LinkedIn CTV Ads Come to Amazon DSP: What B2B Advertisers Actually Get

B2B media buyers have spent two years trying to solve the same problem: streaming TV reaches the buying committee, but nothing in the CTV stack knows who on the couch is a VP of Engineering and who is their teenager. Every workaround has been a proxy. Household income tiers, firmographic overlays inferred from IP, contextual placement against business programming. All of them work a little, and none of them work the way LinkedIn targeting works.

On May 7, 2026, Amazon Ads and LinkedIn announced that LinkedIn's first-party professional signals can now be bought through Amazon DSP. That is a real change and it is worth understanding precisely, because most of the coverage got the mechanics wrong in a way that will cost you a planning cycle if you repeat it.

The short version: this is a buying-seat change rather than an inventory change. If your read of the announcement was "LinkedIn audiences on Prime Video," that is not what shipped, and a media plan built on that assumption falls apart at the first conversation with a rep.

What follows is the actual mechanic, which targeting facets are confirmed and which are not, how the three buying routes compare on measurement and minimums, and who should skip this entirely.

Quick answer

• What shipped. LinkedIn CTV Ads became buyable through Amazon DSP on May 7, 2026.
• The mechanic. LinkedIn packages approved professional signals from its member base, including job title, industry, and seniority, into a curated private marketplace deal inside Microsoft Curate, then delivers that Deal ID into your Amazon DSP seat.
• Where the inventory comes from. Microsoft Monetize, rather than Prime Video, Fire TV, Freevee, or Twitch. Amazon DSP is the buying interface.
• Why you would choose this route. It is the only one of the three that lets you run LinkedIn professional audiences and Amazon retail audiences in the same campaign.
• What you give up. A narrower targeting surface than Campaign Manager, and professional demographic reporting that arrives as a rep-produced wrap report rather than inside the DSP.
• Access. US targeting only, not self-serve, rep-gated, with a reported minimum of $50,000 over 30 days.
• Context. Amazon DSP was the second programmatic route LinkedIn opened, after The Trade Desk in March 2026, with Microsoft Advertising following a week later.

The mechanic, and why the coverage got it wrong

Amazon's announcement says the solution "brings LinkedIn's first-party audience signals from more than one billion members, including job title, industry, and seniority, to streaming TV inventory through Microsoft Monetize."

LinkedIn's version says the same thing: marketers can use LinkedIn targeting "and reach buyers across premium streaming inventory through Microsoft Monetize."

Microsoft Monetize is Microsoft's sell-side platform, the former Xandr and AppNexus stack. It supplies the CTV inventory. LinkedIn builds a curated audience segment from approved professional facets, packages it with that inventory as a private marketplace deal inside Microsoft Curate, and hands the resulting Deal ID to your DSP seat. You then run it in Amazon DSP the way you would run any other PMP line.

The DSP is interchangeable. The supply path is not.

That distinction is the whole story, and it is why the "Amazon does a LinkedIn deal" framing misleads. Amazon DSP is one of three places you can now buy the same LinkedIn CTV product. LinkedIn opened the first programmatic route through The Trade Desk on March 18, 2026, described in LinkedIn's own words as "powered exclusively by Microsoft Monetize." Amazon DSP followed on May 7. Microsoft Advertising became the third on May 14.

Three DSPs, one SSP, one audience graph. What LinkedIn is building is a portable professional identity layer that any buying platform can rent. The DSP you pick changes your workflow, your measurement, and which of your other audiences sit in the same campaign. It does not change the inventory or the targeting.

What is not involved

Three things get asserted in secondary coverage and are not supported by anything either company published:

• No clean room. This is not an Amazon Marketing Cloud integration. If you have been planning an AMC workflow around LinkedIn signals, there is nothing here to build on.
• No identity resolution vendor. No LiveRamp match, no third-party graph. The audience never leaves Microsoft-controlled infrastructure before it reaches your seat, encoded inside the Deal ID.
• No device-graph or IP-household inference. The signals are member-declared profile attributes rather than behavioral inference or licensed firmographics.

For a buyer, that last point cuts both ways. The precision is real because the data is declared rather than modeled. The transparency is limited because you never see the individual match, only the segment behind the deal.

What you can actually target

Both May 7 announcements name exactly three facets: job title, industry, and seniority. Microsoft Advertising's May 14 update publicly added job function and company category to that list.

Everything else is unconfirmed. Company size, member skills, years of experience, and named-account or ABM targeting have not been confirmed as available through the deal mechanism by either company. That is a meaningful gap if your LinkedIn program runs on account lists. Campaign Manager gives you the full targeting surface. The DSP routes appear to give you a subset, and neither company has published where the line falls. Ask your rep before you scope a plan around it.

How the buying routes compare

If you already run LinkedIn CTV, the useful question is whether the route change is worth what you give up in measurement. The figures below on minimums and timelines come from LinkedIn's Help Center documentation as reported by trade press rather than from a public announcement page, so confirm them with your rep before you commit budget.

The row that decides it for most buyers is the second-to-last one. Amazon's own framing is that you can activate LinkedIn audiences on CTV "alongside Amazon audiences within the same campaign, without fragmenting their media buy." If you are already running Amazon DSP with retail audiences and you want a professional cut in the same plan, this route is the only one that gives you that. If you do not run Amazon DSP, the LinkedIn product is the same product elsewhere, and Campaign Manager gives you better reporting.

Who LinkedIn CTV on Amazon DSP is for, and who should skip it

Amazon and LinkedIn both left this vague, so it is worth being direct about it.

It fits an advertiser already spending in Amazon DSP who sells to a professional buyer, wants CTV reach against the buying committee, and can clear a reported $50,000 over 30 days on a single route. Software, financial services, professional services, and B2B hardware are the obvious shapes. So is any endemic seller with a genuine business-buyer segment. If Amazon Business is a real channel for you, this is the first CTV product that can target the buyer rather than the household. We covered Amazon Business pricing strategy separately, and the targeting logic here rhymes with it.

It does not fit a consumer brand looking for an incremental reach source. LinkedIn data describes professional identity rather than purchase intent. A VP of Marketing is not more likely to buy your protein powder because LinkedIn knows their job title, and you will pay a data premium for the privilege. For consumer reach, Amazon's own retail signals and the in-market and purchase-based DSP audiences are the better instrument.

It probably does not fit yet if your program depends on named-account targeting, non-US markets, or clean attribution back to pipeline. All three are open questions on this route.

What the Amazon and LinkedIn announcements left out

• No launch partners, no case studies, no results. LinkedIn's public performance claims for CTV Ads, that it reaches B2B audiences 2.2x more effectively than other CTV platforms and 4.3x more effectively than linear TV, carry a footnote reading "based on early results; as measured by iSpot." The Salesforce and ServiceNow case studies that circulate alongside this news are from the Campaign Manager product in 2025, not from the Amazon DSP route. Do not let a rep present them as evidence for this integration.
• No pricing. Neither company published CPM ranges for the deal.
• No clarity on measurement for blended campaigns. If you run LinkedIn audiences and Amazon retail audiences in one campaign, nothing published explains how you separate their contribution. Our piece on Amazon DSP attribution covers the general problem, and this integration does not solve it.
• No status word. Neither company said beta or GA. LinkedIn says it is available for US targeting, contact your representative. Trade reporting describes it as in testing with limited managed accounts. Plan accordingly.

If you are already committed to B2B CTV

Suppose you have a CTV budget approved and a buying committee to reach. The routes available to you, in rough order of setup cost:

1. LinkedIn Campaign Manager directly. Self-serve, full targeting, native professional demographic reporting. Lowest friction and the best measurement. You give up the ability to blend with anything else in your stack.
2. Amazon DSP with the LinkedIn deal. Worth it specifically when you already run Amazon DSP and the blend with retail audiences is the point. Accept that professional reporting arrives as a rep-produced wrap report on a lag.
3. The Trade Desk with the LinkedIn deal. Same inventory, same facets, reported to retain iSpot measurement. Pick this if measurement matters more than the Amazon audience blend.
4. Contextual and firmographic overlays without LinkedIn data. Bombora contextual segments, ZoomInfo-based products, or straightforward contextual placement against business programming. Less precise, materially cheaper, and worth benchmarking against a LinkedIn deal rather than assuming the premium pays for itself.

Whichever route you pick, plan the measurement before the media. CTV completes at very high rates and produces almost no clicks, so a last-touch model will systematically undercount it. Decide up front whether you are measuring with brand lift, with a geo holdout, or with a pipeline-lagged read, and hold that decision for the full flight.

Where this lands in an Amazon DSP operation

For a team already running Amazon DSP, the practical effect is one more deal line in a seat that already carries retail audiences, frequency caps, and a CTV mix. That is not free. Every additional supply path with its own reporting surface is another thing to reconcile at month end, and the LinkedIn route deliberately hides the professional-demographic view behind a rep deliverable that arrives on someone else's schedule.

Which puts a premium on the rest of the seat being explainable. When a blended campaign underperforms and one of its lines reports late through a third party, the only way to say anything useful about it is to have already ruled out everything else. That is a tooling problem before it is an analysis problem.

Qinetix is our execution layer, and it handles the half of the seat you do control. Bids and budgets move continuously inside zones you set with a floor and a ceiling, DSP extends reach past the search auction, and every move lands in a change log with the rule that caused it. For a seat carrying a new opaque deal line, that log is the difference between answering "did the retail audiences move" in a minute and spending a morning on it.

Qore is where the reconciliation itself gets written down. You describe the monthly seat review once, what you check, in what order, and what counts as off-plan for each line, and it runs on a schedule with the same logic every month, so this month's read is comparable to last month's rather than a fresh opinion. Actions start gated on your approval, and the gate loosens as a routine earns it. Qore is in open beta and you can sign up here. Our Amazon DSP best practices guide covers the operating side in detail.

What none of that fixes is the LinkedIn line itself. The professional-demographic view is produced by LinkedIn and delivered by a representative rather than exposed through the DSP, so no third-party platform can pull it into your reporting, ours included, and any vendor telling you otherwise has not read the documentation. What you can do is make the rest of the seat clean enough that the LinkedIn line is the only unknown in the reconciliation.

A real capability with a narrower shape than the headlines

LinkedIn CTV Ads through Amazon DSP is worth understanding for what it is. You are buying LinkedIn's declared professional signals against Microsoft Monetize inventory, executed in an Amazon seat. The reason to choose it over the other two programmatic routes is the ability to run professional and retail audiences in one campaign. The reason to hesitate is that the targeting surface is narrower than Campaign Manager, the measurement is thinner, and neither company has published a single result from the integration.

If you already run Amazon DSP and sell to a professional buyer, ask your rep for the Deal ID, budget a test rather than a program, and set the measurement design before the first impression serves. If you do not run Amazon DSP, buy the same product where the reporting is better.

The bottom line for B2B media buyers

• The inventory is Microsoft Monetize, not Prime Video. Correct anyone in the planning meeting who says otherwise, because the rest of the plan depends on it.
• Three facets are confirmed: job title, industry, seniority. Named-account, company size, and skills targeting are not. Ask your rep before scoping anything around ABM.
• Pick the route by what you need most. Campaign Manager for targeting depth and native reporting, Amazon DSP for the retail-audience blend, The Trade Desk for iSpot measurement.
• Budget a test rather than a program. Neither company has published a single result from this integration, and the reported minimum is $50,000 over 30 days.
• Refuse the borrowed case studies. The Salesforce and ServiceNow examples come from the 2025 Campaign Manager product, not this route.
• Design the measurement before the media. CTV completes high and clicks almost never, so last-touch attribution will undercount it every time.
• Get the rest of the seat under a standard first. One opaque line is manageable. One opaque line in a seat nobody can explain is a month-end you will lose.
• Consumer brands should skip it. These are professional-identity signals at a data premium, and retail purchase audiences are the more direct instrument.