Integration & ETL Plane · ELT & Ingestion

Fivetran

Managed ELT connectors that replicate data from SaaS applications and databases into warehouses and lakes.

Overview

Fivetran is managed ELT, and the word that matters is managed. It owns the extract and load from source to destination schema, normalising data into a clean relational shape and absorbing schema drift on your behalf; you own the transformation and modelling on the other side.

Pipelines are not code by default. A connection is configured in the dashboard, source plus destination plus tables plus sync frequency, with a REST API and a Python connector SDK available when you want to automate it.

Connector coverage is the headline claim, though the numbers differ by page: the home page says 900 sources and destinations, the connectors page 750 sources and 200 destinations.

Connectors come in four published types, standard, lite for narrower SaaS sources, partner-built, and ones you write yourself with the SDK, which is available on every plan.

What distinguishes it commercially is the billing unit. You pay per monthly active row, a distinct row counted once however often it changes, rather than per compute hour, credit or seat.

Sync frequency runs from one minute to 24 hours, with one-minute syncs gated to Enterprise and above, the tightest published floor among the mainstream ELT tools.

History Mode is a genuine differentiator: built-in type 2 slowly changing dimensions, adding active, start and end columns automatically. Fivetran warns plainly that it raises row counts and therefore cost.

Transformation is included but metered separately by model run, covering no-code quickstart models, dbt projects hosted by Fivetran, dbt Cloud and Coalesce.

Two corporate facts now shape the product. Fivetran completed an all-stock merger with dbt Labs on 1 June 2026, making transformation a first-party capability, and it acquired Census in 2025, which is now Fivetran Activations.

It is entirely proprietary. There is no open-source core, no source-available licence and no free self-hosted edition; the free plan is a usage allowance, not an open product.

Key features and capabilities

The same headings are used for every elt & ingestion entry, so two tools can be read side by side.

Connectors and coverage
  • Counts vary by page, 900 sources and destinations on the home page, 750 sources on the connectors page
  • Four published types, standard, lite for narrower SaaS sources, partner-built, and SDK connectors you write
  • Release status is published per connector, from private preview through beta to generally available
  • Sources span SaaS applications, databases, SAP, event streams, file storage, functions and logs
How pipelines are built
  • Configured in the dashboard as a connection rather than defined in code
  • Sync frequencies from 1 minute to 24 hours, with one-minute syncs needing Enterprise or above
  • Transformation is included, covering no-code models, Fivetran-hosted dbt, dbt Cloud and Coalesce
  • A REST API covers frequency, start time, hybrid agents and SDK connections
Loading and sync behaviour
  • Incremental by cursor, with re-sync available to refetch everything from scratch
  • Deduplication by primary key, with a synthetic hashed key created when the source has none
  • History Mode gives type 2 slowly changing dimensions through active, start and end columns
  • History Mode is unsupported for Kafka destinations and significantly raises row counts and cost
Destinations
  • Warehouses including BigQuery, Snowflake, Redshift, Azure Synapse and Databricks
  • Databases including PostgreSQL and SQL Server, with MySQL and Oracle in beta
  • The Managed Data Lake writes Parquet with Iceberg or Delta metadata into your own storage
  • Catalogues include Fivetran's own managed Polaris implementation, AWS Glue and Unity Catalog
Security, monitoring and lineage
  • Certifications include SOC 1 and 2, ISO 27001, PCI DSS Level 1, HIPAA and HITRUST on Business Critical
  • Column blocking stops a sensitive column replicating; column hashing anonymises while keeping it joinable
  • Customer-managed keys, with staff access to data requiring explicit and revocable approval
  • Private connectivity through SSH, VPN and cloud private links, the last on Business Critical
  • Lineage is not published as a named feature
Extending it yourself
  • A Python connector SDK, available on all plans and billed on rows like any other connector
  • Published plugins that add SDK skills to Claude Code and other AI coding agents
  • REST API for connections, agents and package management
  • dbt Core orchestration, dbt Cloud orchestration and Coalesce support
Where it runs and what it costs
  • Managed software as a service, with the cloud provider selectable on Enterprise
  • Hybrid deployment processes data inside your own network on Kubernetes, Docker or Podman
  • Hybrid deployment is included at no extra cost on Enterprise and Business Critical
  • Billed on monthly active rows, plus model runs for transformation

Pricing

Usage-basedFree to 500,000 monthly active rows

Billed on monthly active rows, a distinct row counted once a month however often it changes, per connection. The free plan covers 500,000 rows plus 3,500 for activations and 5,000 model runs. Standard adds a $5 base charge per connection below a million rows; Fivetran's own worked examples put Google Analytics at $15.88 a month and Marketo at $424.54. Transformations are free to 5,000 runs then $0.01 falling to $0.002. No per-row rate table is published for any tier, so Enterprise is effectively quote-only. Trials run 14 days, per connection.

Vendor pricing page →

Demos and videos

About Fivetran

Fivetran was founded in 2012 by George Fraser and Taylor Brown and is based in Oakland, California. It is private, venture-funded through a $565m Series D at a $5.6bn valuation, and has bought growth as well as built it, acquiring HVR for $700m in 2021 and Census in 2025. In June 2026 it completed an all-stock merger with dbt Labs, with Fraser as chief executive and dbt's Tristan Handy as president, and the combined company claims more than 100,000 data teams. Its last published revenue figure, a $200m run rate, dates from 2023.

Founded 2012 · Oakland, California · fivetran.com

Other elt & ingestion tools

Airbyte

Integration & ETL Plane · ELT & Ingestion

Data integration platform with a large connector catalogue, self-hosted or as a cloud service.

  • Open core

AWS Glue

Integration & ETL Plane · ELT & Ingestion

Serverless data integration service on AWS for discovering, preparing and moving data.

  • Cloud service

Azure Data Factory

Integration & ETL Plane · ELT & Ingestion

Azure's managed service for data integration and ETL/ELT pipelines.

  • Cloud service

Boomi

Integration & ETL Plane · ELT & Ingestion

Enterprise integration platform spanning application and data integration, API management and agentic workflow orchestration, with an embedded option for building integration into your own product.

  • Commercial

Data Factory in Microsoft Fabric

Integration & ETL Plane · ELT & Ingestion

Pipelines and Dataflow Gen2 for ingesting and transforming data in Microsoft Fabric.

  • Cloud service

dlt

Integration & ETL Plane · ELT & Ingestion

Open-source Python library for loading data from APIs and files into warehouses and lakes.

  • Open source

FME

Integration & ETL Plane · ELT & Ingestion

Integration platform from Safe Software handling any data format, with particular depth in spatial and engineering data that general ELT tools do not cover.

  • Commercial

Matillion

Integration & ETL Plane · ELT & Ingestion

Cloud data integration and transformation platform for building pipelines on cloud data platforms.

  • Commercial

Snowplow

Integration & ETL Plane · ELT & Ingestion

Behavioural data pipeline that collects, validates and enriches customer events into your own warehouse. Now under the Snowplow Limited Use Licence, which restricts commercial use of the free edition.

  • Source available

Drafted with AI assistance and checked against the vendor’s own documentation.