

• Attribution Mismatch: Using different windows or models between ad platforms and OMS leads to skewed SKU revenue.
• Ignoring Returns Delay: Returns often occur after the ad attribution window—use historical return curve adjustments.
• Over-Allocating Spend: Double-counting spend across SKUs when campaigns share keywords or audiences inflates denominators.

In short, the Return on Ad Spend metric can be computed at SKU level by joining ad-attributed revenue to per-SKU ad spend and adjusting for fulfillment costs and returns. Proper data integration, careful spend allocation, and testing different attribution windows produce ROAS that reflects true product profitability and informs smarter ad investment decisions.