The Number That Tells You If Paid Actually Pays Back. It Is Not ROAS.

Most teams figure out fairly quickly that platform ROAS is not the same as profit. What tends to happen next is they replace ROAS with blended ROAS or MER and think they have fixed it. Those are better metrics. But they still only measure revenue. A 3x MER on 30% gross margins is an unprofitable business. A 5x MER on 70% gross margins is excellent. Revenue multiples without the margin context below them mislead in a different direction, but they still mislead.

The framework that actually answers the payback question combines three measurements. First, contribution margin per order (CM2): does this sale leave money behind after all variable costs, including the ad spend that generated it? Second, CAC payback period: how long does it take to recover what you spent to acquire this customer, measured in CM2 dollars? Third, LTV:CAC measured on a contribution margin basis: over 12 months, does this customer return enough CM2 to justify the acquisition cost?

We have managed more than $200M in ad-driven revenue across DTC and CPG brands. These are the three numbers that tell us whether paid is working, and the ones we gate scaling decisions on.

If you want the full mechanics of why ROAS misleads, including the double-counting problem and the break-even ROAS formula by margin profile, we covered that in A 4x ROAS Can Still Lose You Money. Here Is the Math.. This article picks up from there.

Contribution Margin: The Baseline

Before payback and LTV mean anything, you need a clean contribution margin number.

Contribution margin is revenue minus all variable costs: cost of goods, outbound fulfillment, payment processing fees, net returns, and ad spend. There are two versions. CM1 is before ad spend. CM2 is after ad spend. For evaluating whether paid ads are profitable, CM2 is the number you need.

CM2 is the margin that remains after everything it cost to produce a paid sale. If it is positive, the sale left the business better off. If it is negative, the business paid more to make that sale happen than it received in net margin.

The median gross margin across publicly traded DTC and CPG brands is 56.6%, drawn from SEC EDGAR 10-K filings (source). That number is dragged upward by beauty and personal care companies operating at 65-71%. Food and beverage brands occupy the bottom quartile, with gross margins ranging from roughly 3% to around 38% for better-positioned brands.

For a food brand with 35% gross margins, after fulfillment, processing, and a normal return rate, CM1 before ad spend is typically 15-18%. That is the entire ceiling for paid acquisition. If ads consume 14% of revenue, CM2 is barely positive. If CPMs rise and ad spend climbs to 20%, CM2 turns negative.

This is why the same 4x ROAS looks very different across categories. For a beauty brand at 70% gross margin, a 4x ROAS usually generates strong positive CM2. For a food brand at 35%, it may be below the break-even threshold. The platform does not know your margin structure. You do.

CM2 per order is the baseline. But it only tells you about a single transaction. Most businesses do not generate all their value from the first purchase, which means CM2 per order alone is not sufficient to evaluate whether the acquisition model works.

The CAC Payback Window

CAC payback period is the number of months it takes to recover what you spent to acquire a customer, measured in CM2 dollars. It is the bridge between the first transaction and the long-run economics.

The formula: Payback period (months) = Blended CAC / Average monthly CM2 per customer.

Two things to get right in that calculation.

Blended CAC should include everything it cost to acquire the customer: platform spend across all channels, creative production, influencer costs, agency fees, and any acquisition-related overhead. Using media spend alone understates the real cost by 20-40% for most brands. If you spent $180,000 in total acquisition activities last month and acquired 3,000 new customers, blended CAC is $60 regardless of what any single platform reports.

Average monthly CM2 per customer is trickier. For a product with a predictable monthly replenishment cycle, it is relatively straightforward: CM2 per order multiplied by average monthly order frequency. For products with less predictable reorder patterns, you need actual cohort data to know what monthly CM2 looks like in practice.

For most pure DTC brands, payback falls in a 6-12 month window. Categories with natural monthly replenishment, like food, beverage, and pet products, compress this to 1-3 months because purchase frequency is high enough to recover CAC quickly. Fashion and home goods, where the repurchase cycle is longer, push toward the 6-12 month end or beyond (source).

Why does the payback window matter beyond the profitability calculation? Cash. A 9-month payback window means you are funding 9 months of acquisition spend before that spend becomes self-sustaining. For a brand growing at 30% year-over-year, a 9-month payback requires significant working capital to bridge the gap between when you pay to acquire customers and when those customers return enough CM2 to cover the cost.

The payback window also frames risk. Most things can change in 9 months. CPMs rise. Platform algorithm shifts reduce conversion rates. A competitor enters the category and raises acquisition costs. Every month of payback is another month of exposure to those risks.

Payback tells us two things: whether the economics are sound over time, and how much cash we have at risk while waiting for them to prove out. Under 6 months with strong retention signals: green light to scale. Six to 12 months: requires a working capital plan and close cohort monitoring. Over 12 months without a subscription revenue stream: fix retention before adding budget.

Subscription models change the payback math substantially. A customer who subscribes at first purchase generates CM2 on their second order within the first month, often before the first-order CM2 has fully cleared. A 15-20% subscription take rate on first orders can move a 9-month payback to under 4 months for that portion of the acquisition cohort.

Segment Payback by Customer Source

The aggregate payback number can hide a wide distribution underneath it. A brand with a 7-month average payback might have two very different customer populations: lower-CAC customers from branded search who reorder every 2-3 months and pay back in 4 months, and higher-CAC customers from cold prospecting who buy once and never return, paying back in 24+ months or never.

Blending both into a single 7-month average makes the economics look manageable when one segment is burning cash.

We segment payback by acquisition source whenever the data supports it. Branded paid search customers typically have different retention profiles than cold prospecting customers. Meta retargeting customers tend to reorder faster than Meta prospecting customers. Direct traffic customers often have the highest LTV:CAC of any channel because there is no acquisition cost.

When you can see which acquisition sources produce the healthiest long-run cohorts, you can shift budget toward those sources and reduce spend on sources generating cheap customers who do not return. The cost-per-acquisition from a source tells you what you paid. The payback period from that source tells you whether it was worth it.

LTV:CAC: The Long-Run Scorecard

LTV:CAC is the lifetime value of a customer divided by the cost to acquire them. It is the ratio that determines whether the acquisition model makes economic sense over a full customer lifetime.

The critical detail: LTV must be measured in contribution margin dollars, not revenue.

LTV measured in revenue includes your variable costs and produces a ratio that looks significantly better than reality. If a customer generates $200 in revenue over 12 months from four orders, and each order has 30% CM2, their actual contribution is $60 in margin. LTV:CAC built on the $200 revenue figure is 3.3:1 if CAC was $60. LTV:CAC built on the $60 contribution margin figure is 1:1. The business broke even on that customer. The revenue-based ratio suggested it was thriving.

We use 12-month CM2 LTV as the standard. It measures what a customer actually returns in margin dollars in their first 12 months, based on real cohort data. LTV projections beyond 12 months introduce speculation that can produce almost any number depending on assumed retention curves. Twelve months is something you can measure and trust.

Based on verified analysis of DTC brand unit economics (source), here is what LTV:CAC looks like across major DTC verticals:

For most DTC brands, the healthy range is 2.5:1 to 4:1. Below 2:1 means you are paying more to acquire customers than they return in margin over 12 months. That is a model problem that more ad spend will not fix.

At the high end: above 5:1 in a transactional category typically signals underinvestment in growth. The economics are so favorable that there is untapped demand you are not reaching.

The 3:1 benchmark you often hear comes from SaaS, where gross margins run 75-90% and subscriptions generate years of recurring revenue. For a transactional DTC brand with 40-60% gross margins and uncertain repeat purchase rates, the right floor is 2.5:1 on CM2, with the specific target depending on your category and margin structure.

Two things move LTV:CAC: reducing CAC through better creative efficiency or higher conversion rates, or increasing LTV through more repeat purchases, subscription enrollment, or stronger retention. The fastest lever most brands can pull is retention.

How Retention Rewrites the Payback Math

The same acquisition investment produces radically different economics depending on what happens after the first sale.

If a customer buys once and does not return, you are paying full CAC for a single order's worth of CM2. In most DTC categories, one order does not produce enough CM2 to cover a typical acquisition cost. That means single-purchase customers are acquired at a loss, and the business is only profitable if enough customers come back.

If that same customer buys twice more within 6 months, driven by email flows or a subscription prompt, the economics transform. You paid the acquisition cost once. Purchases 2 and 3 generate their full CM2 with no additional acquisition spend. LTV:CAC on that customer might reach 3:1 or 4:1 even if the first order alone was unprofitable.

This is why retention infrastructure is not separate from paid acquisition. It is what determines whether paid acquisition works.

Post-purchase email sequences. A 3-email post-purchase flow that reliably drives a second purchase within 60 days compresses the payback window by months. The second purchase carries its full CM2 with no incremental acquisition cost. We have seen payback windows move from 9 months to under 4 months for cohorts with well-built post-purchase sequences versus those without.

Subscription enrollment at checkout. Even a 15-20% subscription take rate on first orders dramatically changes the 12-month CM2 curve for that portion of the cohort. Subscribed customers generate monthly CM2 with no re-acquisition cost, compressing payback and lifting LTV:CAC simultaneously.

Win-back sequences. A win-back flow that re-engages 12-15% of lapsed customers at zero acquisition cost improves LTV for that cohort without touching the acquisition side. Every recovered customer's subsequent CM2 lands entirely in the LTV:CAC numerator.

Abandoned cart recovery. If a paid click drove someone to add to cart, and your SMS or email sequence recovered that cart at 15%, those recovered customers carry no additional acquisition cost. Your effective CPA for the session drops while total CM2 from that paid cohort rises.

Product education sequences. Customers who understand how to use a product and know when to reorder tend to reorder more consistently. For a supplement brand, a 4-email educational sequence that teaches dosing and expected timelines can lift 90-day reorder rates meaningfully. That reorder rate is the primary input to LTV.

We track the retention stack's impact by looking at cohort CM2 curves: what does cumulative CM2 look like for customers acquired in a given month at 30, 60, 90, 180, and 365 days? When that curve rises consistently, paid acquisition is working and retention is doing its job. When it flattens after month 3 because customers stop reordering, the retention infrastructure is where the problem lives.

The scaling implication: do not increase paid acquisition budget before the retention stack is working. If 90-day reorder rate is below 15% and the cohort CM2 curve is flattening, more acquisition spend scales the loss. Fix retention first, then scale.

The Three-Number Scorecard

Here is what we look at before any scaling decision:

All three have to be in range. CM2 positive but payback at 15 months is a cash problem. CM2 positive, payback at 5 months, but LTV:CAC at 1.8:1 means the 12-month economics are borderline and the model depends on retention improvements that have not yet arrived.

ROAS continues to have a role as a relative diagnostic within a channel, useful for comparing creative, ad sets, and audience segments. It does not appear in this scorecard because it is not a verdict on profitability.

When to Scale, When to Fix First

If all three metrics are in range, scale aggressively. The economics support it and additional budget should produce proportional CM2 growth.

If CM2 is positive but thin, payback is 8-10 months, and LTV:CAC is trending toward 2.5:1, you have a model that is working but sensitive to cost increases. Scale carefully. Monitor cohort CM2 weekly. Have a plan for what you do if CPMs rise 25%.

If any metric is in the red zone: solve the underlying problem before increasing budget. A negative or barely-positive CM2 is typically a margin structure problem (pricing, COGS, or fulfillment costs) or a conversion rate problem. A payback window over 12 months is almost always a retention problem. An LTV:CAC below 2:1 can be either, or both.

More spend does not fix broken unit economics. What it does is scale the break.

The mistake we see most often: a brand sees a strong platform ROAS, interprets it as "ads are working," and scales. What they are actually seeing is a revenue attribution claim from a platform whose incentive is their next dollar of spend. By the time the unit economics catch up with the budget increase, the brand has added significant spend to an already marginal model.

How to Track These Numbers

Most brands already have the data. The challenge is assembling it in one place.

For CM2 per order: pull average order value, COGS as a percentage, fulfillment cost, return rate, and processing fees from your 3PL, Shopify, and accounting system. Add blended CAC from all paid channels (total acquisition spend divided by new customers in the period). Calculate CM2 per order and update it monthly as costs shift.

For CAC payback: build a monthly cohort table in a spreadsheet. For each acquisition month, track total CM2 generated by that cohort at 30, 60, 90, 180, and 365 days. Your CAC for that cohort is total acquisition spend for that month divided by new customers acquired. The month where cumulative cohort CM2 equals CAC is the payback point.

For LTV:CAC: use that same cohort table. At the 12-month mark, cumulative CM2 for a cohort divided by that cohort's CAC is your LTV:CAC. Tracking how this ratio changes across cohorts shows whether the business is getting more or less efficient over time.

Tools like Triple Whale, Northbeam, and Peel can automate parts of this for Shopify stores. But a spreadsheet built from Shopify order data and your actual cost inputs is sufficient and is easier to trust because you control all the inputs.

The brands that build this tracking are almost always surprised by what they find. Platform dashboards tell you one story. Cohort-level CM2 data tells a different one.

Frequently Asked Questions