Why is Black Friday won in Q1, not Q4?
Top-quartile brands don't switch on upper-funnel spend for peak, they hold it above 20% all year round. Here's the client data behind why Q4 is won early.
All
Learn More
Cross-channel halo measurement is how a retail brand estimates the sales its advertising drives on channels the ad can’t track, such as Amazon, TikTok Shop and retailers like Sephora or Ulta. It adds those sales back into each marketing channel’s results, so budget decisions reflect everything the ad influenced, not just what one dashboard can see.
A brand cuts its TikTok budget because website sales look weak. A month later, Amazon sales dip too, and nobody connects the two. Each report was right about its own patch and wrong about the whole picture. Halo measurement exists to fix that.
What is the halo effect?
The halo effect is when advertising in one place lifts sales somewhere else. Think of an ad as a shop window and the sale as the till. If customers pay at a different till, the window looks like it isn’t working.
Picture someone scrolling past a TikTok ad for a moisturizer. They don’t click. Two days later they remember the name, search for it on Amazon and buy it. TikTok reports no sale. Amazon sees a shopper arriving from search. The ad did its job, but no report connects it to the sale.
Marketing channels and sales channels are no longer the same thing
Three terms matter here:
- Marketing channel: where the ad runs, such as TikTok, Meta, YouTube, Google or a retail media network.
- Sales channel: where the purchase happens, such as the brand’s website, Amazon, TikTok Shop, Walmart, Target, Sephora, Ulta or a store.
- Retail media network: the advertising arm of a retailer, such as Amazon Ads or Walmart Connect. It sells ad space on the retailer’s own site and app, so it works as both a marketing channel and a sales channel.
Not long ago the path between them was short and fairly linear. A shopper saw an ad, clicked, and bought on the brand’s website or in a store, so the ad and the sale were easy to connect. Today, many brands sell on their own site, on Amazon, on TikTok Shop and through retailers, often all at once. Each of these systems keeps its own records and rarely shares them. When a campaign prompts a sale somewhere else, the credit often doesn’t make it back to the channel that earned it, so that channel looks weaker than it is.
With retail media networks, the ad and the sale can even happen on the same platform, which blurs the line between a marketing channel and a sales channel. Amazon, TikTok, Walmart, Target, Sephora and Ulta now sit in the middle of the journey, and direct-to-consumer (DTC) sales through a brand’s own site are only one destination among many.
The result is a web of links, not a line. Every marketing channel can lead to every sales channel, and most reports only follow the links that happen to land on their own doorstep.
Where does halo land?
- Marketplaces, such as Amazon.
- Retailer sites and apps, such as Walmart, Target, Sephora or Ulta.
- Social shops, such as TikTok Shop.
- Physical stores. These can be affected too, but they are usually measured differently. See our lesson on measuring marketing impact beyond your DTC website.
Halo also runs between retailers. Circana found that nearly 16% of media-driven sales associated with Best Buy’s ad network occurred at competing retailers. An ad in one store can fill the till in another.
Why do standard reports miss it?
Each report measures its own ecosystem, and that is by design.
- Ad platforms see sales on sites they can track, and usually only after a click.
- Amazon Attribution, Amazon’s measurement tool for off-Amazon ads, follows tagged clicks for 14 days. A shopper who sees an ad and searches later is not credited.
- Retail media networks report sales inside their own store. The Interactive Advertising Bureau (IAB) and Media Rating Council (MRC) retail media guidelines note that results measured within a retailer’s own walls may miss the wider market impact.
No shared identifier follows a shopper from a TikTok ad to an Amazon checkout. Halo measurement works from patterns in the totals instead of tracking individuals.
Retailers feel the pressure. NIQ and Unlimitail (June 2026) report that 67% of chief marketing officers plan to increase retail media spend, but only 53% think networks’ measurement is adequate.
How big is the effect?
It varies by brand, category and channel mix, so treat any single number as a range.
- Peer-reviewed research: Zenetti and Pauwels (Journal of Interactive Marketing, 2024) found that off-site purchases account for 16% to 44% of the total effect of on-site retail media ads at one large retailer.
- Modeled data: Fospha’s analysis found that 42% of Amazon sales were influenced by non-Amazon media (August 2025). “Influenced” means the pattern shows up in the data. It is not proof of cause.
How do you measure it?
There are three families of method. Each answers a different question.
None of these replaces the others. Signals point to where halo might be, a test sizes it for one channel, and a model keeps tracking it between tests.
You can spot early signs in your own data:
- Amazon or retailer sales rising in the days after a burst of social spend.
- Branded search climbing when a video campaign launches.
- A channel that looks weak on your website but whose pause is followed by a dip elsewhere.
What halo measurement is not
- It is not the same as incrementality. Halo describes where an ad’s effect lands. Incrementality asks whether the ad caused the sale. A strong halo estimate should answer both.
- It does not make every channel look better. A branded search ad might mostly capture Amazon shoppers who would have bought anyway. Halo measurement shows the full picture, including channels that deserve less credit.
.png)
%20and%20how%20does%20modern%20MMM%20work%3F.png)
