NTB sales means new-to-brand sales: advertising-attributed revenue from shoppers who have not bought from your brand on Amazon during the previous 12 months. It does not mean everyone making their first purchase from your business anywhere. A returning customer outside that window can qualify again. Amazon's definition.

NTB sales, purchases and percentage: which number do you need?

Use the matching sales or purchase denominator from the same campaign, reporting period and attribution scope.

Metric	Calculation	Question it answers
NTB sales	Reported revenue attributed to NTB purchases	How much attributed revenue came from NTB purchases?
NTB sales share	NTB sales / total attributed sales × 100	What share of attributed revenue was NTB?
NTB purchase share	NTB purchases / total attributed purchases × 100	What share of attributed purchases was NTB?
Cost per NTB purchase	Advertising spend / NTB purchases	How much campaign spend was used per attributed NTB purchase?

Cost per NTB purchase is a campaign allocation, not a verified cost per unique newly acquired customer. If the denominator is zero, report the ratio as unavailable rather than dividing by zero. Keep missing NTB data separate from a reported zero.

Worked example: a high NTB percentage can still be expensive

This is an illustrative calculation, not an Arctavia customer result. Suppose one campaign reports £400 spend, £1,000 attributed sales and 20 purchases. Its NTB columns show £600 sales and 12 purchases.

• NTB sales share: £600 / £1,000 = 60%.
• NTB purchase share: 12 / 20 = 60%.
• Cost per NTB purchase: £400 / 12 = £33.33.
• Campaign ROAS: £1,000 / £400 = 2.5.

Those numbers do not establish profit. As a separate simplified scenario, if the £1,000 revenue generated £300 contribution before advertising, spending £400 would exceed that contribution by £100. Returns, taxes and the exact revenue basis also need to be consistent. A strong acquisition percentage cannot replace that calculation.

The 12-month lookback is not the ad-attribution window

The NTB lookback classifies the shopper's purchase history. An attribution window determines whether an interaction receives credit for a later purchase. Record both before comparing reports; they answer different questions. See our Amazon attribution-window guide.

Amazon documents NTB metrics for Sponsored Brands and Display. Availability and naming depend on the report you actually use. Check its metric definitions rather than assuming that an absent column means no new shoppers. Amazon NTB help.

For campaigns supporting view-through measurement, keep the interaction scope consistent. Do not add click-attributed purchases to an inclusive total: the total may already include them. Amazon's conversion metric changes distinguish combined, click, view, promoted and halo metrics.

A practical weekly review

1. Export spend, total sales, purchases and NTB metrics for one comparable campaign group.
2. Record the account, currency, date range, interaction scope and export date alongside the figures.
3. Calculate sales share and cost per purchase. Compare acquisition campaigns with acquisition campaigns, rather than with branded defensive campaigns.
4. Check contribution margin, total business sales and inventory before proposing more spend.
5. Record the proposed change and review it after the reporting period has matured. Keep a comparison group where possible.

NTB attribution alone does not prove advertising caused an additional purchase. A before-and-after increase may also reflect seasonality, price, stock availability or a change in campaign mix.

Frequently asked questions

Is a higher NTB sales percentage always better?

No. The percentage can rise while total sales or profit falls. Check the numerator, denominator and spend together.

Does NTB mean a first-ever customer?

No. Its purchase-history boundary is your brand on Amazon during the lookback period; it is not your complete customer history across channels.

Should I add NTB sales to total ad sales?

No. NTB sales are a subset of the matching attributed sales total. Adding them would count that revenue twice.

Put the acquisition cost beside your margin

Before increasing a budget, calculate your break-even ACoS using your own price, costs and fees. Then review ACoS, ROAS and TACoS with the same reporting definitions. The calculation supports a decision; it does not promise a campaign outcome.

Related Arctavia resources

Use these supporting pages to compare Amazon PPC operating models and implementation choices.

• Amazon Ad Rejections: Check Claims, Images and Destinations
• Amazon Ads ACoS Suddenly Increased: What to Check First
• Amazon Ads ML Bid Review: Guardrails Before Aggression
• Amazon PPC Automation: Data, Decisions and Operating Controls
• Amazon PPC Automation Migration Checklist
• Amazon PPC Automation vs Agency: Which Operating Model Fits Better?