# Tilores Entity Resolution

> Real-time entity resolution engine, as a managed AWS service or self-hosted, that matches records as they arrive with explicit rules and ML scoring, keeps an auditable entity graph, and serves resolved entities through GraphQL.

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

[Website](https://tilores.io/) · [Docs](https://docs.tilotech.io/tilores/) · [Pricing](https://tilores.io/pricing) · [GitHub](https://github.com/tilotech)

Best for: teams that need real-time entity resolution across operational systems with matching rules they can audit.

## Overview

- Core Strength: Real-time entity resolution with an auditable entity graph
- Matching: Explicit rules plus ML scoring, as records arrive
- Golden Records: Best value per field picked at query time
- Platform Fit: AWS-native, also self-hosted; GraphQL API
- Deployment Model: Managed AWS or self-hosted
- Pricing: Quote by entity count; free local Studio
- Notable Caveat: GraphQL only; no steward review queue

## What it's used for

Tilores is used for real-time entity resolution in operational systems: it matches records as they arrive with explicit rules and ML scoring, keeps an auditable entity graph, and serves resolved entities through GraphQL.

## Pricing

Tilores quotes by entity count, and Tilores Studio runs locally for free.

## Deployment

Tilores runs as a managed AWS service or self-hosted in your own AWS account; records are submitted into your Tilores instance.

## Review

Strengths:
- Real-time matching as records arrive, served through GraphQL.
- Explicit rules plus ML scoring, with an auditable entity graph.
- Managed on AWS or self-hosted in your own account.

Caveats:
- GraphQL only, with no steward review queue.
- No stored golden records: the best value per field is picked at query time.
- Records are submitted into the Tilores instance, not matched in your warehouse.

Recommendation: For real-time resolution embedded in your own application, consider Senzing; for pay-per-record matching on S3, AWS Entity Resolution; for batch matching with stable IDs inside your platform, Zingg's paid editions; for a full SaaS hub with stewardship, Reltio.

## Features

### Matching

- ML Matching: Yes — "Deterministic and ML-based matching" in every deployment. ([source](https://tilores.io/pricing))
- Rule-Based Matching: Yes — "Rules are explicit, tunable, and auditable". ([source](https://tilores.io/))
- Real-Time Matching: Yes — Sub-300 ms ingestion; records are matched as they arrive. ([source](https://tilores.io/))
- Incremental Matching: Yes — Records "are matched as they arrive" and the entity graph updates live. ([source](https://tilores.io/))
- Persistent Entity ID: Yes — Entity IDs such as ent_3c8b21a0; "persistent entity memory". ([source](https://tilores.io/))
- Match Explainability: Yes — A Match Explainer lists the rules that linked each record. ([source](https://tilores.io/))

### Golden Records

- Survivorship Rules: Yes — "Best value picked per field across all linked sources". ([source](https://tilores.io/))
- Survivorship on Read: Yes — Query-time functions such as newest(field) pick the "Best value picked per field across all linked sources" per request. ([source](https://tilores.io/))
- Stewardship Review: No — Tilores Studio tunes rules; no review queue. ([source](https://tilores.io/deployment))
- Record History: Unknown — An audit trail of queries and deletions is cited; entity versions are not. ([source](https://tilores.io/deployment))
- Hierarchy Management: No — Graph edges link matched records; no hierarchies. ([source](https://tilores.io/))

### Domains & Modeling

- Multi-Domain MDM: Yes — One entity graph covers people, companies and transactions, with MDM as a use case. ([source](https://tilores.io/))
- Configurable Data Model: No — Only matching rules are configurable. ([source](https://tilores.io/))
- Product Information Management: No — No PIM. ([source](https://tilores.io/))
- Reference Data Management: No — No code-list management. ([source](https://tilores.io/))
- Master Data Authoring: No — No UI to create or edit master records. ([source](https://tilores.io/))

### Data Quality

- Validation at Entry: No — No validation rules on create or edit. ([source](https://tilores.io/))
- Standardization: Unknown — Output shows normalized values (E.164 phones); a cleansing step isn't described. ([source](https://tilores.io/))
- Third-Party Enrichment: No — Connectors are sources (Salesforce, HubSpot, Snowflake), not enrichment providers. ([source](https://tilores.io/pricing))

### Data Distribution

- REST API: No — "GraphQL-native"; no REST API is mentioned. ([source](https://tilores.io/))
- Change Events: Yes — "Webhooks and event streaming". ([source](https://tilores.io/pricing))
- Warehouse Write-Back: Yes — "Resolved entity events flow out to your data warehouse" (Snowflake connector). ([source](https://tilores.io/))
- Data Syndication: Unknown — Entity events flow to warehouses and webhooks; no channel publishing. ([source](https://tilores.io/))

### AI & Agents

- MCP Server: Unknown — No MCP server is mentioned. ([source](https://tilores.io/pricing))
- AI Stewardship Assistance: No — No steward-facing AI. ([source](https://tilores.io/))

### Platform Native

- In-Place Matching: No — Rules apply "after records have been submitted into your Tilores instance", which can run in your own AWS account. ([source](https://docs.tilotech.io/tilores/rules/))
- Snowflake Native: No — Snowflake is a prebuilt connector. ([source](https://tilores.io/pricing))
- Databricks Native: No — AWS-native; no Databricks runtime. ([source](https://tilores.io/))
- Microsoft Fabric Native: No — No Fabric runtime. ([source](https://tilores.io/deployment))

### Deployment

- SaaS Deployment: Yes — Managed Cloud is "Fully managed by Tilores". ([source](https://tilores.io/deployment))
- Self-Hosted Deployment: Yes — "The same engine, deployed inside your own infrastructure." ([source](https://tilores.io/deployment))

### Open Source

- Open Source: No — The engine is patented and proprietary; only tooling repos are Apache-2.0. ([source](https://docs.tilotech.io/tilores/))

## Integrations

6 integrations, each sourced from the vendor docs.

- CRM: Salesforce, HubSpot
- Data Storage: Amazon S3
- Data Warehouse: Snowflake
- Databases: Amazon DynamoDB
- Streaming: Amazon SQS

## Alternatives

- [Senzing](https://toolsfordata.com/tools/senzing/): The closest alternative: real-time entity resolution as an SDK in your own infrastructure, with built-in rules and no model training.
- [AWS Entity Resolution](https://toolsfordata.com/tools/aws-entity-resolution/): Consider AWS Entity Resolution for managed matching on S3 data priced per record, with incremental or near real-time rule-based runs.
- [Quantexa](https://toolsfordata.com/tools/quantexa/): Consider Quantexa when resolved entities feed graph analytics and investigations for financial crime and risk.
- [Zingg](https://toolsfordata.com/tools/zingg/): Consider Zingg for batch matching inside Databricks, Fabric or Snowflake; its paid editions add a persistent ZINGG_ID and incremental runs.
- [Reltio by SAP](https://toolsfordata.com/tools/reltio/): Consider Reltio for a full SaaS MDM hub that also matches in real time, with survivorship and steward review.
