Ask most sellers what Amazon Marketing Cloud is and you'll get a shrug, or a vague answer about "some DSP reporting thing." That's fair — Amazon has done little to explain it in plain language, and the interface assumes you already know what a clean room is and can write SQL. But underneath the jargon, AMC answers questions that standard Seller Central and Advertising Console reports simply cannot: what actually happened, in order, across every ad a specific shopper saw before they bought — or didn't.
For brands spending seriously on Amazon Ads across Sponsored Products, Sponsored Brands, Sponsored Display, and DSP, that gap matters. This guide explains what AMC actually is, who can use it, what it can tell you that nothing else can, and — honestly — whether it's worth the effort for your brand right now.
The short version: Amazon Marketing Cloud is a secure data environment where you can query pseudonymised, event-level signals from your own Amazon Ads campaigns — impressions, clicks, purchases, and more — to answer cross-channel and cross-format questions that standard reporting bundles into simple totals. It's powerful, but it's built for advertisers with real DSP or multi-format spend, not for a seller running three Sponsored Products campaigns.
What Is Amazon Marketing Cloud?
AMC is a "clean room" — a secure environment where advertising event data can be analysed without either party (you or Amazon) exposing raw, identifiable customer information. Amazon feeds pseudonymised, user-level signals into your AMC instance: every impression, click, cart add, and purchase tied to your Sponsored Products, Sponsored Brands, Sponsored Display, and DSP campaigns, plus (if you opt in and meet eligibility) your own first-party customer data.
You then query that combined dataset using SQL, either writing your own queries or using Amazon's library of pre-built templates in Query Studio. The output is always aggregated above a minimum threshold of users — small enough result sets get suppressed automatically, which is how Amazon keeps the environment privacy-safe while still letting you see real patterns.
The key difference from everything else in Seller Central or the Advertising Console: those reports show you campaign-level or keyword-level totals. AMC shows you the sequence — which ad a shopper saw first, what they saw next, and whether that combination led to a purchase, days or weeks later, potentially across formats you'd normally analyse in separate dashboards.
Who Can Actually Access AMC
Eligibility
AMC is available to advertisers running Amazon DSP campaigns, and increasingly to Sponsored Ads-only advertisers with sufficient ad spend and an active advertiser account. Access is requested through your Amazon Ads account team or self-service in the Advertising Console under Measurement & Reporting.
Agencies
Agencies managing multiple client accounts can be granted access to client AMC instances, and can run analysis across accounts they manage — useful for spotting patterns across a portfolio rather than one brand in isolation.
Skills needed
Comfort with SQL is the real barrier for most in-house teams. Query Studio's pre-built templates reduce this significantly, but customising them or building your own analysis still requires someone who can read and adapt a query.
Cost
AMC itself has no separate licence fee — it's included as part of your Amazon Ads relationship. The real cost is analyst time, or an agency/consultancy fee if you bring in outside help to run and interpret queries.
AMC vs Brand Analytics vs Standard Ads Reporting
These three get confused constantly because they all live under the "Amazon data" umbrella, but they answer completely different questions:
What You Can Actually Do With AMC
Once you get past the setup, AMC earns its keep on a specific set of questions that no other report answers directly:
Cross-format overlap and reach
How many shoppers saw your Sponsored Products ad and your DSP display ad before purchasing, versus shoppers who only saw one format? This tells you whether your formats are reinforcing each other or wastefully duplicating reach on the same audience.
New-to-brand analysis, done properly
Standard reporting gives you a new-to-brand percentage per campaign. AMC lets you see which specific combination of touchpoints — a Sponsored Brands video followed by a Sponsored Products click, say — is actually driving new customer acquisition versus which touchpoints are just capturing shoppers who were buying from you anyway.
Path-to-conversion sequencing
See the actual order of ad exposures before a purchase. If awareness-stage DSP impressions consistently precede high-converting Sponsored Products clicks by 5–10 days, that's a concrete argument for budget allocation between upper and lower-funnel formats — backed by your own data, not a generic funnel theory.
Audience building for DSP
Build custom audiences based on actual behaviour combinations — shoppers who viewed a competitor ASIN but didn't purchase, or past purchasers of a complementary category — and push them directly into DSP targeting.
Incrementality and holdout testing
Structure a proper test: expose one audience segment to a campaign, hold out a comparable segment, and measure the actual incremental lift in purchases. This is the closest thing Amazon offers to genuine incrementality measurement, rather than last-click attribution.
Amazon Attribution is not a substitute. Attribution tags your off-Amazon traffic (Google, Meta, TikTok) so it shows up in Amazon reporting. AMC works the other way — it analyses what already happened across your on-Amazon ad exposure at the event level. Serious advertisers use both; they solve different problems.
How to Get Started With AMC
Request an AMC instance
Through your Amazon Ads account manager, or self-service in the Advertising Console if your account is eligible. Instance provisioning typically takes a few days.
Confirm your data sources
By default your instance receives Sponsored Ads and DSP signals. Decide whether to also onboard first-party data (CRM lists, loyalty data) for deeper matching — this requires additional setup and privacy review.
Start with Query Studio templates
Don't write SQL from scratch on day one. Amazon's pre-built templates cover the most common use cases — overlap analysis, path to conversion, new-to-brand — and are a faster way to get a first result while your team gets comfortable with the schema.
Run a small, specific test first
Pick one real question — "does our DSP spend actually lift Sponsored Products conversion for the same ASINs" — rather than trying to build a full measurement framework immediately. A focused first win builds the case for investing further.
Turn findings into budget decisions
AMC output is only valuable if it changes something — a budget shift between formats, a new audience pushed into DSP, a campaign structure change based on sequencing data. Treat every query as a decision input, not a report to file away.
Limitations and Common Mistakes
- Privacy thresholds suppress small results. If your ad spend or audience size is modest, many queries will return no data because the result set falls below Amazon's minimum aggregation threshold. This is by design, not a bug — but it makes AMC frustrating for smaller advertisers.
- Data lag. AMC data is not real-time; expect a reporting lag of a day or more, which makes it unsuitable for daily bid management — that's still a job for standard Ads reporting and automated rules.
- No raw PII, ever. You cannot pull individual customer identities or order-level detail out of AMC, by design. It answers pattern-level questions, not "who bought this."
- SQL is a real barrier. Teams that skip templates and try to write custom queries without SQL experience often get stuck or misinterpret results. Start with templates, bring in specialist help for custom analysis.
- Treating it as a dashboard. AMC isn't a set-and-forget dashboard you check weekly. It's an analysis tool for specific questions. Advertisers who try to make it a routine reporting habit usually abandon it within a few months.
Not sure if AMC is worth setting up for your brand?
We run AMC analysis for clients with meaningful DSP and multi-format spend as part of full account management — and we'll tell you honestly if your account isn't there yet. Book a free audit and we'll assess where your data actually stands.
Book a Free Call →Is AMC Worth It for Private Label Sellers?
Honestly — for most single-digit-million-turnover private label sellers running Sponsored Products and Sponsored Brands only, AMC is not the priority. The privacy thresholds will suppress most of your queries, and the return on the analyst time required won't justify itself yet. Get Brand Analytics, negative keyword hygiene, and campaign structure right first — that's where the bulk of the ACOS improvement lives at smaller scale.
AMC starts to earn its place once you're running meaningful DSP spend alongside Sponsored Ads, managing multiple ASINs or sub-brands where cross-format overlap is a real question, or making budget allocation decisions large enough that a wrong call costs more than the analysis would. For established brands at that stage, it's genuinely the best tool Amazon offers for understanding what your advertising is actually doing — not just what it's spending.
Common Questions About Amazon Marketing Cloud
Do I need to run Amazon DSP to use AMC?
Historically, yes — AMC access was tied to DSP spend. Amazon has been expanding eligibility to Sponsored Ads-only advertisers above a certain spend level, so check current eligibility with your Amazon Ads account team even if you don't run DSP.
Can I use AMC without knowing SQL?
Partially. Query Studio's pre-built templates cover common use cases with a guided interface, so you can get results without writing SQL yourself. Anything beyond the templates — custom logic, unusual segment definitions — requires someone who can write or adapt queries.
Is my customer data safe in AMC?
Yes — that's the entire point of the clean room design. Data is pseudonymised, results are aggregated above minimum thresholds to prevent re-identification, and neither party can extract individual-level records. This is stricter than most third-party ad platforms.
How is AMC different from Amazon Attribution?
Attribution measures the impact of your off-Amazon marketing (Google, Meta, TikTok ads) on Amazon sales by tagging those campaigns. AMC analyses the event-level data from your on-Amazon advertising itself, across formats, to reveal overlap and sequencing. They're complementary, not competing tools.
How much does it cost to set up AMC?
There's no separate Amazon licence fee for AMC access itself. The real cost is time — either an in-house analyst capable of writing and interpreting SQL queries, or an agency/consultancy engagement to run and act on the analysis.