New-to-Brand (NTB)

New-to-Brand (NTB) is Amazon's metric for purchases made by shoppers who have not bought your brand on Amazon in the prior 12 months. It is reported as both an absolute count (NTB orders, NTB sales) and as a percentage of total attributed purchases (NTB%).

NTB is available in Sponsored Brands and Sponsored Display reports. It is not available in Sponsored Products. Brand Registry is required.

Why NTB matters more than ACOS for upper-funnel campaigns

ACOS measures efficiency — how much you spent per euro of attributed revenue. It treats all revenue as equivalent. NTB makes a critical distinction: revenue from an existing customer (low incrementality, likely to have purchased anyway) vs. revenue from a new customer (high incrementality, genuinely acquired).

Two campaigns can have identical ACOS and radically different business value:

Campaign B acquired €2,800 of new-customer revenue for the same spend. Campaign A is mostly re-attribution of customers who would have bought anyway. The ACOS view says they're equivalent; the NTB view says B is more than 2× as valuable.

NTB as a Target ACOS modifier

The practical use: tolerate higher ACOS on high-NTB campaigns, lower ACOS on low-NTB.

A defensible framework:

NTB-adjusted Target ACOS = Base Target ACOS × (1 + NTB% × LTV multiplier)

If NTB customers carry 2× the lifetime value of a one-time buyer (a defensible assumption in repeat-purchase categories), then a campaign running 60% NTB justifies a Target ACOS 60% higher than the baseline.

The reverse applies. A campaign running 5% NTB (almost all attribution to existing customers) should be held to a tighter ACOS than baseline because almost none of the spend is buying incremental growth.

Where high NTB typically appears

• Competitor conquest keywords — shoppers were looking for a different brand and switched. Almost always 70%+ NTB.
• Sponsored Brands on generic high-funnel keywords — shoppers in discovery mode. Typically 40–60% NTB.
• Sponsored Display audience targeting (in-market audiences) — explicit prospecting. 50–70% NTB.

Where low NTB typically appears

• Branded keyword campaigns — shoppers searching your brand name. 5–15% NTB; the shopper already knew you.
• Sponsored Display retargeting (Views Remarketing) — shoppers who already visited your PDP. 10–25% NTB.
• Sponsored Brands on your own brand store traffic — defensive plays. 5–20% NTB.

NTB% as a north-star metric

For brands focused on growth (rather than pure margin defence), tracking blended NTB% as a quarterly metric is often more informative than blended ACOS:

• NTB% rising at constant ACOS = the account is acquiring more new customers per spend dollar. Genuine growth.
• NTB% flat at falling ACOS = the account is becoming more efficient on existing customers but not acquiring more. Margin improvement without growth.
• NTB% falling at constant ACOS = the account is increasingly servicing existing customers. Growth has stalled.

Common mistakes

• Ignoring NTB because it's only in SB/SD. That's the point — the formats designed for upper-funnel acquisition are where NTB matters most. Use it there.
• Optimising SB campaigns on ACOS alone. Under-funds the highest-incrementality spend in the account.
• Treating NTB as gospel. The 12-month lookback misses the long tail of repeat customers; the household-level definition has edge cases. It's the cleanest available metric, not a perfect one.
• Not naming an NTB target alongside an ACOS target. Both should appear on the campaign brief. "Hit 25% ACOS at 50%+ NTB" is a complete brief; "hit 25% ACOS" alone is half a brief.