Key takeaways: Sponsored TV is a self-service streaming TV ad format for brand-registered sellers with no minimum spend. It is worth testing when your listing already converts, you have a video that works without sound, and you can measure the result through branded search and new-to-brand orders rather than direct ACoS. It is not worth it as a first ad type, for a single low-priced product, or without a measurement plan.
Amazon Sponsored TV is worth it for brands that have already built the lower funnel and want to reach shoppers before they search. It runs your video on Amazon's streaming inventory, including Prime Video ad-supported content, Freevee, Twitch and Fire TV, and it is bought through the same Campaign Manager as your other ads with no minimum commitment. Whether it pays depends on three things: whether your listing can convert the traffic it eventually sends, whether your video creative fits a television screen, and whether you have a way to measure lift that does not rely on click attribution.
What Sponsored TV is and how it differs
Sponsored TV sits between Sponsored Brands video and Amazon DSP streaming TV. It uses the self-service console, bills on a CPM basis, and lets you target by audience and by content genre. There is no minimum spend, unlike DSP, and no requirement to work through an agency or managed service.
| Sponsored Brands video | Sponsored TV | Amazon DSP streaming TV | |
|---|---|---|---|
| Where it shows | Amazon search results | Streaming TV screens | Streaming TV, Amazon and third-party inventory |
| Shopper intent | Searching now | Watching TV | Watching TV or browsing |
| Buying model | Cost per click | Cost per thousand impressions | Cost per thousand impressions |
| Minimum spend | None | None | Typically a substantial monthly minimum |
| Targeting | Keywords and products | Audiences and content genres | Full audience and contextual set |
| Clickable | Yes | Remote-based interaction on some devices | Varies |
| Best measured by | ACoS, new-to-brand | Branded search lift, new-to-brand, detail page views | Full-funnel attribution, AMC |
The main difference from everything else in the console is intent. A shopper watching a streaming show did not ask to see your product. The ad has to create the interest, and the sale, if it comes, arrives later through a branded search or a direct visit. Judging it on the same seven-day click attribution as Sponsored Products misses most of what it does.
When it is worth testing
The listing converts already. Sponsored TV sends people to Amazon with a brand in mind. If the detail page or Store they find has a weak main image, thin reviews or a price out of step with the category, the awareness is spent on a page that cannot close. Fix conversion first; every upper-funnel dollar is worth more after.
The video works on a television. A 15 or 30-second product video that shows the product in use, in a real setting, with the brand name visible and legible from across a room, and that makes sense with the sound off. A listing video shot for a phone screen with on-screen text designed for close reading does not translate.
The product has a considered purchase cycle or repeat purchase. Categories where shoppers think for a while, or buy again, give the awareness time to turn into a search. A low-priced impulse product has a thinner path from screen to cart.
You have a measurement plan. Before the campaign starts, record baseline branded search volume from Brand Analytics, new-to-brand order share, and detail page views. Without a baseline, there is nothing to compare against.
When it is not
Not as a first ad type. Sponsored Products carries the sales, and Sponsored TV depends on it. Not for a brand with one product under $15 and no repeat purchase. Not with a budget so small that the impressions cannot reach a meaningful audience; a test needs enough reach to move branded search visibly. And not without Brand Registry, which it requires.
Running the test
Read this flowchart as text
- Brand Registry, listing converts, TV-ready video
- Record baselines: branded search, new-to-brand share, page views
- Run 4–6 weeks at a fixed budget. Change nothing else upper-funnel
- Branded search volume up vs baseline and vs seasonality?
- Creative or audience the likely cause?
- Re-cut video: product in use, brand visible, works muted
- Narrow to in-market and lifestyle audiences that match
- New-to-brand orders and detail page views up too?
- Keep. Set budget as a share of total and review quarterly
- Hold budget. Fix the landing page, then re-test
The "change nothing else upper-funnel" step is essential. If you launch Sponsored TV in the same month as a Sponsored Display awareness campaign and an influencer push, a rise in branded search cannot be attributed to any of them. Test one upper-funnel channel at a time, and account for seasonality by comparing against the same period last year where the data exists.
The creative decision is where most tests are won or lost. We test video the same way we test images, using evidence rather than opinion, and that process is B10 · How we test creative before we commit.
How it fits the rest of the account
Sponsored TV is a brand-building tool, and it belongs in the brand-building objective described in C5 · Building your brand on Amazon: a full-funnel presence measured on branded search growth and new-to-brand share, not on ACoS. In an account run for profit or for launch, it has no place yet. In an account where Sponsored Products is efficient, Sponsored Brands and Display are in their bands, and the goal is to grow the number of people who search for the brand by name, it is one of the two or three tools available. The other is DSP, and how the two compare is covered in When Is Amazon DSP Worth It? Budget, Audiences and Fit.
What it should cost
Amazon does not set a minimum. A test needs enough impressions to reach a meaningful share of the target audience several times over the test period, which puts a practical floor on a useful budget that depends on the audience size. As a guide, a test that cannot fund a few thousand dollars over four to six weeks is unlikely to move branded search enough to read. Beyond the test, a working share for upper-funnel spend in a brand-building account is usually a minority of total advertising, with the majority still in Sponsored Products.
Frequently asked questions
What is the minimum spend for Amazon Sponsored TV?
There is no minimum set by Amazon. The practical minimum is whatever it takes to reach your audience enough times to move branded search measurably, which depends on the size of that audience. A test too small to read is not a saving.
How do I measure Sponsored TV results?
Not on click ACoS. Record branded search volume from Brand Analytics, new-to-brand order share and detail page views before the campaign, run it for four to six weeks with no other upper-funnel changes, and compare. Amazon also reports view-based metrics in the console.
Do I need Brand Registry for Sponsored TV?
Yes. Sponsored TV, like Sponsored Brands and Sponsored Display, is available to brand-registered advertisers only.
Can I use my listing video for Sponsored TV?
Sometimes, but usually not as it is. A listing video is designed for a phone screen and often relies on small text and sound. A TV ad needs the product in use, the brand legible from a distance, and a message that works muted. A re-cut of existing footage is often enough.