New-to-brand metrics on Amazon: what they tell you about your ads
Written by The Pixamp Team
PixampA sale is a sale, until you ask whether it came from a new customer or from someone who was going to reorder anyway. Amazon's new-to-brand metrics answer that question.
What are Amazon new-to-brand metrics?
New-to-brand (NTB) metrics flag whether an order came from a shopper who had not bought any product from your brand on Amazon in the prior 12 months. They split a flat order count into two buckets: customers you acquired and customers you kept.
Amazon reports them inside Sponsored Brands, Sponsored Display, Streaming TV, and Amazon Attribution. According to Amazon Ads documentation, the core fields are new-to-brand orders, new-to-brand sales, and new-to-brand units, each shown next to the total so you can read the share.
The distinction matters because acquisition and retention have different economics. Paying a premium to win a first-time buyer can be sound. Paying that same premium to buy a reorder you would have gotten for free is waste.
How does Amazon count a new-to-brand order?
Amazon looks back 12 months from the order date. If the shopper has no recorded purchase of any ASIN under your brand in that window, the order counts as new-to-brand. One prior purchase anywhere in your catalog, and the shopper is a returning customer for the next 12 months.
A few mechanics shape how you read the number:
- The lookback is brand-level, not product-level. A shopper who bought your shampoo last spring is not new-to-brand when they buy your conditioner today, even though the product is new to them.
- The window is rolling. A customer whose last purchase was 13 months ago resets and counts as new-to-brand again.
- Attribution is Amazon's, so an order credited to your ad had to meet Amazon's click or view window for that ad type.
- The signal is directional at low volume. A handful of NTB orders in a week tells you little. A few hundred starts to mean something.
That 12-month boundary sits behind every NTB number you read. A brand that has sold for years will see a lower new-to-brand rate than a launch, and neither figure is good or bad on its own.
Why separate new customers from repeat buyers?
Because the two tell opposite stories about the same campaign. A high NTB share means your ads are reaching people outside your existing base, which is what acquisition spend is for. A low NTB share on a prospecting campaign means you are paying prospecting prices to serve ads to people already in your funnel.
Blended ROAS hides this. A campaign can post a strong return built almost entirely on repeat buyers who cost little to convert, while the acquisition portion quietly loses money. Split the orders and the real customer acquisition cost comes into focus: ad spend divided by new-to-brand orders, not total orders.
The reverse is just as common. An upper-funnel campaign that looks expensive on blended ROAS can be your best acquisition engine once you isolate the NTB orders it generates.
How to read the numbers for ad spend
Pull NTB alongside totals and compute the share for each campaign and ad type. Then judge each line by the job you gave it.
| Metric | What it isolates | Use it to judge |
|---|---|---|
| NTB orders | First-time buyers in 12 months | Acquisition volume |
| NTB order rate | Share of orders that are new | Whether a campaign prospects or harvests |
| Cost per NTB order | Spend ÷ NTB orders | True acquisition cost |
| NTB sales | Revenue from new buyers | Acquisition return |
| Total orders | Everyone | Overall efficiency |
Read cost per NTB order against the lifetime value of a customer, not against a single order's margin. Acquisition that looks unprofitable on the first purchase can still pay off once you account for the reorders that follow, which is why owning your retail demand and the repeat purchases it drives belong in the same calculation.
Two guardrails keep the read honest. Compare NTB rate across ad types rather than in isolation, since Sponsored Products and Sponsored Brands serve different stages. And watch the trend over weeks, not days, because the 12-month lookback makes any single day noisy.
Where new-to-brand metrics stop
NTB is an Amazon-internal view. It knows what happened on Amazon and nothing about where the shopper came from. A buyer who found you through a Meta ad, a creator post, or a Google search all look identical once they check out.
That matters when external traffic drives the sale. Amazon can tell you an order was new-to-brand, but it cannot tell the ad platform that paid for the click. The platform sees the click and then silence, so its delivery keeps learning from clicks rather than from the new customers those clicks produced.
NTB also cannot net out organic baseline on its own. Some first-time buyers would have found you without any ad. Separating ad-driven acquisition from that baseline needs a holdout, which is where incrementality testing earns its place. NTB tells you a customer is new. A test tells you whether your ad is why.
Where to start
- Website: www.pixamp.io. What Pixamp does, pricing, and the FAQ. First 1,000 clicks free, no card required.
- How it works: www.pixamp.io/#how-it-works. The three-step setup: connect Meta Business Manager, add a retailer button, launch. Live in under an hour.
- Book a demo: www.pixamp.io/#contact. A 20-minute walkthrough on a real retailer page, with the founding team.
New-to-brand metrics tell you who you acquired on Amazon. Pairing them with where that shopper came from is what turns the number into a decision.