The problem with judging every campaign on the first order
Everywhere else in this series the rule is strict. A campaign below your break-even Return on Ad Spend (ROAS) loses money. Cut it. For a store whose buyers come once and vanish, that rule is right.
But some stores are not one-order stores. Subscriptions. Refills. Anything customers use up and reorder. Here the first sale starts a relationship. Repeat orders carry no new ad cost. So part of their profit belongs to the campaign that started them. Judge that campaign on the first order alone? It looks like a disaster worth cutting.
Founders fall off both sides of this horse. Some apply the first-order rule everywhere. They choke campaigns that would have paid back many times over. Others accept any losing campaign as an ‘LTV play’. Lifetime Value (LTV) is the total profit a customer brings over time. The agency said so. And they never check the data.
1. An example showing you the numbers
You run a hair-care subscription brand. The starter set sells for $100. Customers can subscribe for refills. Your main paid social campaign spends $80 to win each new customer. That $80 is your . It is the ad price of one new customer.
On the house numbers, the five non-ad costs eat $70 of every $100 order. That leaves $30. So the first order returns $30 against $80 of ads. A $50 loss. An agency ROAS of 1.25x on a store that breaks even at 3.3x. Every rule in this series says: cut it.
Now follow the same customers for twelve months. A is just that: the group of customers who joined in the same month, tracked together.
One acquired customer, first-order view vs 12-month cohort view
Per acquired customer, whole dollars. Refill orders carry no new ad cost. Each leaves the full $30.
| Line item | First-order view | 12-month cohort view |
|---|---|---|
| Ad spend to acquire (CAC) | -$80 | -$80 |
| First order leaves (after the five non-ad costs) | +$30 | +$30 |
| First-order result | -$50 | -$50 |
| Share of customers who subscribe | not counted | 60% |
| Refill orders per subscriber (12 months) | not counted | 6 × $30 = $180 |
| Repeat contribution per acquired customer | not counted | +$108 |
| Value per acquired customer at 12 months | -$50 | +$58 |
Look at the table above. If you only count the first order, this campaign looks like a loser. You spend $80 in ads to win a customer, and their first order only gives you $30 back. That is $50 lost per customer. But these customers keep coming back for refills, and refill orders need no new ad spend. Count everything they buy over 12 months and the picture flips: the same customer is now worth $58 in profit. You lost money on the first order, but the lifetime value of the customer is positive.
There is one catch: you need cash to wait. Refills arrive about every two months, so it takes around six months before a customer pays back their $80. Win 1,000 new customers and you have $50,000 tied up while you wait. But if you can fund that wait, 12,000 new customers a year leave you about $700,000 ahead. The lesson: losing money on the first sale does not make the campaign a bad one. Judge it on lifetime value, not the first order.
One honest note. Every number in that table must be measured, not hoped. What if subscribers stop after three refills, not six? The twelve-month value falls to $84. Barely above the $80 it cost. The defense lives or dies on your own data.
The sentence that changes how you think about first-order losses
A first-order loss is an investment only when your own cohort data proves the payback. Otherwise it is a loss with a story attached.
Here, $80 buys a customer whose first order returns $30. Refills lift the total to $138 within a year. The math works because it is measured, monthly, from your own data. When the measuring stops, the story takes over.
2. How to build the cohort proof for your acquisition math
The first pass takes two weeks with your order data. After that, a monthly refresh.
- Define your cohorts. Group customers by the month they joined. And by the source that brought them: paid social, paid search, organic. A blended cohort hides what you need to see.
- Track each cohort’s real contribution over time. At 30, 60, 90, 180, and 365 days, add up each cohort’s contribution per customer. Contribution, not revenue. Pull it from your store and subscription data. Never from ad-platform tracking. It flatters the numbers.
- Work out the payback month and the 12-month ratio. The month the running total passes CAC is your payback point. Contribution at 12 months divided by CAC is your ratio. Above 3:1 is strong. 2:1 to 3:1 can be defended. Below 1.5:1 is too thin. This brand’s 1.7x sits in the ‘defend it, keep pushing’ zone.
- Turn the measured LTV into a CAC ceiling. Want at least 2:1? With an LTV of $138, the most you may pay is $69 per customer. So this brand’s $80 campaign is overpaying. Push CAC down. Or accept the thinner 1.7x on purpose, in writing.
- Re-measure monthly and let the data move the ceiling. Cohorts drift. January buyers differ from July ones. Results 15 to 20 percent below plan? Lower the ceiling and decide again.
3. One warning before you act
The biggest threat to this method is not bad math. It is math with a motive. Agencies have a reason to defend the spend that pays their fees. A cohort story you did not check is not proof.
So check hard. Use the campaign’s own cohort. Not the store average. Cheap-click campaigns often buy customers who never subscribe. And mind the cash. Even a winning cohort needs its $50-per-customer gap funded until payback. Never scale past what your spare cash can carry.
4. Frequently asked questions
What LTV-to-CAC ratio should I target?
Software firms target 3:1 or better at 12 months. Subscription brands usually work between 2:1 and 3:1. Fast-payback products, under six months, can defend 1.5:1. Below that, one soft quarter turns it into a plain loss.
My customers reorder but do not subscribe. Does this still apply?
Yes. Swap subscription data for repeat-buy data. Track each cohort’s orders and contribution across 12 months. The math does not care how customers come back. Only that they come back.
The LTV math works on paper, but the campaign drains my cash. Now what?
That limit is real. It wins. A month-six payback still needs every customer funded for six months. Cash cannot cover the gap? Scale down to what you can fund. Or line up funding first. Never spend on paper LTV alone.
How does this square with the break-even ROAS rule from this series?
Same six costs. More orders. Break-even ROAS judges one order. The cohort ratio judges all the orders that follow. One-order store? The first-order rule is the whole answer. Repeat store? Use both. First-order math sizes the loss. Cohort math decides if it is worth funding.
5. Quick reference: what to avoid and what to apply
What to avoid
- Cutting every below-break-even campaign without checking its cohort first.
- Accepting ‘the LTV justifies it’ without your own cohort data.
- Measuring cohorts from ad-platform tracking instead of your own store data.
- Defending a campaign with the store-average repeat rate instead of its own cohort.
- Scaling a money-losing first-order campaign past what your cash can fund.
What to apply
- Define cohorts by joining month and by source.
- Track the running contribution per customer at fixed checkpoints.
- Work out the 12-month LTV-to-CAC ratio and the payback month.
- Set a CAC ceiling from measured LTV, and hold every campaign to it.
- Re-measure monthly; lower the ceiling the moment cohorts weaken.
Want your six costs mapped for you?
Our free 31-Point Profit Diagnostic puts a dollar figure on where your profit is leaking. Or get a quote for done-for-you eCommerce accounting.
Definitions, modeling notes & rate-basis disclosures
Definitions
- The six profit levers
- (1) Discounts, (2) Refunds, (3) Product cost (landed COGS), (4) Sales channel and payment fees, (5) Shipping and 3PL fees, (6) Advertising spend.
- Customer Acquisition Cost (CAC)
- Ad spend divided by the new customers it wins. The price of starting one relationship.
- Lifetime Value (LTV)
- The total contribution a customer produces across all orders. Measured to a set point, usually 12 or 24 months. Contribution, not revenue.
- LTV-to-CAC ratio
- Measured LTV divided by CAC. It decides whether a first-order loss is an investment.
- Cohort
- The group of customers who joined in the same month, from the same source, tracked together over time.
- Payback period
- The months until a cohort’s running contribution equals its CAC. Until then, the campaign eats cash.
- Break-even ROAS
- The single-order bar from this series: 100 divided by the dollars left for ads per $100 order. Here, 3.3x.
- Contribution per order
- Selling price minus all six costs. What one sale really leaves behind.
Modeling notes
- This article uses the series’ standard teaching store: a $100 product, product cost $40, a standing 10% discount, shipping and 3PL $12, refunds 5%, and payment and channel fees 3%. That leaves $30 per order before ads. Only the advertising lever varies. The first order carries the campaign’s full $80 CAC. Refill orders carry no ad cost.
- First order: $30 - $80 = -$50 per customer. Agency ROAS: $100 ÷ $80 = 1.25x, against the 3.3x break-even.
- Cohort: 60% subscribe × 6 refills × $30 = $108 of blended repeat contribution. LTV = $30 + $108 = $138. LTV ÷ CAC = 138 ÷ 80 ≈ 1.7x. Net value per customer = $138 - $80 = +$58.
- Payback: refills land about every two months. Each cycle adds a blended $18 (60% × $30). So $30 + 3 × $18 = $84 ≥ $80 by month six. Annual: 12,000 new customers × $58 = $696,000, rounded to $700,000. CAC ceiling at 2:1 = $138 ÷ 2 = $69.
Rate-basis disclosures
- Baseline: Direct-to-Consumer hair-care subscription brand on Shopify, $100 average order value. Paid social is the main channel for new customers.
- New customers: CAC $80 each. First-order agency ROAS: 1.25x.
- Cohort behavior: 60% subscribe. 6 refill orders per subscriber over 12 months. Refills at the full $100, house cost rates, no ad cost.
- Non-ad variable costs: 70% of the selling price (product 40%, discount 10%, shipping and 3PL 12%, refunds 5%, fees 3%).
- Thresholds: LTV-to-CAC 3:1 strong. 2:1 is the floor for defense. 1.5:1 is the minimum for short paybacks. All figures whole dollars. 1.725x rounded to 1.7x.