Key Takeaways
- Amazon’s MCP server entered open beta in February 2026 with Amazon Marketing Cloud (AMC) included, enabling plain-English queries directly against your AMC instance.
- AMC has always tracked the full customer journey across every ad touchpoint before purchase — the barrier was SQL expertise, and that barrier is now gone.
- Brands can now surface multi-touch attribution data, frequency curves, and new-to-brand rates by channel in minutes, not hours.
- Precise AMC audiences — cart abandoners by recency, high-AOV customers, organic buyer overlaps — can be built and pushed to DSP through plain language descriptions.
- The integration removes the execution barrier. It does not replace the judgment required to ask the right questions and act on the answers correctly.
General Summary
Amazon’s MCP server integration with Amazon Marketing Cloud removes the SQL barrier that has kept most brands from using AMC’s full attribution and audience capabilities. Since February 2026, sellers and agencies can query AMC directly in plain English — asking multi-touch attribution questions, building rule-based DSP audiences, and surfacing frequency curves without writing a single line of SQL. AMC has held this data for years. Path-to-purchase sequences, new-to-brand rates by channel, DSP influence on conversions credited elsewhere — all of it has been available and largely unused because access required technical resources most brands do not have in-house. The integration closes that gap. What it does not change is the strategic judgment required to ask meaningful questions, interpret the results correctly, and make sound budget and audience decisions based on them.
Extractive Summary
AMC shows the full customer journey, including every ad touchpoint in sequence before the purchase, which Seller Central’s last-click attribution cannot. AMC via MCP lets you pull the actual path to purchase in minutes from a plain-English prompt. Precise AMC audiences can now be built through plain language and pushed directly to DSP without SQL expertise. The overlap between paid audiences and organic shoppers is a question AMC can answer, and most brands have never asked it. This integration removes the execution barrier but does not replace the judgment that determines what to execute.
Abstractive Summary
For most Amazon brands, advertising decisions have been built on an incomplete picture. Seller Central shows what converted. AMC has always shown why — which campaigns influenced which customers, how many touchpoints preceded a purchase, which channels are growing the customer base versus selling to people who were already coming back. The gap was never data availability. It was access. Amazon’s MCP integration with AMC resolves the access problem at the point where it mattered most: it removes the requirement for SQL expertise that has kept this data out of reach for the majority of brands and agencies running meaningful ad budgets. What emerges now is a divided landscape — brands that use this to ask sharper questions about attribution, frequency, and audience quality, and brands that do not. The data was always there. The difference going forward is who looks at it.
What Data Has Amazon Marketing Cloud Always Had?
Amazon Marketing Cloud has always tracked the full customer journey — every ad touchpoint in sequence before a purchase — in a way that Seller Central’s last-click attribution cannot. Seller Central shows you what converted. AMC shows you what led to the conversion: which campaign the customer saw first, how many touchpoints they hit across DSP and Sponsored Ads, and which campaigns influenced purchases that Seller Central credits to something else entirely.
That last category is the one that changes budget decisions. When a DSP campaign that appears to be underperforming in Seller Central has influenced 60% of the conversions credited to Sponsored Products, the decision to cut it looks very different. AMC has always surfaced that relationship. The question was whether most brands could actually reach it.
The barrier was SQL. Pulling multi-touch attribution data from AMC required writing custom queries, structuring them correctly for your instance, and having someone available to troubleshoot when they failed. For brands without a dedicated data team — which is most brands — that meant AMC sat largely unused, or was used for a narrow set of reports someone else built months ago.
Amazon’s own documentation records that AMC query development dropped from hours to minutes following the MCP integration. For brands without in-house technical resources, that is not a marginal improvement. It is the difference between using AMC and not using it at all.
What Attribution Questions Can You Now Ask in Plain English?
AMC via MCP lets you pull the actual path to purchase in minutes from a plain-English prompt, which changes how every campaign on the account gets evaluated. The questions that were previously gated behind SQL expertise are now accessible to anyone managing the account.
Practical examples: you can ask for the conversion path for every customer who saw a video ad before converting — and find out whether that video ad is doing attribution work Seller Central never credits. You can ask for a frequency analysis showing how many DSP impressions a customer sees before converting. If the answer is three and your frequency cap is set to one, you are cutting the journey off before it completes.
You can compare new-to-brand rates across Sponsored Products, Sponsored Brands, and DSP for the same period. That comparison tells you which channel is growing your customer base and which is selling to people who were already going to buy. Those two things require different budget logic.
These are not exotic analytics questions. They are questions every account manager should be answering every month. The reason most are not is not lack of interest — it is that reaching the answers required building SQL workflows or paying someone who could. According to Amazon’s MCP documentation, that requirement is now gone for accounts with AMC access.
The frequency analysis alone justifies the setup. Brands running DSP without knowing their effective frequency curve are either under-investing in impressions or burning them past the point of return. AMC has held this data for years. The integration makes it reachable without a data analyst in the room.
How Does the MCP Integration Change Audience Building for DSP?
You can now build precise AMC audiences through plain-language descriptions and push them directly into DSP — without SQL expertise, and without a technical resource to translate what you want into what AMC needs. The audiences themselves are not new. The access to them is.
The specific audience types that become practical include: customers who viewed your product but did not add to cart in the last 90 days; shoppers who added to cart but did not purchase and have not been retargeted; your highest-value customers above a defined AOV threshold, as a seed for lookalike targeting in DSP. Each of these has always been buildable in AMC. The friction was the technical layer between knowing what you wanted and getting it into the platform.
What the integration removes is that dependency. You describe the audience in plain English. The query gets constructed and submitted to AMC. The audience moves into DSP. For brands running meaningful DSP budgets, the quality of AMC audiences directly determines the quality of retargeting.
Cart abandoners are not a single audience. A customer who added to cart yesterday and one who added three months ago are different targeting propositions with different bid logic, different creative requirements, and different conversion probabilities. AMC lets you segment that properly. The brand retargeting “added to cart in the last 14 days, not yet purchased, never retargeted” is running a fundamentally different campaign from one using a generic cart-abandonment list — at the same DSP budget, with meaningfully different precision.
Where Does the Strategic Edge in AMC Actually Come From?
The overlap between your paid audiences and your organic shoppers is a question AMC can answer, and most brands have never asked it. If 40% of the customers your DSP is targeting are already repeat organic buyers, you are spending impression budget to reach people who were returning anyway. That is not a retargeting campaign — it is waste with a DSP line item.
The edge is not in asking the standard attribution questions. It is in asking the questions your competitors have not considered yet, now that the time cost of asking has dropped from hours to minutes.
The correlation between ad spend and organic rank movement is another: which campaigns are driving BSR improvements, not just direct conversions. That changes how top-of-funnel investment gets evaluated, because you are no longer measuring it only by what it closes directly. If Sponsored Brands activity is lifting organic rank on a high-volume keyword, the return on that investment is larger than ROAS shows.
Cross-channel measurement surfaces a third angle: the overlap between your Sponsored Ads and DSP audiences. Double-serving the same customer across both channels wastes impressions and inflates frequency without adding reach. Reducing that overlap does not just save spend — it reallocates impressions toward customers who have not yet been exposed.
These questions existed before February 2026. The constraint was always time and expertise. When asking a question takes two minutes instead of two hours, more questions get asked. More questions mean more findings. Competitors who are not using AMC this way are not finding them.
What Does the Amazon MCP and AMC Integration Not Do?
This integration removes the execution barrier. It does not replace the judgment that determines what to execute, and that distinction matters more than most coverage of this integration acknowledges.
Claude through AMC will run the analysis you ask for. It will not tell you whether you are asking the right question. It will build the audience you describe. It will not tell you whether that audience fits where you are in the funnel, or whether your DSP bids make sense relative to your margin structure. It will show you the frequency curve. It will not tell you what the correct frequency cap is for your category.
The strategic layer — knowing which questions matter, reading results correctly, deciding what to do about them — remains entirely on the account manager. What AMC via MCP removes is the SQL barrier and the hours of technical work. The insight gap does not move.
The right starting point is attribution, not audiences or budgets. Run the path-to-purchase analysis first. Check new-to-brand rates by channel. Understand what you are seeing before making decisions based on it. The data has been in AMC for as long as you have been running campaigns. Take the time to learn what it says before you act on it.
