# Reltio Master Data Management

> Cloud MDM platform, now part of SAP, with real-time ML entity resolution, survivorship computed when records are read, multi-domain mastering, reference data and hierarchies, plus AgentFlow agents and an MCP server for governed agent access.

Source: https://toolsfordata.com/tools/reltio/ · Updated 2026-11-19

[Website](https://www.reltio.com/) · [Docs](https://docs.reltio.com/)

Best for: large enterprises in life sciences, financial services and healthcare that need real-time, cloud-only MDM for customer and other domains, next to a separate catalog for governance.

## Overview

- Core Strength: Real-time SaaS MDM with pretrained LLM matching and AI agents
- Matching: Pretrained LLM matching plus fuzzy and exact rules, in real time
- Golden Records: Field-level survivorship, steward review, unmerge
- Platform Fit: Pipelines to Snowflake, Databricks, BigQuery
- Deployment Model: SaaS
- Pricing: Quote from sales
- Notable Caveat: SaaS only, no self-hosting

## What it's used for

Reltio, part of SAP since May 2026, is used for real-time master data management in the cloud: profiles are matched and merged as data arrives, and survivorship rules compute each attribute's value at read time, per user role.

## Features

### Matching

- ML Matching: Yes — "Pretrained, LLM-driven matching models" (FERN) score records, with "No model training required". ([source](https://www.reltio.com/products/entity-resolution/))
- Rule-Based Matching: Yes — "Deterministic match rules with fuzzy and exact comparison logic" and a match rule builder. ([source](https://www.reltio.com/products/entity-resolution/))
- Real-Time Matching: Yes — "Real-Time, AI-Driven Entity Resolution", with latency "in milliseconds". ([source](https://www.reltio.com/products/entity-resolution/))
- Incremental Matching: Unknown — Not stated explicitly; the closest claim is automatic unmerge "even as data and attributes change". ([source](https://www.reltio.com/products/entity-resolution/))
- Persistent Entity ID: Yes — "Unique, Reltio IDs persist across connected systems", with full traceability. ([source](https://www.reltio.com/products/entity-resolution/))
- Match Explainability: Unknown — A match rule analyzer is named, but no per-match explanation is described on the pages read. ([source](https://www.reltio.com/products/entity-resolution/))

### Golden Records

- Survivorship Rules: Yes — "Field-based survivorship to select attributes from many crosswalk records". ([source](https://www.reltio.com/products/entity-resolution/))
- Survivorship on Read: Unknown — "Dynamic survivorship" is named, but read-time per-consumer computation isn't described on the public (staging) docs. ([source](https://docs.reltio.com/reltio/what-reltio-does-at-a-glance/data-unification-and-mdm-at-a-glance/data-unification-and-mdm-in-detail))
- Stewardship Review: Yes — "Data stewardship workflows and bulk match reviews to resolve potential matches", plus unmerge. ([source](https://www.reltio.com/products/entity-resolution/))
- Record History: Unknown — "Auditable merges" are claimed; version history is not described on the pages read. ([source](https://www.reltio.com/products/agentflow/))
- Hierarchy Management: Unknown — Hierarchies appear under reference data and graph relationship views; parent-child hierarchy management is not described on the pages read. ([source](https://www.reltio.com/products/multidomain-mdm/))

### Domains & Modeling

- Multi-Domain MDM: Yes — "All data domains unified in one platform": customer, product, supplier, location. ([source](https://www.reltio.com/products/multidomain-mdm/))
- Configurable Data Model: Yes — "A flexible data model" where you add sources, attributes or entities. ([source](https://www.reltio.com/products/multidomain-mdm/))
- Product Information Management: Unknown — Product is a domain, but no PIM features are described on the pages read. ([source](https://www.reltio.com/products/multidomain-mdm/))
- Reference Data Management: Yes — Reference data management "standardizes codes, hierarchies, and attributes across systems". ([source](https://www.reltio.com/products/multidomain-mdm/))
- Master Data Authoring: Yes — Hub: "Search, create, update, and delete profiles", with an Inbox for workflow tasks. ([source](https://docs.reltio.com/applications/hub/hub-at-a-glance))

### Data Quality

- Validation at Entry: Unknown — Continuous data quality is described; validation on create/edit is not described on the pages read. ([source](https://www.reltio.com/products/multidomain-mdm/))
- Standardization: Unknown — "Standardized, and enriched core data" is claimed, without cleansing functions described on the pages read. ([source](https://www.reltio.com/products/multidomain-mdm/))
- Third-Party Enrichment: Yes — "Enrich data with third-party sources such as D&B, Moody's, ZoomInfo, NPI". ([source](https://www.reltio.com/products/data-integration/))

### Data Distribution

- REST API: Yes — "REST APIs in real time or batch". ([source](https://www.reltio.com/products/data-integration/))
- Change Events: Unknown — Only inbound streaming from a message queue is described on the pages read. ([source](https://www.reltio.com/products/data-integration/))
- Warehouse Write-Back: Yes — Provisions cleansed data to BigQuery, Databricks and Snowflake (Reltio Data Pipeline). ([source](https://www.reltio.com/products/data-integration/))
- Data Syndication: Unknown — No syndication to external channels or GDSN on the pages read. ([source](https://docs.reltio.com/en/products/reltio-multidomain-master-data-management-mdm))

### AI & Agents

- MCP Server: Yes — An "MCP server" for governed connectivity across apps, warehouses and AI platforms. ([source](https://www.reltio.com/products/multidomain-mdm/))
- AI Stewardship Assistance: Yes — AgentFlow's Resolver agent "automates manual match resolution with AI-driven recommendations"; the Unmerger recommends reversals. ([source](https://www.reltio.com/products/agentflow/))

### Platform Native

- In-Place Matching: No — SaaS only ("Cloud-native SaaS MDM platform", no self-hosting): records are loaded into the Reltio tenant. ([source](https://www.reltio.com/products/multidomain-mdm/))
- Snowflake Native: No — Snowflake is a pipeline destination; Reltio does not run inside Snowflake. ([source](https://www.reltio.com/products/data-integration/))
- Databricks Native: No — Databricks is a pipeline destination; Reltio does not run inside Databricks. ([source](https://www.reltio.com/products/data-integration/))
- Microsoft Fabric Native: No — Microsoft Fabric is not mentioned. ([source](https://www.reltio.com/products/data-integration/))

### Deployment

- SaaS Deployment: Yes — "Cloud-native SaaS MDM platform" with zero-downtime upgrades. ([source](https://www.reltio.com/products/multidomain-mdm/))
- Self-Hosted Deployment: No — SaaS only; no customer-run option is described. ([source](https://www.reltio.com/products/multidomain-mdm/))

### Open Source

- Open Source: No — Commercial SaaS. ([source](https://www.reltio.com/products/multidomain-mdm/))

## Integrations

8 integrations, each sourced from the vendor docs.

- Data Enrichment: Dun & Bradstreet, Moody's, ZoomInfo
- Data Integration: MuleSoft, SnapLogic
- Data Platform: Databricks
- Data Warehouse: Snowflake, Google BigQuery

## Alternatives

- [Informatica IDMC by Salesforce](https://toolsfordata.com/tools/informatica/): Consider Informatica for MDM inside one suite with catalog and governance.
- [Semarchy](https://toolsfordata.com/tools/semarchy/): Consider Semarchy for no-code MDM that's fast to deploy, including Snowflake-native.
- [Profisee](https://toolsfordata.com/tools/profisee/): Consider Profisee on a Microsoft stack: fast to implement, with deep Fabric and Purview integration.
- [Stibo Systems](https://toolsfordata.com/tools/stibo-systems/): Consider Stibo Systems if product data is the main domain: strong PIM and a long enterprise track record.
- [Ataccama ONE](https://toolsfordata.com/tools/ataccama/): Consider Ataccama for MDM next to data quality and governance, with full self-hosting.
- [Zingg](https://toolsfordata.com/tools/zingg/): The open-source option: ML entity resolution on your own cloud data platform, with persistent IDs in the paid editions.
