I had a call last month with a DTC skincare brand doing about £4M in revenue. Their marketing manager was buzzing.
"Our ROAS is 4.2x on Meta. We've never had it this good."
I pulled up their P&L.
They were losing money.
Not a little bit. They were burning through £30K a month in cash and had about four months of runway left. Their "amazing" ROAS was a mirage.
This happens more often than you'd think. And I'd argue it's one of the biggest reasons brands stall between 7 and 8 figures.
The problem isn't that ROAS is useless. It's that ROAS doesn't tell you what you think it tells you.
So let me walk you through the framework we use with every brand we work with at Elevate. The one that actually tells you whether your marketing is making money or quietly bleeding you dry.
Feel free to jump ahead to the section most relevant to you:
- Why ROAS Breaks Down at Scale
- The Contribution Margin Framework (CM1, CM2, CM3)
- The Maths: Two Brands, Same ROAS, Wildly Different Outcomes
- Channel-Level Contribution Margin Breakdown
- What Good Looks Like (2026 Benchmarks)
- How to Implement This Framework This Week
Why ROAS Breaks Down at Scale
ROAS (Return on Ad Spend) measures one thing: how much revenue you generated per pound spent on ads.
That's it. Revenue. Not profit. Not margin. Not cash flow.
At lower spend levels, this works fine. If your margins are healthy and your fulfilment costs are predictable, a 4x ROAS probably means you're making money.
But as you scale, everything changes...
- Your CPMs rise as you push into colder audiences
- Discount-heavy campaigns inflate revenue but compress margins
- Shipping costs increase (especially on heavier or bulkier products)
- Return rates creep up on new customer segments
- Payment processing fees stack up
A brand running at 4x ROAS with 70% gross margin and low fulfilment costs is in a completely different position to a brand at 4x ROAS with 50% gross margin and £8 average shipping cost.
Same ROAS. Wildly different outcomes.
According to research from Saras Analytics, chasing ROAS-led campaigns often leads to discount-heavy strategies that inflate top-line numbers while destroying unit economics on the backend. The brands that figure this out early are the ones that make it past the 8-figure mark.
Enter Contribution Margin: The Metric That Actually Matters
Contribution margin tells you how much money each sale actually contributes to covering your fixed costs and generating profit after all variable costs are accounted for.
Not revenue. Actual money in your pocket.
The framework breaks down into three levels, and each one strips away another layer of cost:
CM1 (Gross Margin)
Revenue minus your landed cost of goods sold (COGS).
This is the starting point. What does it actually cost you to make or buy the product and get it to your warehouse?
CM1 = Revenue - Landed COGS
CM2 (Post-Fulfilment Margin)
CM1 minus all the variable costs of getting the product to the customer.
This includes shipping, packaging, payment processing fees (Stripe, Shopify Payments), marketplace fees, and returns/refunds.
CM2 = CM1 - Shipping - Packaging - Payment Fees - Returns
CM3 (Post-Marketing Margin)
CM2 minus your variable marketing spend.
This is the big one. CM3 tells you how much each pound of revenue actually contributes to your business after you've paid for the product, shipped it, and paid to acquire the customer.
CM3 = CM2 - Variable Ad Spend
CM3 is where the truth lives.
If your CM3 is negative, you are literally paying to give products away. Your ROAS could be 10x and you'd still be losing money if your margins and fulfilment costs are eating everything.
As Wayflyer puts it in their 2026 guide, a healthy CM3 of 20% or above is what separates sustainable ecommerce brands from the ones quietly running out of cash.
Let's Do the Maths. Two Brands. Same ROAS. Totally Different Reality.
This is where it gets real. Let me walk you through two brands. Both selling a product at £50. Both running at 3.5x ROAS on Meta.
Brand A: The "Good ROAS" Trap
- Revenue per unit: £50
- COGS: £20 (CM1 = £30, or 60%)
- Shipping + packaging: £7
- Payment processing (2.9%): £1.45
- Returns (15% return rate): -£7.50 averaged across orders
- CM2 = £30 - £7 - £1.45 - £7.50 = £14.05
- Ad spend per acquisition (3.5x ROAS): £14.29
- CM3 = £14.05 - £14.29 = -£0.24
Negative. They're losing 24p on every single new customer acquired through paid. At 500 new customer orders a month from paid channels, that's £120 per month in negative contribution... before they've paid rent, salaries, software, or anything else.
Brand B: Same ROAS, Different Universe
- Revenue per unit: £50
- COGS: £12 (CM1 = £38, or 76%)
- Shipping + packaging: £3.50
- Payment processing (2.9%): £1.45
- Returns (5% return rate): -£2.50
- CM2 = £38 - £3.50 - £1.45 - £2.50 = £30.55
- Ad spend per acquisition (3.5x ROAS): £14.29
- CM3 = £30.55 - £14.29 = £16.26
Brand B is making over £16 on every new customer from paid channels. That's a 32.5% contribution margin.
Same ROAS. Same revenue. One is dying. One is thriving.
This is why ROAS alone is dangerous.
The Channel-Level Breakdown (Where It Gets Really Powerful)
Here's where this framework changes how you think about your entire marketing mix.
Most brands look at blended ROAS or blended CAC across all channels. That's like knowing your car uses "some petrol" without knowing how much each journey costs.
You need CM3 by channel.
When we onboard a new client at Elevate, one of the first things we do is map contribution margin across every acquisition channel. And almost every time, the picture looks completely different from what the brand expected.
Here's a simplified example of what we typically find:
| Channel | Revenue | CM2 | Ad Spend | CM3 | CM3 % |
|---|---|---|---|---|---|
| Meta (Prospecting) | £100K | £45K | £28K | £17K | 17% |
| Meta (Retargeting) | £40K | £18K | £4K | £14K | 35% |
| Google Brand | £60K | £27K | £3K | £24K | 40% |
| Google Shopping | £80K | £36K | £22K | £14K | 17.5% |
| Email/SMS | £50K | £22.5K | £1.5K | £21K | 42% |
| Organic | £30K | £13.5K | £0 | £13.5K | 45% |
Look at that. Google Brand and Email have the highest CM3 percentages. Organic is printing money. Meta prospecting is the lowest margin channel... but it's feeding every other channel on that list.
This is the nuance that gets lost when you only look at ROAS.
If you cut Meta prospecting because the ROAS is "only" 3.5x, you'll watch Google Brand searches drop, email revenue fall, and organic traffic decline within 4-6 weeks. I've seen it happen dozens of times.
The smart play isn't to optimise each channel in isolation. It's to understand how they work together and what each one actually contributes to the bottom line. (If you want to understand the full picture of how customers move between channels, our customer journey mapping guide goes deep on this.)
This is exactly why we built our approach at Elevate around the full picture. CRO, organic, paid, and email working together. When you only optimise one channel, you miss the interconnections that actually drive profitable growth.
What Good Looks Like in 2026
Based on the brands we work with and wider industry data, here's what healthy contribution margins look like right now:
CM3 Targets by Stage
- Pre-scale (under £2M revenue): 15-20% CM3 is acceptable. You're investing in growth and that's fine.
- Scaling (£2M-£10M): 20-25% CM3. This is where discipline separates survivors from casualties.
- Scaled (£10M+): 25%+ CM3. If you're below 20% at this stage, something is structurally broken.
LTV:CAC Ratio
The standard benchmark is 3:1 or better. But here's the thing... that ratio is meaningless if you don't know your actual contribution margin. An LTV of £150 against a CAC of £50 looks brilliant until you realise your CM2 on each order is only £15 and it takes eight repeat purchases to hit that LTV number.
I talked about a related version of this problem in our piece on DTC marketing strategies. The brands that win aren't just acquiring customers cheaply. They're acquiring customers profitably.
CAC Payback Period
Healthy brands in 2026 are recovering their acquisition cost within 3-6 months. If it's taking 12+ months, your cash flow will strangle you before the LTV materialises. This is especially critical for VC-backed brands burning through runway.
A recent breakdown from Eightx found that the average ecommerce CAC has increased 40-60% since 2021, which makes getting your contribution margins right more important than ever. You simply can't afford to be sloppy with your numbers at today's acquisition costs.
How to Implement This Framework This Week
Right. Enough theory. Here's exactly what to do.
Day 1-2: Calculate Your CM1
Pull your COGS data. Not the rough estimate you've been using... the actual landed cost including manufacturing, raw materials, import duties, and warehouse receiving costs.
If you're a DTC brand on Shopify, this should live in your product records. If it doesn't, that's your first problem to fix.
Day 3: Map Your CM2
Add up every variable cost between your warehouse and the customer's door. Shipping, packaging, payment processing, marketplace fees, and returns. Calculate this as a percentage of revenue and as an absolute number per order.
The return rate one catches people out. If you're in fashion and running a 20% return rate, that's a massive hidden cost that never shows up in your ROAS calculation.
Day 4-5: Build Your CM3 by Channel
Take your CM2 and subtract your variable ad spend, broken down by channel. Use your ad platform data alongside your analytics to attribute spend correctly.
If you're on Shopify, tools like Lifetimely, Triple Whale, or Polar Analytics can help. But honestly? A well-structured Google Sheet that you update weekly is better than a fancy dashboard you never check.
Day 6-7: Make Decisions
Look at the data. Ask yourself:
- Which channels have the highest CM3 percentage?
- Which channels are negative or near-zero?
- Where am I spending the most on the lowest-margin channel?
- What would happen if I shifted 20% of my lowest-margin channel spend to my highest-margin channel?
Then test it. Run a two-week experiment. Shift budget. Measure the impact on blended CM3, not ROAS.
The Bigger Picture
Here's the thing that separates the brands that make it from the ones that don't.
The brands that stall at 7 figures are optimising for ROAS and platform metrics. They're making decisions based on what Meta's dashboard tells them, in isolation.
The brands that push through to 8 and 9 figures? They're optimising for contribution margin across the entire business. They understand that a £1 spent on Meta prospecting might show a 2.5x ROAS in-platform but generates £3 in organic search revenue, £2 in email revenue, and £1.50 in direct traffic over the following 90 days.
They think in systems, not silos.
If you're sitting between £3M and £30M and you don't have contribution margin by channel calculated and updated at least monthly, I'd genuinely suggest making that your number one priority this week. Not a new ad campaign. Not a website redesign. Not a new creative strategy.
Get your numbers right first. Everything else gets easier after that.
Key Takeaways
- ROAS measures revenue, not profit. Two brands with identical ROAS can have wildly different profitability based on their cost structure.
- CM3 (contribution margin after marketing) is the metric that matters. It tells you how much each sale actually puts in your pocket.
- Calculate CM3 by channel, not just blended. You need to know which channels are profitable and which are feeding other channels.
- Don't kill your top-of-funnel because the ROAS looks low. Meta prospecting might have the worst CM3 but it's generating demand that shows up everywhere else.
- Aim for 20%+ CM3 if you're scaling past £2M. Below that, you're probably not building a sustainable business at today's acquisition costs.
If you're looking at your numbers and feeling slightly sick, that's not a bad thing. It means you're about to get a lot smarter about where your money goes. And if you want a second pair of eyes, we offer a free 15-minute Loom audit where we walk through your website and marketing setup with specific, practical ways to tighten up your unit economics and reduce CPA without destroying your backend margins.
