Why Contribution Margin Should Drive Budget Decisions More Than ROAS

· Articles

Why Contribution Margin Should Drive Budget Decisions More Than ROAS

Two campaigns spend $10k each. Most buyers see a 4.0 ROAS on one and a 2.8 on the other, then move budget into the 4.0. In the example below, that decision costs the brand $6.8k.

ROAS shows how much revenue came back for every dollar spent. It says nothing about how much of that revenue the business keeps, which is the number budget decisions should be built on. We manage paid media across Google, YouTube and Meta for 8-figure ecommerce brands, where this gap decides whether Q4 makes money. This post walks through the math, shows where a single ROAS target breaks, and lays out how to rebuild your targets from contribution margin before BFCM.

ROAS Tracks Revenue While Budget Decisions Run on Profit

ROAS divides attributed revenue by ad spend. The formula ignores cost of goods, pick and pack, shipping, payment processing, discounts and returns. Depending on the category, those costs can take anywhere from a third of each order to well over half.

Contribution margin is the share of revenue left after those variable costs. Subtract ad spend from that and you get contribution after ad spend, the money that pays for salaries, software and rent before anything reaches profit. Two campaigns with identical ROAS can produce very different contribution depending on what they sell and at what price.

The Math Behind the Two Campaigns

Campaign A sells a product line with a 25% margin after COGS, shipping and fees. At a 4.0 ROAS, $10k of spend brings in $40k of revenue. That leaves $10k of contribution before ads, which drops to $0 once the $10k of spend comes out.

Campaign B sells a line with a 60% margin. At a 2.8 ROAS, the same spend brings in $28k of revenue. Contribution comes to $16.8k before ads and $6.8k after.

Break-even ROAS is the fastest way to see the gap. It equals 1 divided by your margin before ad spend. Campaign A breaks even at exactly 4.0, while Campaign B breaks even at 1.67. A single account-wide ROAS target treats these campaigns as the same business when their economics sit far apart.

Discounts Push ROAS and Contribution Margin in Opposite Directions

This gap widens during BFCM. Take a $100 product with $45 of variable costs per order, which gives a 55% margin and a break-even ROAS of 1.82. At 30% off, the price drops to $70 while costs stay at $45. Margin falls to 35.7%, so break-even ROAS jumps to 2.8.

Conversion rate usually climbs during a sale, so ROAS in Google Ads and Meta Ads Manager often looks stronger than it did in October even as each order contributes far less. If you hold the same ROAS target through Black Friday weekend, you can lose money on every order while the dashboard reports your best week of the year. Rebuild promo-period targets from the discounted margin before the sale goes live.

For BFCM on Google, pair the rebuilt targets with a seasonality adjustment. The adjustment tells Smart Bidding to expect a short conversion rate spike, so bidding responds to the sale from day 1 instead of catching up after it ends. Google built the tool for events lasting a few days, which fits Black Friday weekend.

Average ROAS Hides What the Next Dollar Is Doing

When you scale a campaign, the last $1k you add rarely performs like the first $1k. An account averaging 3.0 ROAS can carry a marginal return well below break-even, with the average masking it for weeks.

This is why CPMr belongs in the same report as contribution margin. CPMr (Cost Per Mille Reached) is CPM multiplied by frequency, the true cost of reaching 1,000 unique people. Phil Kiel, our Director of Paid Social, popularized it as a diagnostic inside Taikun.

When CPMr climbs week over week with frequency rising and CPM flat, your extra budget is cycling through the same audience instead of finding new people. Attributed ROAS often holds up longer than it should in that situation, because retargeting and warm-audience ads keep claiming credit for purchases already in motion. Contribution after ad spend, measured against total paid spend at the business level, tends to show the slowdown before platform ROAS does.

Neither view settles attribution by itself, because blended contribution can’t tell you which ad caused a sale. Platform ROAS can’t tell you whether that sale made the business money, which is why both belong in the same weekly report.

YouTube Needs a Business-Level View

YouTube makes the case for contribution after ad spend even stronger. Demand Gen and video campaigns reach viewers who often convert later through branded search, direct traffic or a Meta retargeting ad. Each of those channels claims the sale in its own reporting, while YouTube’s attributed ROAS looks weak. Judge YouTube spend by what happens to total contribution after ad spend when you scale it or pull it back, ideally through a geo holdout. Cutting it on platform ROAS alone can remove the spend feeding your best-looking campaigns.

Where ROAS Still Belongs

ROAS is fast and available at ad level. Both major platforms can bid against it directly, through tROAS in Google Ads and the minimum ROAS bid strategy in Meta. Contribution margin often lags, since returns land weeks later and COGS figures only update when someone updates them. Use ROAS for in-platform decisions within a campaign, with a target derived from contribution margin instead of a round number like 3.0.

The target carries the most weight on Google, where tROAS tells Smart Bidding what to chase across Shopping, Performance Max and Search. A margin-blind target there turns into margin-blind bidding in every auction. Split Shopping and Performance Max campaigns along the same margin bands you use for targets, then set tROAS for each.

Accepting negative first-order contribution can make sense when repeat purchase rates support it. Make that a deliberate decision with a payback period attached, so it doesn’t surface in the P&L 3 months later.

How to Set It Up This Week

  1. Pull variable cost per product or collection, including COGS, fulfillment, shipping, payment fees and expected return rate. Calculate contribution margin before ad spend as a % of revenue.
  2. Convert each margin into break-even ROAS by dividing 1 by the margin. Add the profit buffer you want on top to set a working target.
  3. Map each campaign to the product lines it mostly sells. Replace the single account-wide target with line-level targets, set as tROAS in Google Ads and minimum ROAS in Meta.
  4. Report contribution after ad spend from Shopify or your profit tool against total paid spend across Google, YouTube and Meta every week, next to CPMr. That view shows the point where scaling stops adding profit.
  5. Rebuild every target from the discounted price before any promotion launches. Schedule a Google Ads seasonality adjustment for the sale window at the same time.

Start with your 3 highest-spend campaigns and work out their break-even ROAS this week. At least 1 of them is often judged against the wrong number. Finding that out after Black Friday costs far more than finding it now. For the reach side of the picture, read our breakdown of how to calculate CPMr.

Back to all articles