NTB means New-to-Brand. In retail media, it measures buyers who purchase a brand for the first time inside a defined lookback window. It matters because it shows whether a campaign created real acquisition or simply captured shoppers who were already buying the brand.
That distinction is the whole point.
Most campaign reports are comfortable with activity. Impressions delivered. Clicks recorded. Sales attributed. ROAS calculated. The chart looks green, so the campaign is called successful.
But retail media should answer a harder question:
Did the campaign bring new buyers into the brand?
That is where NTB becomes useful.
What New-to-Brand actually means
A New-to-Brand buyer is a shopper who buys a brand after not buying that brand during the chosen lookback period.
If a shopper buys Dove shampoo today and has not bought Dove shampoo, Dove conditioner or another Dove haircare SKU in the last 12 months, that shopper can be counted as New-to-Brand for Dove haircare.
That sounds simple. It is not.
The definition depends on three decisions:
What counts as the brand. What lookback window is used. Which sales channels and loyalty identifiers are included.
Change any of those and the NTB number changes.
That is why NTB is useful, but also easy to misuse. It is not a magic number. It is a measurement rule. The rule has to match the buying cycle of the category and the commercial question the brand is trying to answer.
Why NTB matters in retail media
Brands grow by increasing penetration. They need more people buying them, not only more sales from the same heavy buyers.
Retail media is one of the few channels that can prove this with transaction data. A retailer can see who bought the brand before the campaign, who was exposed to the campaign and who bought after exposure.
That makes NTB more than a reporting metric. It becomes a budget argument.
If a campaign only reaches existing buyers, the report may still show sales. Existing buyers buy. Loyal shoppers keep buying. A campaign can attach itself to that behavior and claim credit it did not really create.
NTB forces the conversation away from easy attribution and toward growth.
The question changes from:
How many sales were associated with the campaign?
To:
How many new buyers did the campaign help create?
That is a better question for brand managers, trade marketers and finance teams.
How NTB is measured
The standard NTB calculation starts with campaign buyers, then checks their transaction history.
The simplest version:
NTB rate = New-to-Brand buyers divided by all campaign buyers.
Cost per NTB = campaign spend divided by New-to-Brand buyers.
NTB revenue = sales from New-to-Brand buyers.
Those metrics are useful because they separate acquisition from repeat purchase. They also help compare campaigns with different roles.
A loyalty campaign should not be judged mainly by NTB. Its job may be repeat purchase or basket expansion.
A launch campaign should care deeply about NTB. Its job is to create trial.
A category expansion campaign should care about whether the brand recruited shoppers from the category, not only whether existing buyers bought more.
The metric is only useful when the campaign objective is clear.
The lookback window problem
NTB depends on the lookback window.
A 12-month window is common because it is easy to explain. But categories do not all behave the same way.
Milk, beer, snacks and detergent do not share the same buying cycle. A shopper who has not bought beer in three months may be meaningfully inactive. A shopper who has not bought detergent in three months may simply not need detergent yet.
Using the same lookback window across every category creates false precision.
For high-frequency FMCG categories, a shorter window can make sense. For slower categories, a longer window may be needed. For seasonal categories, the window must account for seasonality.
This is where many NTB reports break. They apply a clean formula to messy buying behavior and pretend the answer is final.
It is not final. It is a definition.
Good measurement makes the definition visible.
A simple FMCG example
A brand launches a new protein cereal in a grocery retailer.
The media plan targets shoppers who buy breakfast products, health- oriented snacks and high-protein dairy alternatives. The campaign runs across onsite search, category pages and in-store screens near breakfast and health aisles.
After the campaign, the retailer reports:
Total campaign buyers.
Buyers who bought the cereal before. Buyers who had never bought the cereal brand in the lookback window. Repeat purchase after trial. Basket companions bought with the cereal.
The NTB number answers one question:
Did the campaign create new brand buyers?
If most sales came from existing brand buyers, the campaign may still have protected share or increased frequency. That can be valuable. But it is not acquisition.
If a large share came from shoppers who already bought adjacent breakfast products but had not bought this brand, the campaign has a stronger growth story.
That is the commercial value of NTB. It shows whether the media reached beyond the comfort zone.
NTB is not the same as incrementality
This is the mistake to avoid.
NTB tells you whether a buyer was new to the brand.
Incrementality tells you whether the campaign caused the purchase.
A shopper can be New-to-Brand and still not be incremental. They may have been about to buy the brand anyway because of a price promotion, a store display, distribution changes or seasonal demand.
That is why NTB should be read with a control group or baseline where possible.
Without a comparison, NTB is a classification.
With a comparison, NTB becomes stronger evidence.
The best version is incremental NTB: new buyers above what would have happened without the campaign.
That is a much harder number to get. It is also the number that deserves more trust.
How to use NTB in planning
Start with the campaign job.
If the job is acquisition, define the NTB target before launch. Decide the lookback window, the brand boundary, the eligible shopper pool and the measurement method.
Then decide which audiences can actually create new buyers.
Existing buyers are easy to convert. Category buyers who do not buy your brand are harder, but more valuable for growth. Adjacent category buyers may be even more interesting if the product has a clear usage occasion.
NTB should shape the audience plan, not only the report.
It should also shape the creative.
A message to an existing buyer can remind, repeat or trade up. A message to a New-to-Brand prospect has to create a reason to try. The buyer does not have the same memory, habit or confidence.
If the audience is different, the message should be different.
Common mistake
The common mistake is celebrating NTB without asking what kind of NTB it is.
A buyer can be new to the brand, new to the SKU, new to the category or simply newly visible in the retailer data.
Those are not the same.
New-to-SKU can be useful for product launches, but it does not always mean brand growth. A shopper may move from one SKU in the same brand family to another.
New-to-Brand is stronger, but still needs a clear brand definition.
New-to-Category is stronger again, but harder to prove.
If the report does not define the boundary, the metric can overclaim.
Good NTB reporting says exactly what is being counted.
What a good NTB report should include
A strong NTB report should show:
The lookback window. The brand or SKU boundary. NTB buyers. NTB rate. Cost per NTB. NTB revenue. Repeat behavior where enough time has passed. Incremental NTB where a control or baseline exists. Known data gaps, such as unidentified cash transactions.
That last point matters.
Retail data is powerful, but it is not perfect. Loyalty coverage, online identification, POS matching and store-level data quality all affect the read.
The more transparent the report, the more useful the metric.
FAQs
What does NTB mean in retail media?
NTB means New-to-Brand. It counts buyers who purchased a brand after not buying that brand during a defined lookback window.
Is NTB the same as new customer acquisition?
It is a form of acquisition measurement, but it depends on the retailer data and the lookback window. It should not be treated as causal proof unless it is compared with a control group or baseline.
What is a good NTB rate?
There is no universal good rate. It depends on the category, brand size, campaign objective, audience strategy and lookback window. A launch campaign should expect a different NTB profile than a loyalty campaign.
Why can NTB be misleading?
It can overclaim if the lookback window is wrong, if SKU switching is counted as acquisition or if the campaign is not compared with what would have happened anyway.
How should brands use NTB?
Use NTB before the campaign as an objective, not only after the campaign as a report line. Define the audience, window, brand boundary and proof method before spend starts.
Related reading
Retail Media Guide. Closed-Loop Measurement. Incrementality. iROAS. Category Entry Points.
Bottom line
NTB is useful because it asks whether retail media created new buyers, not just reportable sales.
But NTB is only trustworthy when the definition is clear. Set the lookback window, define the brand boundary, separate acquisition from repeat behavior and compare against a baseline where possible.
New-to-Brand is not the whole growth story.
It is the start of a serious one.
Related Reading
- Retail Media Learning Agenda: how each campaign should improve the next one
- Retail Media Optimization: what to improve after launch
- Retail Advertising Case Studies: what proof to look for before trusting results
- Retail Advertising Analytics: how to judge media when sales happen in stores
- Retail Media Dashboard: what buyers need to see before renewing spend
Ready to see how this works in practice?
Footprints AI helps brands and retailers measure what matters. See our customer success stories or get in touch to discuss your retail media strategy.
