Fivetran

by Fivetran Inc

Cloud Free Tier freemium API Available
EnterpriseData Pipeline

Automated data integration platform for analytics pipelines. Fivetran is a managed extract-load-transform (ELT) data integration platform that ingests data from operational systems into cloud data warehouses without requiring customers to write or maintain pipeline code. Founded in 2012 in Oakland by George Fraser and Taylor Brown, both ex-McKinsey, Fivetran shipped its first connector to Snowflake in 2017 and has expanded the connector library to more than 600 named sources by Q2 2026, covering SaaS APIs, databases (Postgres, MySQL, SQL Server, Oracle, MongoDB), file systems, event streams, and ad platforms.

Performance Scores

8.1

3 rankings evaluated

Score range: 7.8 – 8.6

Key Facts

technical

technical facts about Fivetran
AttributeValueAs ofSource
Connector count (May 2026)600+ named source connectorsMay 2026Fivetran connectors directory

pricing

pricing facts about Fivetran
AttributeValueAs ofSource
Fivetran pricing (July 2026)Fivetran plans (verified July 2026): Free, Standard, Enterprise, and a new Business Critical tier (CMEK, PCI DSS Level 1, private networking); the former "Starter" plan is gone. Billing is per Monthly Active Row (MAR) with a $5 base charge per standard connection (1-1M MAR, not on Free). Free = 500,000 MAR/month plus 3,500 MAR activations and 5,000 model runs. Annual commitments save up to ~22%.Jul 2026Fivetran pricing
Free tier (May 2026)Starter free tier: 500K Monthly Active Rows, 1 user, 14-day historyMay 2026Fivetran pricing page
Standard MinimumStandard plan starts at approximately $500/month (consumption-based, MAR pricing)Apr 2026Fivetran Pricing
Free TierFree plan with up to 500,000 monthly active rows (MAR), suitable for small pipelinesApr 2026Fivetran Pricing

business

business facts about Fivetran
AttributeValueAs ofSource
Last disclosed valuation$5.6B post-money, $565M Series E (September 2021, Andreessen Horowitz led)May 2026Fivetran Series E announcement

General

General facts about Fivetran
AttributeValueAs ofSource
Connectors500+Apr 2026Official Website
Funding$730M total funding (valued at $5.6B in 2021)Apr 2026Company Blog
Pricing ModelMonthly Active Rows (MAR) — billed per distinct row updated per monthApr 2026Official Website
WarehousesSnowflake, BigQuery, Redshift, Databricks, Azure SynapseApr 2026Documentation
Sync FrequencyEvery 5 minutes (Standard+), 6 hours (Free tier)Apr 2026Official Website
Schema MgmtAutomatic schema migration and drift detectionApr 2026Documentation
Notable CustomersSquare, ClassPass, Urban Outfitters, AutodeskApr 2026Case Studies
Employees700+ employees (as of 2025)Apr 2026Company Website
SecuritySOC 2 Type II, HIPAA, GDPR compliant, column-level hashingApr 2026Official Website

capability

capability facts about Fivetran
AttributeValueAs ofSource
Connectors600+ pre-built connectors as of 2026Apr 2026Fivetran Connectors
dbt IntegrationNative dbt Core transformations supported post-load on Fivetran-managed scheduleApr 2026Fivetran Docs

company

company facts about Fivetran
AttributeValueAs ofSource
Customer CountApproximately 3,000+ customers including paying enterprise accounts as of 2026Apr 2026Fivetran Customers

Strengths

  • 500+ pre-built connectors with automated schema migration
  • Log-based change data capture for major databases
  • Strong reliability with documented monitoring and alerting
  • First-class dbt integration via Fivetran Transformations
  • 500+ pre-built connectors with automatic schema handling
  • Zero-maintenance managed pipeline operation
  • Automatic incremental loading and CDC for supported sources
  • Strong data quality monitoring with automatic anomaly detection
  • 750+ managed source connectors with automatic schema handling (July 2026)
  • Reverse ETL via Fivetran Activations (former Census, acquired 2025)
  • dbt Labs merger (completed June 2026) unifies ingestion and transformation

Limitations

  • Consumption pricing on Monthly Active Rows can grow unpredictably
  • Limited control over connector internals — closed source
  • Custom connectors require Fivetran Connector SDK or third-party tooling
  • Expensive at scale due to MAR-based pricing
  • Limited transformation capabilities natively
  • No workflow orchestration beyond extract-load
  • Vendor lock-in with proprietary connector format
  • Monthly Active Rows pricing is hard to predict at high row churn
  • Closed-source managed service; no self-hosted option

Based on evaluations in 3 rankings: Best Data Integration Platforms in 2026, Best Automation Tools for Data Teams in 2026, Best ETL & Data Pipeline Tools 2026

Pricing Plans

View official pricing →

Free

Free

Usage within free limits; no $5 base charge

  • 500,000 MAR/month
  • 3,500 MAR activations
  • 5,000 model runs/month
  • !500K MAR/month
Get started →
As of Jul 2026 · Source
Most Popular

Standard

$/mo

Consumption (per MAR) + $5 base/connection; save up to ~22% annually

  • Full 500+ connector catalog
  • Per-MAR consumption billing
  • !$5 base charge per standard connection (1-1M MAR)
Get started →
As of Jul 2026 · Source

Enterprise

$/mo

Consumption (per MAR), volume discount

  • Private networking, SSO
  • Advanced governance
  • !Custom MAR pricing
Get started →
As of Jul 2026 · Source

Business Critical

Contact us

Highest tier; custom

  • CMEK, PCI DSS Level 1
  • Private networking
  • !Custom MAR pricing
Get started →
As of Jul 2026 · Source

About Fivetran

Fivetran is a managed extract-load-transform (ELT) data integration platform that ingests data from operational systems into cloud data warehouses without requiring customers to write or maintain pipeline code. Founded in 2012 in Oakland by George Fraser and Taylor Brown, both ex-McKinsey, Fivetran shipped its first connector to Snowflake in 2017 and has expanded the connector library to more than 600 named sources by Q2 2026, covering SaaS APIs, databases (Postgres, MySQL, SQL Server, Oracle, MongoDB), file systems, event streams, and ad platforms.

The product architecture is intentionally opinionated. Fivetran owns the schema mapping (column types are inferred and updated on schema drift), the change-data-capture mechanism where supported (log-based CDC for Postgres, MySQL, Oracle, MongoDB), and the destination loading logic. Customers connect a source and a destination warehouse, set a sync frequency (1, 5, 15, 60, 360, or 1440 minutes), and Fivetran handles incremental updates from then on. Native destinations include Snowflake, BigQuery, Databricks, Redshift, Synapse, and Postgres-compatible engines.

Pricing is consumption-based on Monthly Active Rows (MAR), with a free Starter tier (500K MAR/month, single user, 14-day history) and paid tiers Standard, Enterprise, and Business Critical that scale on row volume, connector count, and feature set. As of May 2026 the company's public pricing page indicates Standard plans typically start in the low thousands of USD per month for production volumes; enterprise pricing is custom. Fivetran completed a $565M Series E in September 2021 at a $5.6B valuation led by Andreessen Horowitz, and has not announced a primary round since.

In 2023 Fivetran acquired HVR Software, a CDC-focused database replication vendor, adding higher-volume database-to-warehouse replication to the existing SaaS-API-heavy connector portfolio. The combined platform now positions against Stitch, Airbyte, and warehouse-native ingestion tools (Snowflake Streaming, BigQuery Datastream). Customers include Square, DoorDash, Forbes, Vimeo, JetBlue, Crocs, and the Smithsonian; the company reports 5,000+ paying customers as of 2025 disclosures.

flowchart LR
  A[(Postgres / MySQL / SaaS API)] -->|Fivetran Connector| B[Fivetran Service]
  B --> C{CDC supported?}
  C -- Yes --> D[Log-based incremental]
  C -- No --> E[API polling + schema diff]
  D --> F[(Cloud Warehouse)]
  E --> F
  F --> G[dbt or warehouse SQL transform]

Editor's Note: Across 9 ShadowGen Fivetran deployments in 2024-26, the median month-3 MAR bill landed 38% higher than the pre-deployment projection because operational source rows (especially Salesforce Activity history and Shopify webhook events) compound faster than buyers expect. The mitigation that works: enable connector-level schema exclusions before first sync, not after, since Fivetran bills on rows synced even if columns are later excluded. Where it shines: Postgres CDC over WAL replication is the cleanest production database-to-warehouse path we have shipped, with 15-minute median end-to-end latency on tables under 50M rows. — Rafal Fila, ShadowGen

Integrations (8)

BigQuery native
Databricks native
HubSpot native
Redshift native
Salesforce native
Snowflake native
Stripe native
dbt native

Written & reviewed by · Last updated: · Last verified:

Other ETL & Data Pipelines Tools

See How It Ranks

Questions About Fivetran

Learn More