TL;DR Short answer: Amplitude data comes in two kinds. Raw events are what you want in a warehouse, while active user counts and session lengths are Amplitude's own calculations and will not exactly match numbers you compute yourself. For very high event volumes, Amplitude's native warehouse exports are the most direct route; ELT tools suit Amplitude alongside other sources. Seven options:
Airbyte Amplitude warehouse exports Fivetran Hevo Data CData Sync Estuary Custom code with the Export API The Amplitude constraint: the Export API returns events in hourly batches with a processing delay and has size and rate limits, so very large projects backfill slowly through the API and the most recent hours always lag. Amplitude ETL tools move raw events, cohorts and usage metrics out of Amplitude and into a warehouse, where product behaviour can be joined with revenue, CRM and support data. Amplitude answers product questions quickly; a warehouse is where product usage meets the rest of the business.
Why move Amplitude data into a warehouse? Amplitude excels at self-serve product analysis. Teams also need its events elsewhere: feature adoption joined with subscription and expansion revenue, product-qualified leads scored and sent to sales, churn models combining usage with billing and support, and full event history held under their own retention and governance rules. That requires raw Amplitude events in a warehouse.
How do the Amplitude ETL tools compare? Tool Best for Raw events Deployment Pricing model Airbyte Amplitude with other sources Yes, incremental Cloud, self-hosted, Enterprise Flex Free self-hosted; Cloud volume-based; Pro and Flex capacity-based Amplitude warehouse exports High-volume raw event export Yes, scheduled export Managed by Amplitude Depends on Amplitude plan Fivetran Managed warehouse pipelines Yes Cloud, hybrid Monthly active rows Hevo Data No-code pipelines Yes Cloud Events CData Sync Relational tables, self-hosting Depends on configuration Cloud or self-hosted Subscription by connector Estuary Streaming alongside other sources Yes Cloud, private deployment Data volume Custom code Specific, narrow exports Yes Wherever you run it Free, engineering time
Which Amplitude ETL tools should you consider? Seven options, from warehouse ELT to Amplitude's own exports:
1. Airbyte Airbyte is an open-source ELT platform with 700+ connectors. Its Amplitude source syncs raw events through Amplitude's Export API, along with cohorts, annotations, event types, active user counts and average session length, with incremental sync on the time-based streams.
You authenticate with your project's API key and secret key and choose the data region, Standard or EU. Airbyte fits when Amplitude needs to sit next to billing, CRM and support data in one warehouse, self-hosted, on Airbyte Cloud or as Enterprise Flex.
Pros Cons Raw events plus cohorts and annotations Bound by Amplitude's Export API limits Incremental sync on time-based streams Large projects backfill slowly through the API Amplitude next to billing, CRM and 700+ other sources Usage metrics follow Amplitude's own definitions
2. Amplitude warehouse exports Amplitude can export raw event data on a schedule directly into warehouses such as Snowflake and BigQuery, and into cloud storage such as Amazon S3. Because it runs inside Amplitude rather than through the public API, it is the most direct way to keep a complete copy of very high event volumes. Availability depends on your Amplitude plan.
Pros Cons Built for high event volumes Availability depends on your Amplitude plan No API limits to manage Amplitude data only, with no other sources Managed by Amplitude Fewer options for scheduling and transformation
3. Fivetran Fivetran is a fully managed ELT service with an Amplitude connector that loads events and related data into a standard warehouse schema, handling authentication and schema changes for you. Fivetran completed its merger with dbt Labs in June 2026.
Pros Cons Fully managed, with a standard schema Monthly active row pricing grows with event volume Handles schema changes automatically Connector behaviour cannot be modified Pairs naturally with dbt for modelling Warehouse-only destinations
4. Hevo Data Hevo Data is a no-code managed pipeline platform with 150+ connectors, including Amplitude. It can transform data before it loads, using Python or a drag-and-drop editor, and prices on events, with a free tier for small volumes.
Pros Cons No-code setup for product and growth teams Event-based pricing grows quickly with Amplitude volume Transformation before load Cloud-only, with no self-hosted option Free tier for small projects Smaller catalogue than Airbyte or Fivetran
5. CData Sync CData Sync is a replication tool built on CData's drivers, which expose Amplitude data as relational tables and load it into common databases and warehouses. It can run in the cloud or on your own infrastructure, which suits teams with strict data residency needs.
Pros Cons Presents Amplitude data as relational tables Commercial licence priced per connector Cloud or self-hosted deployment Smaller ecosystem than the largest ELT platforms Many database and warehouse destinations Less community content than open-source tools
6. Estuary Estuary captures Amplitude data into collections and materialises it into warehouses and databases, with latency set anywhere from near real time to batch. It suits teams that want Amplitude alongside database change data capture and other streaming sources in one platform.
Pros Cons Streaming and batch from one platform Smaller connector catalogue than the largest ELT tools Many warehouse and database destinations Pricing on data volume needs modelling Private deployment option Still bound by Amplitude's Export API delay
7. Custom code with the Export API Amplitude's Export API returns raw events for a time range as compressed files, and the Dashboard REST and Behavioral Cohorts APIs cover metrics and cohorts. Custom code suits narrow, specific needs, but you own batching, retries, size limits, deduplication and scheduling.
Pros Cons Full control over what you export You own retries, batching and scheduling Raw events in Amplitude's own format Size and rate limits on large ranges No licence cost Ongoing maintenance
What should you check before syncing Amplitude? Most Amplitude pipeline problems come from regions, timing and definitions. Check these first:
Data region: Amplitude projects are hosted in the standard US region or the EU region, each with its own API endpoints. Set the region that matches your project, or requests will fail.API keys: each Amplitude project has its own API key and secret key. Multiple projects, such as separate apps or environments, need one connection each.Export delay and limits: the Export API serves events in hourly batches after a processing delay and limits how much data one request can return. Expect recent hours to lag, and split very large backfills into smaller time ranges.Computed metrics: active user counts and average session length are Amplitude's own calculations. Numbers you compute from raw events will differ slightly, so decide which definition your reports use.Duplicates and identity: events can be resent, and users can have several device IDs before they log in. Deduplicate on the event's insert ID and model user identity in the warehouse.Personal data: event properties often include user identifiers and sometimes personal data. Restrict access in the warehouse and drop properties you do not need.Amplitude's APIs expose raw data, definitions and computed metrics:
Raw events: every event with its properties, user and device IDs, timestamps and insert ID. User properties: attributes attached to users at the time of each event. Cohorts: behavioural cohort definitions and, through the Cohorts API, their members. Active users and sessions: Amplitude's daily, weekly and monthly active user counts and average session length. Annotations: chart annotations marking releases and campaigns. Event types: the catalogue of tracked events, useful for documenting your tracking plan. How do you start pulling data from Amplitude? If you decide to test Airbyte, setting up an Amplitude pipeline takes three steps.
Step 1: Set up Amplitude as a source In Amplitude, open your project settings and copy the project's API key and secret key. In Airbyte, open Sources, choose New source and select Amplitude , enter both keys, choose the data region and set a start date for the first sync.
Step 2: Set up a destination Pick where your Amplitude data should land: a warehouse such as Snowflake, BigQuery or Databricks, a database, or another Airbyte destination.
Step 3: Configure the connection Select events, cohorts, annotations and the usage metric streams you need, use incremental sync, and schedule daily runs. The process is the same whether you deploy Airbyte Open Source or start with Airbyte Cloud .
Which Amplitude ETL tool should you choose? For very high event volumes where Amplitude is the only source, use Amplitude's native warehouse exports if your plan includes them. For Amplitude joined with billing, CRM and support data, use Airbyte, or Fivetran or Hevo for a managed pipeline. CData Sync suits teams that need relational tables or self-hosting, Estuary suits streaming stacks, and custom code covers narrow, specific exports.
For a wider view of the market, see our guide to the top data integration tools .
Sync Amplitude to your warehouse with Airbyte Replicate Amplitude raw events, cohorts and usage metrics incrementally, next to your billing, CRM and 700+ other sources, self-hosted, on Airbyte Cloud or with Enterprise Flex.
Try Airbyte Cloud free for 14 days or talk to our team .
Frequently Asked Questions (FAQs) 1. How do you export raw Amplitude events to a data warehouse? Use an ELT tool such as Airbyte, Fivetran or Hevo with your project's API key and secret key to sync events through Amplitude's Export API. For very high volumes, Amplitude's native warehouse exports send raw events directly to Snowflake, BigQuery or cloud storage, depending on your plan.
2. Why doesn't my active user count match Amplitude's? Amplitude's active user counts use its own definitions of users, sessions and time zones. Counts computed from raw events will differ slightly, so pick one definition for reporting and document it.
3. Why are the latest Amplitude events missing from my warehouse? Amplitude's Export API serves events in hourly batches after a processing delay, so the most recent hours are not yet available. They will appear in a later sync.
4. Can one connection sync several Amplitude projects? No. Each Amplitude project has its own API key and secret key, so you need one connection per project. Label each one so the data can be combined in the warehouse.
Suggested reads Mixpanel ETL Tools
Google Analytics ETL Tools
Data Analytics Tools
Reverse ETL Tools
.webp)