Guides
Operational playbooks for AI-assisted Amazon work.
Durable implementation guides for teams using standards-based AI clients, MCP, and Amazon operating data.
Counting New-to-Brand Customers in AMC Without Double-Counting the Same Buyer
How an Atlas-grounded AMC query counts genuine first-time buyers, why the conversion-time table is the only safe pick for a recurring workflow, and the five caveats the IQ docs scatter across four pages.
Sponsored Ads x DSP Overlap: The 4-Way AMC Analysis
How an Atlas-grounded agent runs the improved Sponsored Ads and demand-side platform overlap query in AMC, why Sponsored Products exposure means impressions here, and the caveats that keep the result honest.
Choosing the Right AMC Attribution Data Source for Custom Models
An operator asks for a custom attribution view in AMC. An agent grounded in Amazon Agent Atlas returns a five-row data-source matrix, six decision rules, and the footnotes that would have eaten the rest of the week.
AMC Lookalike Audience Evaluation: Compare Prime Day Lift Against Last Year's Rule-Based Audience
After a Prime Day lookalike audience activates, an agent grounded in Amazon Agent Atlas compares it with last year's rule-based audience baseline using flight-time conversions, NTB share, cost per NTB customer, and standards-based handoffs across chat, code, and workflow clients.
Customer Lifetime Value Segmentation: Beyond ROAS
An operator asks which customer cohorts are actually worth the most over time. An Atlas-grounded agent returns the composite Lifetime × Value SQL — and the year-over-year diagnostic that catches cohort drift before it costs you.
Segmenting Your Highest-Value Customers: Three Seed Strategies
A CRM lead asks for a DSP audience of best customers and people who look like them. An agent grounded in Amazon Agent Atlas returns three seed strategies — SnS, Multi-Purchase, Total Spend — each sized into the 1,000 to 450,000 buffer before activation, with the AMC table-variant gotcha that would have killed the first push.
Finding the Three-Time Buyers Who've Gone Quiet: An AMC Audience for Lapsed Loyalists
A retention lead wants an audience of customers who bought three times last year and went silent. An agent grounded in Amazon Agent Atlas returns an AMC Audiences query that actually compiles, plus the size constraint that would have killed the first activation.
AMC Lookalike Audience for Prime Day: Seed SQL, Sizing Query, and the 500-500,000 user_id Window
Six weeks out from Prime Day, an operator asks for a lookalike audience built on best customers. An agent grounded in Amazon Agent Atlas returns three seed strategies, a sizing query, API payload guidance, and the playbook footnotes that decide whether the audience activates.
Mapping the Path to Conversion Across DSP and Sponsored Ads
An operator asks what the customer journey actually looks like across DSP, SP, SB, and SD. An agent grounded in Amazon Agent Atlas returns the right AMC instructional query, the table joins that compile, and the campaign-grouping rule that keeps the output from collapsing into NULLs.
Building an AMC Cart-Abandoner Audience: A Workflow That Compiles on the First Try
An operator asks for a DSP retargeting audience. An agent grounded in Amazon Agent Atlas returns SQL that runs — plus the footnotes that would have eaten the rest of the afternoon.
Quantifying Subscribe & Save Lift in AMC: The Signal Most Agents Don't Know Exists
A marketing director wants to know if Subscribe & Save is actually moving revenue. An agent grounded in Amazon Agent Atlas returns a working AMC lift analysis built on a signal Amazon only added in February 2024.
Turning AMC Workflows into Agent Skills
Why Amazon Marketing Cloud is a good fit for repeatable, auditable agent workflows.
Why MCP Needs Run Logs
Tool access is only useful in production when every action is scoped, visible, and reproducible.
Human Approval for AMC Activation
Audience generation can be automated, but activation should stay explicit.