Can You Advertise on Netflix Through Amazon DSP?

What actually changed, and when

Until late 2025, buying a Netflix ad meant going through Netflix's own advertising sales team or its Microsoft-built ad platform. That changed on 10 September 2025, when Amazon Ads and Netflix announced that Netflix's ad-supported inventory would become directly buyable inside Amazon DSP, with the integration rolling out through the fourth quarter of 2025. If your DSP seat was already active before that date, no new contract or onboarding step is required to see Netflix as a supply source — it shows up alongside Prime Video, Disney's inventory and everything else Amazon DSP already reaches.

The practical shift is less about access and more about planning. Before the integration, a brand running Amazon DSP and a brand running Netflix ads were managing two separate buys, two separate reporting systems and two separate frequency caps on the same household. Now that can be one line item inside one plan.

Which markets are actually live

Amazon's own announcement names eleven initial markets: the United States, United Kingdom, France, Spain, Mexico, Canada, Japan, Brazil, Italy, Germany and Australia. The integration is publicly described as covering roughly a dozen global markets in total, and South Korea is called out specifically as a market where Netflix runs an ads tier that Amazon DSP does not yet serve — a useful reminder that "Netflix is on Amazon DSP" is not the same claim as "Netflix is on Amazon DSP everywhere Netflix sells ads."

Check the market list in your own DSP console before you build a global plan around Netflix. A campaign built for a market where the integration has not landed yet will simply not deliver Netflix impressions, and a thin delivery report is easy to misdiagnose as a targeting problem when it is actually a market-availability problem.

What you are buying, and what you are not

Netflix through Amazon DSP is programmatic access to Netflix's ad-supported inventory, planned and measured alongside your other Amazon DSP line items, with Amazon's clean-room and audience tooling layered over the buy. It is not a shortcut to Netflix's own upfront or direct-sold premium placements — those still run through Netflix's own sales relationships. What you gain is the ability to include Netflix in a single reach and frequency plan alongside Prime Video, Disney's DRAX inventory and everything else your DSP seat already touches, instead of stitching three vendor reports together by hand.

Amazon's own DSP audience layer travels with the buy. Because targeting rides on Amazon's shopping, browsing and streaming signals rather than on Netflix's own viewing data alone, you can point a Netflix line item at an audience defined by what people looked at or bought on Amazon last week — something Netflix's own ad platform, working from viewing history alone, cannot construct on its own.

A worked example

Say you set aside $8,000 of a monthly streaming budget specifically for Netflix. Premium ad-supported streaming inventory runs in the industry-reported range of roughly $20–$50 CPM, so at a representative $30 CPM that buys about 266,000 impressions for the month — before any frequency cap trims delivery to unique households. For comparison, across 27 advertisers in our own Amazon DSP book over a 31-day window this summer, all streaming and connected-TV inventory blended to a $15.51 CPM, well under the premium end of that range, because a blended figure mixes cheaper ad-supported placements in with pricier premium ones. Build your Netflix-specific budget off the premium end of the range, not off a blended average that includes inventory Netflix is not part of.

What to do when it does not deliver, or does not convert

If a Netflix line item is not spending, check market eligibility and bid floor first — new supply sources are frequently under-bid relative to more established inventory in the same campaign, so the campaign quietly favors Prime Video or Fire TV instead. If it is spending and not converting on a last-click basis, that is closer to expected than broken: Netflix inventory sits at the top of the funnel, viewers are not clicking a streaming ad mid-episode, and the honest read comes from branded search movement and Amazon Marketing Cloud rather than from a conversions column that was never going to move much on a non-clickable format.

Also check that Netflix is not sitting inside a blended streaming line item with Prime Video and Fire TV. If the three are lumped together, a slow Netflix ramp gets masked by faster-delivering Amazon-owned inventory, and you never actually learn whether Netflix specifically is working — you only learn whether the average is acceptable.

The common mistake

The mistake we see most is a brand moving its entire streaming budget onto Netflix the week the integration is announced, on the assumption that newer inventory must be better inventory. Netflix is one more supply source in a DSP that already reaches Prime Video, Disney's DRAX bundle, Fire TV, Twitch and third-party publishers — it is additive, not a replacement, and it earns its place in a plan the same way any other supply source does: by testing it against a target and reading the result honestly, including when that result is that a $30 CPM environment does not clear the bar a $15 environment did. reMKTR builds Netflix into an existing Amazon DSP plan as one line among several rather than as a standalone campaign, precisely so it gets judged against the rest of the book instead of in isolation.