The ROAS Myth: Why High ROAS Is Not a Business Goal
Softies broke ROAS on purpose in a 61-day BFCM window — and contribution-margin dollars and new customers still rose.
Joel Brda, Founder & CEO of Human. Last updated: September 18, 2026.
A high ROAS can look like a win and still be underinvestment. Efficiency is not the same as growth. When the ratio becomes the goal, brands protect a trophy number while contribution-margin dollars and new customers stall.
If your agency or in-house team is “winning” on ROAS while new-customer volume stalls, this is the trade you are already paying for — you just are not seeing it on the dashboard.
This page covers Growth ROAS vs Traditional ROAS, the Softies “we broke our ROAS on purpose” trade from a 61-day BFCM window, and a teaching table that holds contribution margin roughly constant while customer counts diverge. For what contribution margin (CM3) is and how to calculate it, read Ecommerce Contribution Margin: The Number That Tells You If Yesterday Made Money.
Is a high ROAS actually a problem?¶
Yes — when ROAS becomes the goal instead of a guardrail. A ROAS that’s too high usually means untapped growth.
You can hit a ROAS target and still shrink your profit. Platforms reward efficiency. Boards like a clean dashboard. Meanwhile the customer count stops growing because nobody will spend into demand that still clears a profitable floor. The ratio looks healthy. The business does not scale.
Treat ROAS as a lever with a floor, not a trophy to maximize. For the cost stack behind that floor, use the contribution margin guide. For whether yesterday itself made money, read How to Know If Your Ecommerce Store Made Money Yesterday.
What is Growth ROAS vs Traditional ROAS?¶
Traditional ROAS maximizes the ratio. Growth ROAS scales spend to maximize profitable customer acquisition while unit economics still work — accepting that ROAS falls as volume rises.
| Traditional ROAS | Growth ROAS | |
|---|---|---|
| Goal | Highest ratio | Max profitable new customers + contribution-margin dollars |
| What “good” looks like | 8x on a capped budget | More customers at a still-profitable ROAS floor |
| Failure mode | Plateau; flat new customers | Scaling past the CM$ floor (then you pull) |
Know your floor before you push. That floor comes from contribution margin, not from last month’s “good ROAS” habit. Start with the cost stack in the CM guide.
Same contribution margin, more customers — what does that prove?¶
You can hold contribution margin roughly constant (~$76K) and still leave 1,667 customers on the table by protecting an 8x ROAS.
This is an example adapted from the book High ROAS is Bad For Your Ecommerce Business (Joel Brda, 2025).
| Scenario | ROAS | Ad Spend | Revenue | Contribution Margin | Customers |
|---|---|---|---|---|---|
| High ROAS | 8x | $20,000 | $160,000 | $76,000 | 2,133 |
| Growth Play | 5x | $38,000 | $190,000 | $76,000 | 2,533 |
| Scale Play | 3x | $95,000 | $285,000 | $76,000 | 3,800 |
Same profit. Up to 1,667 more customers. The Traditional play “wins” the ratio. The Scale play grows net new customer count massively. That is the mindset shift of growth-focused ROAS.
What is the opportunity cost of efficiency?¶
Playing it safe on ROAS costs you the customers and contribution-margin dollars you never bought. Underspending can be riskier than spending into a still-profitable floor.
Every protected 8x that leaves demand on the table is a quiet tax on next quarter’s customer count. Efficiency feels responsible. Opportunity cost does not show up as a red KPI — it shows up as flat new customers and a team that thinks the account is “healthy.” Growth ROAS makes that trade visible on purpose.
What happened when Softies broke their ROAS on purpose?¶
In a 61-day BFCM window (Nov–Dec, year-over-year), Softies’ ROAS moved 7.8x → 3.6x (−55%). Ad spend scaled 10.4x (Meta 31x). Contribution-margin dollars +102%, new customers +181%, gross revenue +191%. Same brand. Same products. One decision changed.
We broke our ROAS on purpose.¶
Softies is a loungewear brand. Human ran a Growth ROAS decision through BFCM: scale spend while unit economics still worked, and stop treating a high ROAS as the win condition.
| Metric | Prior | Current | Change |
|---|---|---|---|
| ROAS | 7.8x | 3.6x | −55% |
| Ad spend | $92K | $959K | 10.4x ($46K/mo → $479K/mo; Meta 31x) |
| CM rate (% of net revenue) | 65.0% | 45.4% | −19.6pp |
| Gross revenue | $1.44M | $4.19M | +191% |
| Contribution margin $ | $818K | $1.65M | +102% |
| New customers | 6,605 | 18,554 | +181% |
CM rate fell. Contribution-margin dollars rose. That is the point: CM3 is a dollar number, not a rate trophy and not a ROAS trophy. Why dollars ≠ rate is covered in the contribution margin guide.
Where the spend scaled¶
| Channel | Prior | Current | Change |
|---|---|---|---|
| Meta | $23K | $708K | 31x |
| $69K | $251K | 3.6x | |
| Total ads | $92K | $959K | 10.4x |
Meta was the scale lever.
If Softies had protected the prior ROAS trophy, they would have bought far less volume. Rate compressed and the ROAS fell about half; dollar contribution margin, revenue, and new customers all rose. That is what “high ROAS is bad” means in operator language — not that ROAS is useless, but that maximizing it can starve the business.
See the full Softies case study.
What should I look at instead of chasing a higher ROAS?¶
Look at contribution-margin dollars (Human’s CM3), new customers, and whether yesterday made money — not a prettier ROAS.
- Contribution margin dollars (CM3) — what’s left of net revenue after COGS, shipping and fulfillment, merchant fees, and marketing spend. Formula and ladder: Ecommerce Contribution Margin. Manual math: Ecommerce Contribution Margin Calculator.
- New customers — the key growth number traditional ROAS quietly starves.
- Did yesterday make money? — daily contribution margin, not a prettier ROAS: How to Know If Your Ecommerce Store Made Money Yesterday.
For marketing efficiency ratios, see The Ecommerce Guide to Marketing Efficiency Ratio (MER). Do not substitute MER for contribution-margin dollars when you are deciding whether to push spend.
How does Human run this decision on an account?¶
Paid, Creative, GEO, SEO, Design, Development, and CRO sit on the same contribution-margin scoreboard. Reviews open on CM$, new customers, and revenue. They close on push, hold, or pull.
Goal bands run in both directions. A ROAS that’s too high usually means untapped growth. A soft month is a call on the 8th, not a surprise on the 30th. Profit Compass is the tool Human runs inside every engagement — it surfaces contribution margin and daily contribution margin so the software and the team share one scoreboard, not a bolt-on dashboard.
When a customer review begins, the question is not “did we hit ROAS?” It is whether contribution-margin dollars and new customers moved in the right direction inside a still-profitable floor. If ROAS is climbing while customer count stalls, Human treats that as a signal to push — not a reason to celebrate. If spend is past the floor, the call is pull. Softies’ BFCM window is the public proof of that push: spend scaled, ROAS fell, and contribution-margin dollars and new customers rose.
This is how Human works with ecommerce brands that want profitable scale, not just higher spend. Softies’ BFCM results on this page are the clearest example of that trade.
FAQ¶
Is a high ROAS bad? It can be. When ROAS is the goal, brands under-buy profitable growth. Growth ROAS treats the ratio as a floor; Softies’ 61-day BFCM trade shows ROAS down and contribution-margin dollars and new customers up.
What is Growth ROAS vs Traditional ROAS? Traditional ROAS maximizes the ratio. Growth ROAS maximizes profitable new customers and contribution-margin dollars while unit economics still work, accepting that ROAS falls as volume rises.
Can lowering ROAS increase profit? Yes, when contribution-margin dollars and customer count grow inside a profitable floor. Softies’ 61-day table is the named example on this page.
What are Softies’ numbers? In a 61-day BFCM window (Nov–Dec, year-over-year), Softies’ ROAS moved 7.8x → 3.6x (−55%). Contribution-margin dollars +102%, new customers +181%, gross revenue +191%, with ad spend up 10.4x.
Is contribution margin the same as ROAS? No. ROAS is revenue / ad spend. Contribution margin (CM3) is what’s left of net revenue after COGS, shipping, fees, and ads. Read the CM guide.
What should I track instead? CM3 dollars, new customers, and daily “did yesterday make money.” Links above.
Do I need Human the agency to see this? No. The next step on this page is a free Profit Compass audit with Human — we connect your accounts and show the numbers. If you only want the software, you can request Early Access at profitcompass.io.
See Your Contribution Margin This Week¶
Every new client engagement at Human opens with a Profit Compass audit. We connect Meta, Google, Shopify, and GA4. Profit Compass pulls revenue and customer counts (counts only, no personal data). If you keep COGS in Shopify, that comes through too. Usual assumptions: merchant fees and shipping & handling. Human does the work. You’ll see whether a “great” ROAS is leaving contribution margin and customers on the table — before you spend a dollar with us. If the numbers say your current strategy is working, we’ll tell you that too.
Start With a Free Audit / See Your Contribution Margin This Week Grow with Human / Book a free consultation
Prefer the software only? Request Early Access at profitcompass.io.
Author¶
Joel Brda is Founder & CEO of Human and co-builder of Profit Compass with Scott Williams. He is the author of High ROAS is Bad For Your Ecommerce Business (2025).
