The DTC guide to true margin, not just revenue.

This is the shape of the calculation: revenue steadily eaten by real costs at each stage, from what it costs to make the product, to what it costs to ship and discount it, to what it costs to acquire the sale in the first place. What follows is the set of definitions and data rules that make CM1, CM2, and CM3 accurate for your own product line, rather than a rounded illustration on somebody else's.

Gross margin, CM1, CM2, CM3: what each one actually measures

Most DTC teams use "margin" to mean three or four different numbers, often in the same meeting. Here's how to keep them straight, and the specific data rules that make each one accurate rather than approximately right.

The revenue line comes first, and it's where most calculations already go wrong

Before you touch a single cost line, get the revenue number right. Use pre-tax item revenue, the subtotal, not the order total that folds in shipping charged to the customer and tax. If your Shopify reporting already pulls revenue ex-VAT (common with Shopify's built-in reports), don't apply a VAT deduction again. Doing that strips tax out twice and understates revenue, which understates every margin figure built on top of it.

Exclude gift card sales from revenue entirely. A gift card sale is deferred revenue: the cash comes in now, but the sale that actually earns it happens later, when the card gets redeemed. Counting it as revenue at the point of purchase overstates your numbers for that period and creates a mismatch when it's redeemed later.

Gross margin

Gross margin is the simplest cut: revenue minus the direct cost of the goods sold, with nothing else subtracted. It answers one question: how much does each product cost to make or buy, relative to what it sells for. It doesn't touch shipping, discounts, refunds, or marketing. Most finance teams already track this, but it's the least useful number for a growth or marketing conversation, because it says nothing about what it actually costs to acquire and deliver the sale.

CM1: gross margin, done properly

CM1 is revenue minus COGS, but the accuracy of this number lives entirely in how well you've mapped costs to products.

Unit cost isn't sitting on the line item in your order data. It has to be built from a separate product-variant cost map, joined by SKU. That join is where most homegrown spreadsheets quietly fall apart: new SKUs launch without a cost entered, variants get renamed, and the cost map drifts out of date. So alongside any CM1 figure, always report your COGS coverage rate, the share of total order value that has a matched cost. If your coverage rate is 80%, your CM1 is only as trustworthy as that 80%, and the untracked 20% is very likely understating your true cost, not overstating it. A CM1 number with no coverage rate next to it is a number nobody should make a decision on.

One line item needs special handling here: gift-with-purchase (GWP) items. These show zero revenue on the order, but they carry a real unit cost, and that cost has to be included at CM1. What you should exclude is the GWP unit itself from any average order value calculation. Leave it in and your AOV looks lower than it is; leave the cost out of CM1 and your margin looks higher than it is. Either mistake distorts the number in the direction that makes the business look better than it actually is, which is exactly the kind of error that doesn't get caught until it matters.

CM2: after the costs that scale with each order

CM2 takes CM1 and subtracts the variable costs tied to fulfilling and discounting each individual order: shipping, discounts, and refunds.

Discounts split into two categories that behave very differently in your data. Standard discount codes are easy to capture; they're logged against the order and show up cleanly in most reports. Markdown or sale pricing usually isn't captured the same way. If a product is simply priced lower at checkout rather than discounted via a code, that discount is often invisible in the field your reporting normally checks. If your brand runs markdown pricing, calculate the true discount depth separately, against the original list price, or your reported discount rate will look understated even though the actual margin impact is real.

Refunds need a similar split by order age. For orders more than 30 days old, use actual refund data; by that point, most returns that are coming in have already come in. For very recent orders, only use actual refunds recorded so far, unless you have documented historical return-rate data you can apply as a provision. Don't estimate a return-rate percentage without that history behind it. A guessed rate dressed up as a real number is worse than no number at all, because it looks precise when it isn't.

CM3: after marketing spend

CM3 takes CM2 and subtracts the marketing cost of acquiring that revenue: your paid spend across ad platforms for the period. This is the number that answers the question a CM1 or CM2 figure can't: after everything it actually cost to make, ship, discount, and market the product, what's left. It's also the number most likely to swing sharply week to week, since ad spend and revenue rarely move in lockstep.

One more identification rule matters across all three CM levels, especially if you're segmenting by new versus returning customer: identify a new customer by whether this is that customer's very first order, not by account creation date and not by total order count. Both of those alternatives break the moment you look at historical data, since accounts get created for all kinds of reasons unrelated to a first purchase, and order counts can be wrong if past orders were imported, merged, or refunded to zero.

Contribution margin benchmarks across DTC brands ($5M–$75M revenue) put the median around 25%, with top-quartile brands running 54–56%. Below 20% is a warning sign regardless of category, below 15% is critical territory.

Sourced from Finaloop's P&L dataset across 800+ DTC brands, via Commerce Catalyst's 2026 benchmark report.

The fill-in template

Copy this structure into a spreadsheet. Run it at the order-batch or weekly level, whichever matches how your team already thinks about the numbers.