AWS Entity Resolution
A managed service that matches and unifies customer records scattered across multiple data sources to build unified customer profiles
Overview
AWS Entity Resolution is a fully managed service that matches customer records distributed across different data sources, identifies the same entities, and unifies them. It performs identity resolution across customer data scattered in CRM, e-commerce sites, call centers, and other systems using attributes such as name, address, email, and phone number as matching keys. It provides both rule-based matching (combinations of exact and partial match conditions) and ML-based matching (machine learning similarity scoring), allowing selection based on data quality and use case. It integrates directly with Glue Data Catalog and S3 data sources, and when incorporated as part of an ETL pipeline, automates record deduplication and unification before data warehouse ingestion.
Matching Workflow Design
In Entity Resolution, Schema Mapping defines the fields of the input data, and Matching Workflow matches the records. ID Mapping is a separate workflow that translates source data to a target ID system. Schema Mapping maps columns from each data source to standard attributes (name, address, phone number, email, etc.), enabling unified handling of data with different schemas. Matching Workflow defines matching rules that specify which attribute combinations determine same-entity identification. Rule-based matching allows declarative specification of compound conditions such as "email address exact match" OR "phone number exact match AND last name phonetic match." In ML-based matching, a pre-configured ML model analyzes the entire record using name, email address, phone number, address, and date of birth fields, and assigns each match group a confidence score from 0.0 to 1.0. Processing results are assigned as Match IDs to each record, uniquely identifying groups of records belonging to the same entity.
Integration with Data Provider Services
Entity Resolution integrates with third-party data providers on AWS Data Exchange, offering enrichment capabilities that supplement your data with external data. Supported providers are LiveRamp, TransUnion, and Unified ID 2.0 (as of October 2026); for example, TransUnion enhances customer records with over 200 data attributes. In ID mapping with LiveRamp, using the RampID as a token to represent customers avoids sharing customer data directly with advertising platforms. An ID namespace defines the source and target of an ID mapping workflow, and IDs can also be translated across two AWS accounts. Combined with Clean Rooms, it can also be leveraged for data collaboration with partner companies.
Operational Patterns and Cost Optimization
Entity Resolution pricing is based on the number of records processed, including records that do not result in matches (pricing page, as of October 2026). Rule-based matching supports automatic incremental processing: as soon as new data is available in S3, only the new records are compared against existing records (and only incremental records are charged). Combined with Glue jobs, you can build event-driven pipelines that auto-execute triggered by new file arrivals on S3. For matching accuracy tuning, adjust thresholds to balance False Positives (incorrectly identified as same) and False Negatives (same entity missed) according to business requirements.
References (Official AWS Resources)
The primary sources for this page are the official AWS website and documentation. Check the official pages below for the latest specifications and pricing.
- AWS Glue official page
- AWS Lake Formation official page
- AWS Clean Rooms official page
- AWS Documentation (official)
If this page and the official documentation disagree, treat the official documentation as authoritative.