The DTC Profit Metrics That Matter

There is no shortage of metrics available to a DTC brand. The problem is that most of them are downstream of a handful that actually determine outcomes, and time spent on the derivatives is time not spent on the drivers.

Five numbers cover almost everything, and the order they come in matters as much as the numbers themselves.

The sequence

1. Gross margin. Net sales minus landed cost of goods. This sets the ceiling on everything. At 22% no amount of operational skill produces a profitable business, and every analysis below this line is premature until it is fixed.

2. Contribution margin rate. After shipping, payment fees, packaging and fulfilment. The fastest-moving number in the business and the earliest warning system. Watch it weekly, as a percentage.

3. Contribution margin per order against acquisition cost. Whether growth funds itself. Both sides in currency, no attribution model required. If this is negative, scaling makes things worse and everything else is noise.

4. Profit after advertising. Contribution margin for the period minus all acquisition spend including agency fees and invoice tax. Reconciles to real money; cannot be inflated by overlapping attribution windows.

5. Net profit, with founder salary at market rate. Quarterly. The only number that says whether the business works.

Work down the list and stop at the first failure. That is where the problem is, and fixing anything below it will not help.

Why the order matters

Each metric strips out a different category of cost, so a weak result at any level points at a specific cause rather than a vague one.

A store with poor net profit and healthy figures at every level above it has a scale problem — the unit economics work, there are not enough orders to carry the fixed base. That is the case in the example: every order contributes $7.08 above acquisition cost, and 1,240 of them cannot cover $11,200.

A store with poor net profit and a negative gap at level three has an economics problem. Growth deepens the loss.

These look identical from the bottom line and demand opposite responses. The first should probably spend more; the second must not. Reading only net profit gives no way to tell them apart, which is why the sequence exists.

The metrics that mislead

Revenue. Rises when you discount, when you buy more traffic, when you lower a free-shipping threshold. Every one of those can reduce profit.

ROAS. Uninterpretable without contribution margin, and inflated by attribution that platforms grade themselves. Its useful content is captured by profit after advertising.

Average order value. Rises when you bundle heavy items, which can lower margin per order. A number that moves for both good and bad reasons is a poor monitor.

Lifetime value as revenue. Two customers with identical LTV can differ entirely in profit if one returns half of what they buy.

Gross margin used as profit. The most expensive single error in ecommerce, because it overstates the acquisition ceiling by the whole cost of fulfilment.

None of these are worthless. All of them are frequently used to make decisions they cannot support.

Rates, not totals

Every metric here should be watched as a rate or a per-unit figure, not as a currency total.

Totals rise with volume and reassure. Contribution margin in dollars going up while the rate falls is a business getting bigger and less healthy, and only the rate shows it.

The exception is profit after advertising, which is genuinely useful as a total because it represents money the business kept — and even there, tracking advertising as a percentage of contribution margin alongside it adds the diagnostic the total lacks.

Cadence

Weekly: contribution margin rate, acquisition cost, and the gap between margin per order and CAC. Ten minutes.

Monthly: gross margin, profit after advertising, full P&L with comparison columns.

Quarterly: net profit with founder salary, cohort curves at matched ages, product ranking rebuilt with current costs.

Anything measured more often than it can be acted upon becomes noise, and noise trains people to ignore the display.

Getting the inputs honest first

The sequence only works if the numbers feeding it are right, and four errors recur often enough to check before trusting any of them.

Uncosted variants. Counted as pure margin, inflating gross margin, contribution margin and the acquisition ceiling simultaneously. Count the orders shipping with no cost of goods attached — a handful can move a store-level figure by points.

Founder salary at zero. Makes net profit look several times better than it is, and describes a business that only functions while somebody works unpaid.

Platform cost per purchase used as CAC. Understates acquisition cost by excluding agency fees, invoice tax and welcome discounts, and by counting repeat buyers as new customers.

Returns left out of contribution margin. In categories with meaningful return rates this overstates the ceiling by more than any other single omission.

All four push in the same direction: they make the business look healthier than it is. That is not coincidence — each one omits a cost, and there is no equivalent error that omits revenue. A store that has never checked these is almost certainly working from figures that are too generous rather than too harsh.

What none of these tell you

Two blind spots worth naming, because these five metrics are frequently treated as complete.

Cash. Every figure here can be healthy while the bank balance falls, most commonly because inventory was purchased. Profit and cash diverge routinely, and only a cash view with inventory commitments shows the gap.

Concentration. A store where one product, one channel or one supplier produces most of the margin passes every test above and is fragile. The check that reveals it is asking what these numbers look like with the largest single dependency removed.

Both are quarterly questions rather than weekly ones, and both have ended businesses that were profitable on every metric they were watching.

Tooling for these metrics

Best DTC profitability software compares the products that report these metrics continuously. If you are running a direct-to-consumer brand specifically, NetNet for D2C brands sets out which of them NetNet computes and how.