The most expensive problem I see isn’t creative quality or media buying. It’s brands making scaling decisions off a P&L that can’t tell them whether they can afford a new customer.
Take a brand doing $120M in revenue on $20M of marketing spend. MER of 6.0. On paper, one of the healthiest businesses in DTC.
Split it by customer type and the aMER is 1.3. nCAC is $182. Repeat orders are 78% of revenue and the new customers are flat. That business isn’t growing. It’s harvesting. The 6.0 is a rearview mirror pointed at customers they bought two years ago.
I built a model that fixes this. It’s free & linked at the bottom.
The blended P&L hides some key issues
A standard P&L mixes new and returning customers into one revenue line.
Fine for your accountant. Useless for deciding whether to spend more tomorrow.
Returning customers cost almost nothing to convert. Email, SMS, and subscriptions do that work for a fraction of the cost. Average them in with new customers and the cheap revenue subsidizes the expensive revenue until the whole thing reads as profitable.
You need the full P&L to know your profit but you need a new customer P&L to understand the current state of acquisition and if it’s profitable.
It’s the difference between profit today and future profitability.
Meta does the same thing to you. Left alone it harvests the cheapest conversions available, and the cheapest conversions are people who were going to buy anyway.
Everything below is first-time orders only.
Sections 1-4: Get to contribution margin
Four lines. One example brand runs through the whole post.
Gross revenue, first-time orders only. $200k of new customer revenue out of $500k total. 40% of the business is new, 2,000 orders, $100 AOV. Under 30% new and you’re living off the base.
Net revenue. Deduct discounts and returns. 30% discount rate, 12% return rate, leaves $116k.
Gross margin. Deduct COGS of $42,400. Leaves $73,600, a 63% margin on net revenue. If this line is thin, stop reading. No media buying efficiency saves a product that doesn’t make money.
Contribution margin. Deduct $10k shipping, $3,180 in payment fees, $8k fulfillment. $52,420.
That’s per-order profit before you spend a dollar acquiring anyone. No email, no SMS, no retention in it.
Section 5: Your ad spend is not your acquisition cost
Most brands take that $52,420, subtract Meta spend, and stop.
Ad spend was $50k. Acquisition CM reads +$2,420. Thin, positive, keep scaling.
Here’s the same month fully stacked:
Meta ad spend: $50,000
UGC and creative production: $7,500
Whitelisting and creator licensing: $2,000
Agency fee at 10% of spend: $5,000
Affiliate and influencer commissions: $1,500
Marketing software and tools: $1,000
$67,000. A 1.34x multiplier on reported spend. The same month now reads -$14,580.
Per customer: reported CPA of $25, fully loaded nCAC of $33.50, contribution margin of $26.21. Every new customer costs $7.29 before retention does anything.
None of those costs should be cut. Creators, licensing, media buying, and attribution are worth paying for, and brands that skip them usually have worse economics. They’re acquisition costs and they belong in the number you decide with.
Assume true CAC runs 20-30% above platform CPA until you’ve stacked it yourself.
Sections 7-8: Two caps, and which one binds
You need two limits, and they do different jobs.
Cap 1 is a minimum blended MER. It protects cash flow (or caps burn). Derive it from your P&L: 50% gross margins minus 10% OPEX means marketing can be at most 40% of revenue, which is a blended MER floor of 2.5. Our example brand is at 8.3, so cash flow is not the constraint.
Cap 2 is a maximum fully-loaded nCAC. It protects unit economics. Project cumulative profit per customer, pick a payback month you can justify, and the profit at that month is your cap. Ours is $35.
The model reads both every month and returns one signal: room to scale, or CPA constrained. Whichever cap is tighter is your actual constraint.
Section 9: What retention owes you
Two key numbers:
Cohort breakeven is per customer: -$7.29. That’s what retention has to earn back from each buyer.
Cash breakeven is the running total: -$14,580 in month 1, -$141,165 by month 6.
The first tells you whether the business model works. The second tells you whether you can survive proving it. A brand can be fine on cohort economics if LTV is strong and still run out of money.
It’s worth sharing this with whoever owns email, SMS, and subscription. Not “improve retention.” Recover $14,580.
Model this by cohort and policy period, not off a blended historical LTV. Old cohorts bought under a different offer, a different discount, and a different creative mix. They will not predict new ones.
Section 10: What happens if you turn ads off
First-time revenue slowly disappears (ads have a lag but that lag is supported by other ads). Apply your churn rate to what’s left.
At 8% monthly churn, our brand drops from $500k to $276k. A 44.8% decline.
That gap is your burn rate, and it’s the number that tells you whether ad spend is buying growth or renting revenue.
Which of the four are you
Put aMER and nCAC next to blended MER and almost every brand lands in one spot.
Retention coaster. MER 6.0, aMER 1.3, nCAC $182, 78% repeat. Returning customers carry the business. Rebuild prospecting with hooks written for cold traffic and accept a lower blended MER for a quarter.
Leaky bucket. MER 2.3, aMER 2.1, 10% repeat. MER barely clears aMER because there’s no repeat business to average in. Retention before scale.
Too efficient to grow. MER 8.4, aMER 4.6, marketing at 12% of revenue. Can afford far more customers than it buys. Spend until marginal aMER hits breakeven.
Scaling past breakeven. MER 1.6, aMER 0.9, nCAC $238, payback past 12 months. Cut the worst 20-30% of spend, reset to the efficient core, rebuild through creative testing.
Three things to watch out for:
Blending offers. If you sell at multiple price points, build a separate P&L per offer. A $39 trial and a $150 bundle have different economics and blending them hides both.
Using it on first-purchase-profitable businesses. If you have to make money on order one, skip the MER floor entirely and manage to marginal CPA.
Reading one month. Promo months swing acquisition CM hard. The signal is the trend across months, not the level in any one.
Get the model
curtishowland.com/download/first-customer-pnl
Export new vs returning from Shopify, fill the blue cells, and read sections 6 through 8. Everything else calculates.
Your blended P&L tells you what your business earned. The first customer P&L tells you what your business is worth buying.
Hope this helps!
Curtis