1 · The reality: most e-commerce accounts fly blind
Across HeySquad missions in Belgian e-commerce from 2023 to 2026, the same pattern keeps coming back. The marketing manager or the Ads agency steers on 1 single indicator: the Meta ROAS. Everything else is treated as secondary. When we dig in during an audit, we discover that:
- The announced Meta ROAS credits the whole sale to the last click, so it inflates the importance of the last touchpoint
- Product margin (cost of goods, shipping, payment fees) is not built into the profitability calculation
- Fixed costs (team, tools, premises) get forgotten when reasoning about "campaign profitability"
- A customer's 12-month value and the repeat-purchase rate do not appear in any steering meeting
The consequence: accounts that look healthy collapse silently. Meanwhile, Meta ads have gotten more and more expensive. According to eMarketer, Meta overtakes Google in global advertising revenue for the first time in 2026 (243.46 billion $ vs 239.54 billion $), with an ad cost that exploded 70% over 12 months. The pressure on margins has never been this strong.
You can no longer steer your e-commerce account with a single indicator. You need the full cascade.
2 · Why ROAS on its own lies
Take 3 made-up e-commerce accounts with the same 4.5× Meta ROAS. Are they all healthy?
Account | Meta ROAS | Average basket | Product cost | Fixed costs | Customer value 12 months | Real verdict |
|---|---|---|---|---|---|---|
A · B2C beauty, customers who come back | 4.5× | 45 € | 25% | 35% | 113 € (4 purchases, 50% come back) | Healthy (profit per euro of ads ~3.1, profitable, customer value far above acquisition cost) |
B · Fast fashion B2C, thin margin | 4.5× | 35 € | 55% | 30% | 46 € (2 purchases, 30% come back) | In the red (profit per euro of ads ~1.9 that looks OK, but after 30% fixed costs, net loss on every order) |
C · Pro B2B, single purchase | 4.5× | 280 € | 40% | 20% | 280 € (1 purchase only) | Profitable but fragile (profit per euro of ads ~2.6, profitable from the first order, but no cushion if ad cost rises) |
Same ROAS, 3 completely different business realities. ROAS on its own does not tell A, B and C apart. This is exactly the trap that Ads agencies and marketing managers have been falling into for the past 2-3 years.
What separates A, B and C is the margin cascade and the real value of a customer. That is what the HeySquad calculator below measures.
3 · The full cascade: 14 indicators we look at on a mission
In an e-commerce audit at HeySquad, we trace back 14 indicators organized across 3 levels. Margin on costs (which we note MC1, MC2, MC3) is a cascade: you start from revenue and strip out costs step by step to see what actually remains. Each indicator sheds light on a specific business angle.
Level 1: per-order profitability
Indicator | What | Why it matters |
|---|---|---|
Gross margin per purchase | Basket - (product cost + shipping + payment fees) | Real gain per order, before ads and fixed costs |
Target profit per purchase (MC3) | Basket × Desired profit % | What you want to earn per order |
Target acquisition cost | Gross margin - fixed costs - desired profit | The cost to win a customer beyond which you eat into your profit |
Critical acquisition cost | Gross margin - fixed costs | The break-even threshold: no gain, no loss |
Level 2: the margin cascade (monthly)
Indicator | What | Healthy benchmark |
|---|---|---|
Expected orders (monthly) | Revenue target / Average basket | n/a |
Monthly marketing budget | Target acquisition cost × Number of orders | n/a |
Daily budget | Monthly budget / 30 | n/a |
Gross margin (MC1) | Revenue ex-VAT - (product cost + shipping + commissions) | < 55% of revenue |
Margin after ads (MC2) | Gross margin - marketing costs | > 25-35% of revenue |
Monthly net profit (MC3) | Margin after ads - fixed costs | Business target |
Level 3: the signature account-health indicators
Indicator | What | Healthy benchmark |
|---|---|---|
Target / critical ROAS | Basket / Target acquisition cost (or critical) | Compares your real measured ROAS to the theoretical target |
POAS (profit per euro of ads) | Gross margin / Ad spend | ≥ 1 profitable · < 1 loss on every sale |
Customer value over 12 months | Basket × (1 + (purchases − 1) × repeat rate) | A customer's true lifetime value (the first order always counts, later orders are weighted by the repeat rate) |
Customer value / acquisition cost | Customer value / Target acquisition cost | < 3 warning · 3-5 healthy · > 5 excellent |
Time to break even | Target acquisition cost / Gross margin per purchase × frequency | < 6 months excellent · 6-12 months healthy · > 12 months warning |
This is the cascade that tells accounts A, B and C apart from the example in §2. ROAS is the 11th indicator in the cascade, not the 1st. Steering on it alone is professional malpractice in 2026.
4 · POAS, not ROAS: the metric that makes ROAS obsolete
POAS is your profit per euro of ads: the ratio between the gross margin you generate and what you spend on advertising. ROAS looks at revenue per euro spent (flattering, because it ignores margin). POAS looks at what is actually left once you strip out the product cost and variable costs.
Formula: `POAS = Gross margin / Ad spend`
How to read it:
- POAS ≥ 1 : you are profitable. Every euro of ads brings back at least 1 euro of gross margin.
- POAS < 1 : you lose money on every sale, no matter what your ROAS suggests.
- POAS ≥ 2 : you have a real machine. You can scale up without risk.
Take account B from the §2 example again (low-margin fast fashion). 4.5× ROAS announced. 35 € basket. 55% product cost. 2% payment fees. Gross margin per purchase = 35 - (35 × 0.57) = 15 €. If the cost to win a customer via Meta = 35 / 4.5 = 7.80 €, then POAS = 15 / 7.80 = 1.92 → in reality still profitable, but the margin is fragile and the smallest ad-cost increase tips it over.
Now take an account with a 4.5× ROAS but a 35 € basket and 70% product cost (a tiny margin). Gross margin = 35 × 0.3 = 10.50 €. POAS = 10.50 / 7.80 = 1.35. The account is on the edge. If the Meta ad cost rises 70% as it did between January 2025 and January 2026, the POAS drops below 1.
This is the ROAS-only blindness effect: you steer on the wrong indicator, and you do not see the moment your account tips into a loss.
5 · The customer value that flips the decision
POAS only looks at the first order. But in recurring B2C e-commerce, annual-contract B2B or a B2B2C marketplace, a customer places several orders over 12 months. Customer value is what they bring you over their whole lifetime, not just their first order.
Simplified formula: `Customer value 12 months = Average basket × (1 + (Number of purchases − 1) × Repeat rate)`. The first order always counts, it is the acquisition you pay for; only the later orders are weighted by the repeat rate. A customer who buys only once but stays profitable therefore keeps a value equal to their basket, not a value crushed by the repeat rate.
Take account A, B2C beauty where customers come back, again. 45 € basket. 4 purchases per year on average. 50% of customers come back. Customer value = 45 × (1 + 3 × 0.5) = 113 € (the first order plus 3 later orders weighted at 50%).
If the target acquisition cost = 25 €, then customer value over acquisition cost = 113 / 25 = 4.5 → a healthy ratio.
If the same company underestimated this value and reasoned on the first order alone (a profit per euro of ads of 0.9 on the first order), it would conclude "loss on every sale" when in reality the 3rd or 4th purchase more than compensates.
The ratio between a customer's value and their acquisition cost is something no ROAS shows you. See server-side tracking which makes this measurement clean on the server side.
6 · The HeySquad e-commerce account health calculator
We put this whole cascade into a calculator. You give 10 figures (average basket, product cost, fixed costs, shipping and payment fees, monthly revenue target, purchase frequency, purchase cycle length, share of customers who come back over 12 months), and it outputs the 14 indicators from the previous section and, above all, a health score out of 100.
The score weighs 5 axes: profit per euro of ads (30 points), customer value over acquisition cost (25), net margin as a percentage of revenue (20), time to break even (15) and adherence to the target ROAS (10). Below 50, the account is in danger. Between 50 and 75, it is healthy but improvable. Above 75, it is a viable machine.
The verdict guides the action, not the other way around. Below 50, you tackle the urgent: renegotiate the supplier margin, cut the low-margin products, redo the tracking. Between 50 and 75, you optimize: get more customers to come back, work on the product cost, target audiences better. Above 75, you scale up: more Ads budget, a new channel, international.
We run this calculation on your account during an audit. See paid acquisition for account structures by maturity, and strategy for the broader business framework.
7 · How to read your score by e-commerce model
If you are recurring B2C e-commerce (beauty, food, everyday fashion)
Your key levers: high customer value (4+ purchases per year), a share of customers who come back > 40%. Your first-order profit per euro of ads can be borderline (1-1.5) if the overall customer-value-over-acquisition-cost is > 3. Focus the audit on customer value and purchase frequency.
If you are pro B2B e-commerce, single purchase or one-off contract
Your key levers: a high average basket (200-2,000 €), a high gross margin per purchase, a low customer value because purchases are one-off. Your first-order profit per euro of ads must be > 2 to absorb the low repeat rate. Focus the audit on the gross margin (MC1) and the target acquisition cost.
If you are B2B2C e-commerce (marketplace, distributor)
Your key levers: a mix of buyer profiles, a variable average basket, higher payment fees (platform commissions). Your profit per euro of ads must be > 1.5 minimum to absorb the middleman costs. Focus the audit on payment fees and fixed costs.
In every case, ROAS on its own is not enough. It is the full cascade we look at.
