How to Export Amazon UAE Seller & Ad Data to Google BigQuery

Amazon seller data and Amazon advertising data come from two separate APIs, so you build two pipelines into BigQuery and join them only at the reporting layer. Seller orders, refunds and inventory come from the Selling Partner API (SP-API). Sponsored campaign performance comes from the Amazon Ads API. Keep marketplace revenue and ad-attributed sales in separate columns, and reconcile both with Seller Central before making budget decisions.

Separate seller reporting from advertising

Seller orders, inventory and returns don't come from the same report as sponsored campaign performance. Confirm access, marketplace coverage and supported reports for each account.

This guide describes a reporting architecture. AKM does not currently list a native Amazon.ae seller-and-ads connector on its integrations page, so don't assume that connecting other AKM sources enables these Amazon datasets.

Prerequisites

You need an authorized selling partner application or suitable seller exports, approved advertising access or advertising exports, a Google Cloud project and a BigQuery dataset. Limit personal data to what the reporting use case actually needs.

Build the pipeline

Choose source reports. Define order, refund, inventory and advertising fields separately. Record report period, marketplace and currency.
Request and retrieve seller reports. Follow Amazon's report retrieval documentation, including generation status, document retrieval and any compression it indicates.
Retrieve advertising reports separately. Follow the current Amazon Ads API documentation. Metric names and attribution periods vary by report and ad product.
Load with stable keys. Stage validated files, then replace reporting windows or merge on business keys. Keep order lines separate from order totals to avoid counting revenue twice.
Refresh revised periods. Allow for late advertising attribution and later returns. Monitor failures and show freshness for each source separately.

The load pattern is the same one we use for other platforms. See the TikTok to BigQuery and Snapchat Ads to BigQuery guides for matching table keys.

Define ACoS, TACoS and business-level ratios

ACoS (%) = advertising spend / advertising-attributed sales × 100 for the same ad product and reporting basis.
TACoS (%) = Amazon advertising spend / defined total Amazon sales × 100 for a matched period.
Business revenue-to-ad-spend ratio = defined Amazon and DTC revenue / defined total paid media spend.

Illustrative example: AED 5,000 of Amazon advertising spend, AED 20,000 of attributed sales and AED 50,000 of total Amazon sales give 25% ACoS and 10% TACoS. Handle zero revenue explicitly, and say whether refunds and taxes are included.

If you also sell direct-to-consumer, the business-level ratio works like blended ROAS. Our guide to tracking blended ROAS and MER explains how to set its definitions.

Verify before making budget decisions

Reconcile one completed period with Seller Central and the advertising console, using the same marketplace, dates, currency and attribution basis. Check that reloading doesn't change counts unexpectedly, and that a refunded order updates your revenue model.

A combined dashboard can show correlations between outside campaigns and marketplace sales. It doesn't prove that Meta or TikTok drove extra Amazon sales; that needs a proper experiment or measurement design.

Use the same acceptance checks for any custom or third-party pipeline. Before promising an automated export, check which sources AKM connects today on the integrations page.

Separate seller reporting from advertising

Seller orders, inventory and returns don't come from the same report as sponsored campaign performance. Confirm access, marketplace coverage and supported reports for each account.

Build the pipeline

Choose source reports. Define order, refund, inventory and advertising fields separately. Record report period, marketplace and currency.

Request and retrieve seller reports. Follow Amazon's report retrieval documentation, including generation status, document retrieval and any compression it indicates.

Retrieve advertising reports separately. Follow the current Amazon Ads API documentation. Metric names and attribution periods vary by report and ad product.

Load with stable keys. Stage validated files, then replace reporting windows or merge on business keys. Keep order lines separate from order totals to avoid counting revenue twice.

Refresh revised periods. Allow for late advertising attribution and later returns. Monitor failures and show freshness for each source separately.

The load pattern is the same one we use for other platforms. See the TikTok to BigQuery and Snapchat Ads to BigQuery guides for matching table keys.

Define ACoS, TACoS and business-level ratios

ACoS (%) = advertising spend / advertising-attributed sales × 100 for the same ad product and reporting basis.

TACoS (%) = Amazon advertising spend / defined total Amazon sales × 100 for a matched period.

Business revenue-to-ad-spend ratio = defined Amazon and DTC revenue / defined total paid media spend.

If you also sell direct-to-consumer, the business-level ratio works like blended ROAS. Our guide to tracking blended ROAS and MER explains how to set its definitions.

Verify before making budget decisions

Use the same acceptance checks for any custom or third-party pipeline. Before promising an automated export, check which sources AKM connects today on the integrations page.