DTC metrics glossary

Revenue an ad platform attributes to its own ads, divided by what you spent on that platform. Useful for optimizing inside the platform; unreliable for setting total budget, because each platform counts sales others also claim.

Formula Platform-attributed revenue ÷ platform ad spend

Revenue divided by ad spend with new and returning customers counted together. It can look healthy while new-customer acquisition weakens, because repeat buyers who would have ordered anyway lift the number.

Revenue from first-time buyers only, divided by ad spend. The cleanest single read on whether prospecting spend is actually finding new customers.

Formula First-order revenue from new customers ÷ ad spend

Total store revenue divided by total marketing spend across every channel. The business-level number, immune to platforms double-counting each other.

Formula Total revenue ÷ total marketing spend

Revenue from new customers divided by total marketing spend. MER can stay flat while aMER falls, which is how many DTC plateaus start.

Formula New-customer revenue ÷ total marketing spend

The ROAS at which an order covers its variable costs (product, shipping, fees, discounts and returns) and earns nothing. Below it, every sale loses money. It differs by product.

Formula 1 ÷ contribution margin % (before ad spend)

ROAS after product cost, shipping, fees, discounts and returns, calculated product by product. Shows which products actually make money on paid traffic.

The sales that would not have happened without an ad or channel. Platform ROAS shows what an ad can be linked to; incrementality shows what it caused. Measured with holdout or geo tests.

Sales on Amazon driven by ads aimed at your own website. They never show up in Meta or Google ROAS, so social and video ads are often undervalued for brands that also sell on Amazon.

What it costs to acquire one new customer. Calculated on new customers only; dividing all spend by all orders understates it.

Formula Acquisition spend ÷ new customers

What a customer is worth over a set period, usually 6 or 12 months. Most useful as gross profit rather than revenue, and split by first product, channel and offer.

What customers acquired in the same month are worth over time. Comparing cohorts shows whether newer customers are better or worse than older ones.

Months until a customer's profit covers what it cost to acquire them. Cheap-to-acquire channels often pay back slowest.

Months until each channel's customers pay back their acquisition cost. Answers which channel deserves the next dollar better than CPA does.

Share of customers who place a second order within a set window, such as 90 days. Read it by first product and channel, not just store-wide.

Share of new subscribers who take their second order. The number that decides whether subscription acquisition pays back.

How much revenue comes from first-time buyers versus existing customers. The fastest check for whether growth is durable.

When a meaningful share of customers only buy during promotions. Shows up as revenue concentrated in sale weeks and repeat buyers who never pay full price.

12-month value of customers, split by the first product they bought. Shows which products are your real entry points.

The product whose first-time buyers are most likely to go on and buy other products. Often not the best seller.

Which product customers buy next, and how often, after each first purchase.

Each product's share of ad spend set against the strength of the customers it creates. Highlights products that are under or over-funded.

Margin after the full cost of returns: refund, both shipping legs, handling, lost resale value and the ad spend that bought the order.

The full cost of getting a product into your warehouse: unit cost, freight, duties and tariffs, customs fees and inbound handling. Changes in tariffs change every break-even ROAS.

The order value above which shipping is free. Best set from your own order value distribution and judged on contribution per session, not AOV.

Revenue a tool credits to itself under its own rules. Klaviyo, Meta and Google can each claim the same order, so added together they can exceed real sales.

A one-question "How did you hear about us?" survey shown after checkout. Captures podcasts, creators, word of mouth and AI assistants that click-based attribution misses.