Amazon releases geographic optimization API for DSP advertisers

Amazon unveiled the Geographic Insights and Activation (GIA) API in beta as part of its Ads API v1 suite, enabling programmatic access to location-based optimization features within Amazon DSP. The technical documentation appeared in Amazon's developer resources for the Advanced Tools Center, though no specific announcement date was disclosed in the available materials.

The API provides six distinct endpoints designed to manage geographic optimization strategies at the postal code level. Advertisers can create location indexes that serve as containers for geographic performance data, upload versioned datasets containing index values by postal code, and configure smart location groups that automatically adjust bids based on percentile-based performance ranges. The system supports both direct index values and constituent index values calculated from brand sales and category sales data.

GIA addresses a specific challenge facing advertisers running campaigns through Amazon DSP. Many brands experience concentrated performance in certain geographic regions while remaining relatively unknown in others, creating untapped opportunities that traditional targeting methods fail to capture. The API builds on the visual interface Amazon launched on June 23, 2025, which enabled advertisers to identify high-potential markets using first-party retail signals.

The technical implementation requires three prerequisites: an Amazon DSP account, Amazon Marketing Cloud access, and active ad groups. Advertisers using AMC can combine their proprietary geographic data with Amazon's retail insights, creating custom optimization strategies tailored to specific business objectives. This integration differentiates GIA from standard geographic targeting by enabling data-driven bid adjustments rather than simple inclusion or exclusion rules.

The API's location index structure operates through a versioned system. Advertisers first create a named location index using the POST /adsApi/v1/create/locationIndexes endpoint, which returns an index ID for subsequent operations. Each index can contain multiple versions, allowing advertisers to update their geographic strategies as market conditions change without creating entirely new indexes. The system automatically tracks which version is the latest, simplifying version management for campaigns running continuously.

Location index versions contain the actual performance data at the postal code level. According to the technical specification, advertisers can upload data in two formats. Direct index values provide pre-calculated performance scores for each postal code, ranging from 0 to 100. Constituent index values require brand sales and category sales figures for each postal code, with the system calculating index values based on the ratio between these metrics. The versioning process initially sets status as "IN_PROGRESS," then updates to either "SUCCESS" or "FAILED" once processing completes.

Smart locations represent the operational component of the API. These optimization rules consume location index data to create automated bid adjustment strategies. When creating a smart location through the POST /adsApi/v1/create/smartLocations endpoint, advertisers specify an index ID, associate it with a specific ad group, and configure percentile-based ranges that determine how bids should be adjusted. The system supports both inclusion and exclusion strategies across these ranges.

The percentile configuration allows granular control over bid adjustments. An example configuration might decrease bids by 80% for postal codes in the 0-30th percentile range, increase bids by 100% for the 30th-80th percentile range, and again decrease bids by 80% for the 80th-100th percentile. This approach enables advertisers to concentrate spending in mid-performing regions while reducing exposure in both low-performing and already-saturated markets. Each smart location configuration automatically generates an adjustment rule ID that links to the associated ad group.

The update functionality operates with PATCH behavior, meaning modifications only affect specified fields while preserving existing configurations. When advertisers change the ad group ID associated with a smart location using the POST /adsApi/v1/update/smartLocations endpoint, the system automatically creates a new adjustment rule ID and applies the bid adjustment strategy to the newly specified ad group. This design allows rapid testing of different optimization strategies across campaign structures.

Amazon DSP has systematically expanded its targeting and optimization capabilities throughout 2024 and 2025. The platform consolidated region and location targeting interfaces in May 2024, introduced goal-based bidding in September 2024, and expanded Related Product Targeting to third-party supply in August 2024. These enhancements reflect Amazon's strategy to provide sophisticated automation tools that leverage first-party data signals while maintaining advertiser control over campaign execution.

The timing of the API release aligns with broader developments in Amazon's measurement and data infrastructure. Amazon made Marketing Cloud directly accessible to sponsored ads advertisers on September 16, 2025, eliminating traditional onboarding barriers through enhanced interface capabilities. This democratization of advanced analytics tools creates a foundation for more advertisers to develop the custom geographic datasets required for GIA implementation.

AMC's role in the GIA workflow extends beyond simple data storage. The platform functions as a privacy-safe clean room where advertisers can combine Amazon's pseudonymized shopping signals with their own customer data and third-party sources. This unified data environment enables sophisticated geographic analysis that accounts for both online and offline customer behavior. Advertisers can identify patterns such as strong brand awareness in certain regions with low purchase conversion, or conversely, high purchase rates in areas with limited brand visibility.

The API's data model represents a departure from traditional geographic targeting approaches. Rather than selecting specific cities or postal codes to include or exclude, advertisers define performance indexes that automatically segment markets into relative performance bands. This abstraction layer means the same optimization strategy can adapt as market conditions shift, with the system automatically recalculating which postal codes fall into each percentile range based on the latest index version.

Geographic optimization for programmatic display advertising has gained prominence as privacy regulations limit audience targeting options. Google Ad Manager expanded postal code targeting to Spain and Italy on June 16, 2025, demonstrating industry-wide movement toward location-based precision as an alternative to behavioral targeting methods affected by cookie deprecation and privacy frameworks.

The postal code granularity in GIA provides significant precision for advertisers with sophisticated market intelligence. A national retailer expanding into new regions could identify specific postal codes where category demand is high but brand penetration remains low, then allocate increased advertising resources to those areas through automated bid adjustments. Conversely, established brands might reduce spending in areas where organic brand strength already drives strong performance without incremental advertising support.

Amazon's technical documentation includes example scenarios demonstrating practical applications. A hypothetical "Seattle Coffee Shops Performance Index" illustrates how local businesses might use the system to optimize regional campaigns. The examples show postal codes in the Seattle area with index values ranging from 88 to 95, with corresponding bid adjustment multipliers designed to emphasize mid-range opportunities while constraining spend in extreme performance bands.

The API endpoints follow RESTful conventions with POST methods for all operations, including retrieval functions. This design pattern differs from traditional REST implementations where GET methods typically handle read operations. The specification indicates that all request and response bodies use JSON format, with success and error arrays allowing batch operations to partially succeed when processing multiple entities simultaneously.

Integration with Amazon's existing advertising technology stack positions GIA within a comprehensive suite of optimization tools. Amazon DSP added Spotify's global audio and video inventory in October 2025, expanding available placement options across 696 million monthly users. Geographic optimization strategies can now extend across these diverse inventory sources, enabling consistent regional targeting approaches whether advertising appears in streaming audio, connected TV, or display environments.

The beta designation indicates Amazon continues refining the feature based on advertiser feedback and system performance. Beta programs typically involve limited availability to qualified advertisers who meet specific technical and spending thresholds. As the system processes real-world geographic optimization strategies across diverse campaign types, Amazon likely monitors performance impact, API reliability, and user experience to inform eventual general availability decisions.

Operational considerations for implementing GIA include data refresh frequency and index versioning strategy. Advertisers must determine appropriate cadences for updating location index versions based on their market dynamics. Rapidly changing markets might require weekly or monthly updates, while stable categories could operate on quarterly refresh cycles. The versioning system accommodates these different approaches without requiring complex data migration or campaign restructuring.

The marketing community's interest in geographic optimization reflects fundamental economic principles of advertising efficiency. Allocating budget toward highest-opportunity markets while reducing spend in saturated or underperforming areas should theoretically improve overall campaign return on ad spend. However, the effectiveness depends heavily on the quality of the underlying geographic intelligence. Advertisers with sophisticated first-party data and strong AMC implementation capabilities stand to benefit most from these tools.

Amazon's approach differs from competitors by integrating retail transaction data into the optimization framework. While other platforms offer geographic targeting, few can match Amazon's visibility into actual purchase behavior across millions of products and geographic markets. This data advantage becomes particularly valuable when combined with an advertiser's own customer intelligence, creating compound insights that neither party could generate independently.

Technical limitations and considerations appear throughout the specification. The system processes location index versions asynchronously, meaning immediate feedback on data validation issues may not be available. Advertisers must implement appropriate error handling and status monitoring to ensure index versions process successfully before activating smart locations that depend on them. Failed index versions would prevent optimization strategies from functioning as intended.

The postal code format follows country-specific conventions with optional country prefixes. Examples in the documentation use "US-98101" format for United States addresses, indicating that international implementations must adapt to local addressing systems. This flexibility accommodates Amazon DSP's global footprint while requiring advertisers to understand format requirements for markets where they operate campaigns.

Bid adjustment multipliers in the API use decimal values representing percentage increases or decreases from base bids. A value of 2.0 represents a 100% increase (doubling the bid), while 0.2 represents an 80% decrease. This mathematical approach provides precise control but requires advertisers to carefully calculate appropriate adjustment levels based on their cost per acquisition targets and campaign economics.

The smart location naming convention allows descriptive identifiers that help advertisers manage multiple optimization strategies. The example "Coffee Shop High Performance Optimizing" demonstrates how advertisers might label strategies to indicate their purpose and configuration. As campaigns scale to include multiple smart locations across different ad groups and optimization approaches, clear naming conventions become essential for operational management.

Integration complexity varies based on advertiser technical capabilities and existing infrastructure. Organizations with sophisticated marketing technology stacks and dedicated engineering resources can build automated workflows that regularly update location indexes, monitor processing status, and adjust smart location configurations based on performance data. Smaller advertisers might implement more manual processes, updating indexes quarterly and maintaining simpler optimization strategies.

The API's relationship with other Amazon Ads APIs creates potential for comprehensive automation. Advertisers could theoretically combine GIA with campaign management APIs, budget optimization APIs, and reporting APIs to create fully automated geographic optimization systems. Such implementations would require significant development investment but could enable sophisticated market entry and expansion strategies executed at scale.

Market entry scenarios represent a particularly compelling use case. Brands launching in new countries or regions could use GIA to systematically identify and prioritize geographic expansion opportunities. By analyzing category performance across postal codes, advertisers could discover unexpected market pockets where demand signals suggest receptive audiences despite limited current brand presence. This intelligence could inform both advertising strategy and broader business decisions about distribution and market development.

The percentile-based approach to defining optimization ranges offers advantages over absolute threshold methods. Markets naturally evolve over time, with performance distributions shifting as competitive dynamics and consumer behaviors change. Percentile-based rules automatically adapt to these shifts without requiring manual threshold updates. The 30th-80th percentile range consistently captures middle-performing markets regardless of whether absolute index values drift higher or lower over time.

Amazon's decision to build GIA on the Ads API v1 data model suggests longer-term architectural planning. Version 1 APIs typically represent significant platform investments expected to support extended product lifecycles. This foundation should provide stability for advertisers building integrations, reducing the risk of disruptive changes that would require substantial rework of optimization systems.

The competitive landscape for geographic optimization tools remains relatively sparse in programmatic advertising. While most platforms offer basic geographic targeting, few provide automated bid adjustment systems driven by custom performance indexes. Amazon's integration of retail data, clean room capabilities, and optimization automation creates a differentiated offering that leverages unique data assets competitors cannot easily replicate.

Implementation guidance in the technical documentation focuses on API mechanics rather than strategic best practices. Advertisers seeking to maximize GIA effectiveness will likely benefit from establishing clear hypotheses about geographic opportunity, developing rigorous testing frameworks to validate optimization strategies, and maintaining disciplined processes for index updates and performance monitoring. These operational practices will matter as much as technical implementation quality.

The announcement positions Amazon to capture advertiser interest in data-driven localization strategies. As national brands increasingly recognize the importance of regional customization, tools that automate geographic optimization while leveraging first-party insights become strategically valuable. GIA addresses this demand by combining Amazon's unique data position with programmatic automation that scales beyond what manual campaign management could achieve.

Timeline

• January 5, 2021: Amazon launches Amazon Marketing Cloud beta for eligible agencies, advertisers, and tool providers
• May 2, 2024: Amazon DSP streamlines location targeting by consolidating region and location targeting interfaces
• May 20, 2024: Amazon Marketing Cloud unveils High Value Audiences solution for customer segmentation
• June 16, 2025: Google Ad Manager expands postal code targeting to Spain and Italy
• June 23, 2025: Amazon launches Geographic Insights and Activation feature for Amazon DSP with visual interface
• August 27, 2024: Amazon DSP expands Related Product Targeting to third-party supply
• September 16, 2025: Amazon makes Marketing Cloud directly accessible to sponsored ads advertisers
• October 1, 2025: Amazon DSP adds Spotify's global audio and video inventory across nine markets
• October 15, 2024: Amazon Ads expands measurement and targeting capabilities with AMC enhancements
• Date undisclosed: Amazon releases Geographic Insights and Activation API beta in Advanced Tools Center

Summary

Who: Amazon released the Geographic Insights and Activation API for advertisers using Amazon DSP and Amazon Marketing Cloud. The beta targets self-service and managed service advertisers with technical capabilities to integrate programmatic optimization tools.

What: The GIA API provides six endpoints for creating location indexes, uploading postal code-level performance data, and configuring smart locations that automatically adjust bids based on percentile-based geographic performance ranges. The system supports both direct index values and constituent calculations from brand and category sales data.

When: The API documentation appeared in Amazon's Advanced Tools Center for the Ads API v1 platform, though no specific announcement date was disclosed in available technical materials. The API builds on the GIA visual interface that Amazon launched on June 23, 2025.

Where: The API operates within Amazon DSP across all regions where the platform is available. Implementation requires Amazon Marketing Cloud access, which is available in North America, Europe, Japan, and Australia. The system processes geographic data at the postal code level with country-specific format support.

Why: The API addresses advertiser demand for automated geographic optimization driven by first-party data insights. It enables brands to systematically identify regional growth opportunities, adjust bids toward underpenetrated markets with high category demand, and reduce spending in saturated or underperforming areas. The programmatic approach scales beyond manual campaign management while leveraging Amazon's unique visibility into shopping behavior across geographic markets.