Amazon Marketing Cloud Guide 2026: What It Is & When to Use

TL;DR

Amazon Marketing Cloud (AMC) is a free, privacy-safe clean room built on AWS that lets advertisers analyze cross-campaign customer journeys and build custom audiences using pseudonymized Amazon data. Originally restricted to DSP advertisers, AMC opened access to all sponsored ad advertisers in late 2025. It answers questions standard reporting cannot, like which ad touchpoints actually drive incremental sales, but it requires SQL knowledge or an AI assistant to operate.

Amazon sellers hear “AMC” tossed around in agency pitches and conference talks, but most still don’t have a clear picture of what it actually does or whether they need it. That confusion is understandable. Amazon Marketing Cloud sits at the intersection of advertising analytics, data privacy, and cloud infrastructure, three topics that each carry their own jargon.

This guide cuts through the noise. It explains what AMC is, how the mechanics work, what changed in 2025 and 2026, and when your brand should seriously consider using it.

If you’re evaluating whether your Amazon advertising strategy needs this level of depth, a free brand audit is a good starting point.

What Is Amazon Marketing Cloud?

Amazon Marketing Cloud is a secure, cloud-based clean room solution that lets advertisers run analytics and build audiences across pseudonymized signals from Amazon Ads and their own first-party data. Amazon launched it in 2021, and it runs on AWS Clean Rooms.

The canonical definition from Amazon Ads describes it as a space where advertisers can “perform analytics and build audiences across pseudonymized signals, including Amazon Ads signals as well as their own inputs.”

But what does “clean room” actually mean in plain language?

The Clean Room Concept, Simplified

Think of a secure reading room at a library. You can study the documents inside. You can take notes on patterns you observe. But you cannot photocopy individual pages or take any single document home. That’s essentially what a data clean room does with customer data.

Amazon assigns a unique pseudonymized user ID to every shopper who interacts with an ad (impression or click). Inside AMC, advertisers can follow these anonymized IDs across their entire journey, from a DSP display impression to a Sponsored Products click to a purchase. But no individual user data ever leaves the room. Every query output must aggregate at least 100 users before Amazon returns results, protecting customer privacy while still providing actionable patterns.

How Amazon Marketing Cloud Works

Understanding the mechanics requires knowing three things: what goes in, how you query it, and what comes out.

Data Inputs

AMC unifies signals from multiple sources:

Amazon Ads signals: Sponsored Products, Sponsored Brands, Sponsored Display, Sponsored TV, and Amazon DSP campaign data
Advertiser first-party data: CRM lists, purchase histories, website behavior (uploaded by the advertiser)
Third-party signals: Data from onboarded providers

This combination is what makes AMC different from standard Amazon Ads reporting. The Ads Console shows you campaign-level metrics. AMC shows you user-journey-level patterns across campaigns and data sources. For a deeper look at how campaign data flows through AMC, see our guide on using AMC for campaign performance.

Query Mechanism

Historically, AMC required SQL queries to extract insights. You write (or modify) a SQL query, run it against the pseudonymized dataset, and receive aggregate results. This was the primary barrier to adoption for most brands.

In 2025 and 2026, Amazon introduced AI-powered alternatives. The AMC AI Audience Generator lets advertisers describe their ideal audience in plain English, and AMC generates the working SQL query automatically. By March 2026, Amazon enhanced its Ads Agent with in-editor AI controls accessible via right-click, making it faster to edit, explain, and fix queries without switching contexts.

Outputs

AMC returns two categories of results:

Aggregate analytics: Cross-campaign performance reports, path-to-purchase maps, incrementality measurements, frequency distributions
Custom audiences: Segments built from engagement records, conversion events, and behavioral patterns that can be activated directly for media buying across sponsored ads, video ads, and display ads

The privacy threshold is firm. All aggregations must contain a minimum of 100 users. No exceptions.

Who Can Access AMC (the 2026 Reality)

This is where most existing guides are outdated, and it’s the single biggest change to understand.

The Old Model (Pre-September 2025)

Before September 2025, AMC access required an executed Amazon DSP Master Service Agreement (MSA), active or recently active DSP campaigns, and a technical resource comfortable with SQL. This effectively limited AMC to larger brands spending five or six figures monthly on Amazon advertising.

The New Model (September 2025 Onward)

Amazon opened AMC access to anyone running Sponsored Products, Sponsored Display, Sponsored Brands, or Sponsored TV campaigns. No DSP requirement. No separate registration needed.

According to Marketing Dive’s coverage, Amazon gave all advertisers running sponsored ad campaigns access to its clean room service, a significant shift from the previous registration or partner requirement. This matters because nearly 48% of marketers and agencies worldwide had been avoiding clean room technology due to budget constraints.

Cost: AMC is free to eligible advertisers. You don’t pay for access or for running queries.

What you still need: SQL knowledge (or willingness to use the AI tools), an understanding of what questions to ask, and for API-level access, an AWS account.

Key Use Cases for Amazon Marketing Cloud

AMC’s value comes from answering questions that standard Amazon Ads reporting simply cannot address. Here are the use cases that matter most.

Path-to-Purchase Analysis

Standard reporting assigns credit to the last click. AMC lets you map the actual customer sequence, for example: DSP view on day 1, Sponsored Brands click on day 14, Sponsored Products purchase on day 30. This reveals which upper-funnel touchpoints are creating demand that last-click attribution ignores entirely.

Practitioners report that running path-to-purchase templates with extended lookback windows can quantify sales that never touched branded search before converting. Those represent true incremental discovery, the kind of insight that reshapes how you allocate budget across your campaign architecture.

Incrementality Measurement

This is arguably AMC’s most financially impactful use case. It answers: “Would these customers have bought anyway without this ad?”

For brands at $10M or more in annual revenue, even a 15-point reduction in CPC spend on cannibalized keywords frees up hundreds of thousands in annual budget. That money can be redirected to genuine incremental demand generation. But the cannibalization only becomes visible through AMC incrementality analysis, not through ACOS or ROAS reports in the Ads Console.

This connects directly to scaling sales without losing margin, something standard reporting makes dangerously easy to get wrong.

New-to-Brand Analysis for Sponsored Products

Amazon has offered new-to-brand (NTB) metrics for years, but only for display, video, and Sponsored Brands campaigns. Sponsored Products, the workhorse of most Amazon ad strategies, was excluded. With AMC, you can calculate NTB purchases for Sponsored Products campaigns through a custom query. This closes a significant blind spot.

Custom Audience Building and Activation

AMC enables brands to construct precise audience segments and activate them for DSP campaigns:

Buyers who purchased in the last 90 days
Buyers of a specific product line
Shoppers who viewed but didn’t convert
High-frequency repeat purchasers

Each segment gets built inside AMC and pushed directly into campaign targeting.

Frequency Optimization and Campaign Overlap

When multiple campaigns target similar shoppers, brands overpay for the same audience without knowing it. AMC identifies overlap across tactics so spend can shift toward more incremental audiences. This is particularly valuable for understanding the interaction between Sponsored Ads and DSP, a connection that the standard Ads Console treats as completely separate.

If you’ve noticed PPC clicks spiking without corresponding orders, over-frequency across overlapping campaigns is a common culprit that AMC can diagnose.

Cross-Device Purchase Tracking

Amazon assigns unified user IDs that persist across devices. AMC lets you track cross-device purchases from ad exposure to conversion, revealing paths that device-siloed reporting misses entirely.

Prime Video Viewership Signals (2026)

As of November 2025, AMC includes Prime Video viewership signals in open beta for US, Canada, Japan, and Australia. Advertisers can analyze content engagement patterns across Prime Video for their customers, connecting streaming behavior to shopping behavior for the first time.

AMC vs. Standard Amazon Ads Reporting

This is the comparison most sellers need before deciding whether AMC is worth the effort.

Capability	Ads Console	Amazon Marketing Cloud
Campaign-level metrics (clicks, spend, ACOS)	Yes	Yes
Cross-campaign customer journeys	No	Yes
Multi-touch attribution	No (last-click only)	Yes
New-to-brand for Sponsored Products	No	Yes (custom query)
Custom audience creation	Limited	Full flexibility
Overlap/frequency across campaign types	No	Yes
First-party data integration	No	Yes
Lookback window	60-90 days typical	Up to 25 months
Real-time data	Near real-time	Latency of several hours
SQL or technical skill required	No	Yes (or AI tools)

The short version: if your strategic questions are limited to “what’s my ACOS?” and “which keywords convert?”, the Ads Console is sufficient. If you need to understand how campaigns interact, which audiences are incremental, and how upper-funnel spend influences bottom-funnel conversions, AMC is where those answers live.

For brands noticing discrepancies between ad platform conversions and backend orders, AMC provides a reconciliation layer that standard reporting lacks.

What AMC Cannot Do

Honest boundaries matter. AMC is powerful, but it has real limitations.

No product research. AMC reports on advertising interactions, not organic search volume, competitor sales, BSR trends, or keyword gaps. It cannot tell you which product to launch next.

Data latency. AMC data can lag by several hours, making real-time or same-day optimization impractical. It’s a strategic analysis tool, not a bid management tool.

Not a campaign execution platform. You don’t build or manage campaigns inside AMC. It generates insights and audiences. You still execute in the Ads Console or DSP.

Strategy still required. As one practitioner put it bluntly on LinkedIn: “Your blended TACOS looks solid at 15%, but your EBITDA is eroding month over month. You’re flying blind on incrementality, burning 20% of your media budget on audiences that would convert anyway.” AMC surfaces that problem. It doesn’t solve it. Brands still need clear business questions, clean inputs, and a plan for how insights will change actual media decisions.

Complexity for beginners. AMC holds a 4.4 out of 5 rating on G2 from 73+ reviews, but users consistently note the learning curve. Sellers in Amazon’s own forums describe confusion about setup, APIs, and documentation. One Seller Central thread captures it well: “We really want to start using this, but are confused about how to set it up, the APIs involved, etc.”

Major AMC Updates in 2025 and 2026

Amazon has been aggressively expanding AMC’s capabilities. Here’s what changed:

AI-powered query generation (2025): The AMC AI Audience Generator allows advertisers to create audiences using natural language prompts instead of SQL. Describe what you want, and AMC writes the query.

Enhanced Ads Agent (March 2026): AI-powered controls are now embedded directly in the query editor via right-click menus. You can edit, explain, and fix SQL without switching contexts.

25-month lookback window (November 2025): Amazon expanded the ad traffic lookback from 13 months to 25 months, giving advertisers access to more than two complete calendar years of historical data.

Prime Video signals (November 2025): Open beta connecting streaming viewership behavior to advertising analytics.

Access expansion (September 2025, January 2026): The elimination of DSP requirements, opening AMC to all sponsored ad advertisers.

The overall direction is clear. Amazon is making AMC more accessible, more automated, and more central to full-funnel measurement. The days of AMC being exclusively a tool for technical teams writing raw SQL are ending.

When Your Brand Should Consider AMC

Not every brand needs AMC right now. Here’s a practical framework.

You probably don’t need AMC yet if:

You spend less than $5,000 per month on Amazon Ads
You only run Sponsored Products campaigns
Your strategic questions are limited to keyword-level performance
You don’t have DSP campaigns or plans to run them

AMC becomes valuable when:

You run multiple ad types (SP, SB, SD, DSP) and can’t see how they interact
You suspect you’re paying for conversions that would have happened organically
You need new-to-brand data for Sponsored Products
Your annual Amazon revenue exceeds $1M and you’re optimizing for contribution margin, not just ACOS
You have first-party data (CRM, website) you want to match against Amazon shopping signals

On the agency question: Most brands without a dedicated analytics team will benefit from working with an AMC-experienced partner, at least initially. The setup, query design, and insight translation require skills that sit outside typical Amazon seller operations.

If your brand is running multi-campaign Amazon advertising and you want to understand whether AMC fits your growth strategy, EZCommerce’s Amazon services include AMC strategy as part of the advanced advertising toolkit.

Related Terms

Data clean room: A secure environment where multiple parties can analyze shared data without exposing individual records.

Amazon DSP (Demand Side Platform): Amazon’s programmatic display and video advertising system, historically the gateway to AMC access.

Pseudonymized signals: Data where personal identifiers have been replaced with artificial IDs, allowing analysis without exposing individual identities.

SQL query: A structured command used to retrieve and analyze data from databases. AMC’s primary interface, now supplemented by AI tools.

Incrementality: The measurement of whether an ad caused a conversion that wouldn’t have happened otherwise.

New-to-brand (NTB): A metric tracking purchases from customers who haven’t bought from your brand in a defined lookback period.

TACOS (Total Advertising Cost of Sales): The ratio of total ad spend to total revenue (organic + paid). AMC insights feed directly into better TACOS management by identifying where ad spend is genuinely incremental.

First-party data: Information collected directly by a brand from its own customers (email lists, purchase histories, website behavior).

AMC Certification: A free credential issued by Amazon Ads Academy that validates an individual’s ability to use AMC’s UI, write SQL queries, and generate advanced advertising performance reports.

Ready to see where your Amazon advertising strategy has gaps? Request a free brand audit that includes an assessment of your campaign structure, reporting blind spots, and readiness for advanced tools like AMC.

Frequently Asked Questions

Is Amazon Marketing Cloud free?

Yes. AMC is available at no cost to eligible advertisers. There are no fees for access or for running queries. The cost is in the expertise needed to write queries and interpret results, not in the tool itself.

Do I still need Amazon DSP to use AMC?

No. As of September 2025, any advertiser running Sponsored Products, Sponsored Brands, Sponsored Display, or Sponsored TV campaigns has access to AMC. DSP is no longer a prerequisite, though brands running DSP campaigns get additional signal depth.

Do I need to know SQL to use Amazon Marketing Cloud?

Not anymore, strictly speaking. Amazon’s AI Audience Generator and Ads Agent let you create audiences and run queries using natural language prompts. That said, SQL knowledge still gives you far more flexibility and control over custom analyses.

How is AMC different from the regular Amazon Ads Console?

The Ads Console shows campaign-level, last-click metrics. AMC shows user-journey-level, multi-touch analytics across campaigns. AMC can answer questions like “how do my DSP impressions influence Sponsored Products conversions?” that the Ads Console simply cannot.

What is the lookback window in AMC?

As of November 2025, AMC offers a 25-month lookback window for ad traffic data, expanded from the previous 13 months. This gives advertisers more than two full calendar years of historical data for trend analysis and seasonality comparisons.

Can AMC help me find new products to sell?

No. AMC reports on advertising interactions and customer journeys. It does not provide organic search volume, competitor sales data, BSR trends, or keyword gap analysis. It is an advertising analytics and audience tool, not a product research tool.

How long does it take to get data from AMC?

AMC data has a latency of up to several hours. It is not a real-time optimization tool. Think of it as a strategic analysis platform for weekly or monthly planning, not for intraday bid adjustments.

Should I hire an agency to manage AMC for my brand?

It depends on your internal capabilities. If you have a data analyst comfortable with SQL and Amazon’s advertising ecosystem, self-service is feasible. If not, working with an experienced partner for initial setup, query design, and insight translation usually delivers faster, more reliable results. Many brands start with a partner and gradually build internal capability as they learn which queries and workflows matter most.