The 500,000 free MAR allowance is available only on the Free Plan. After upgrading to a paid plan (Standard, Enterprise, or Business Critical), usage is billed based on actual MAR consumption at the plan rate.

You can subscribe to Fivetran directly through the AWS Marketplace. Simply search for Fivetran, select your preferred plan, and complete the subscription to link billing to your AWS account.

Subscribe through the Google Cloud Marketplace.. Search for Fivetran, select a pricing plan, and enable billing through your Google Cloud organization.

Yes. You can switch from monthly, pay-as-you-go credit card billing to an annual invoiced contract in two ways:

• Through the Fivetran product: Build and initiate an annual contract directly in the Fivetran UI.
  
• Through your renewal manager: Your renewal manager can assist with setting up and finalizing an annual contract.
  

Note: Annual contracts require a minimum spend of $12,000 and may include discounts based on commitment level.

See the Fivetran pricing page for a full comparison of plan features.

You can view and customize a pricing estimate by using the Pricing Estimator or directly in the Billing & Usage dashboard under Account Settings.

The monthly estimate projects expected MAR-based spend for the current month. Because billing is based on actual usage, final costs may vary depending on how data changes over time. To better predict costs, monitor the Billing & Usage dashboard and use the Pricing Estimator to model expected MAR and sync frequency.

No. Fivetran provides free resyncs. You are billed only for data that has changed. Monthly Active Rows (MAR) include rows that are inserted, updated, or deleted during the billing cycle. If a resync modifies rows, those changes count toward your MAR total.

Not directly. Monthly Active Rows (MAR) is based on the number of unique rows inserted, updated, or deleted, not on sync frequency. However, higher sync frequency may require an upgrade to a more advanced plan, such as Enterprise or Business Critical.

Potentially. Every plan includes 5,000 free transformation model runs per month. If usage exceeds this limit, additional model runs are billed. On the Free Plan, exceeding 5,000 runs requires an upgrade to continue using transformations.

• Syncs: Scheduled jobs that pull data from source systems.‍
• Monthly Active Rows (MAR): The number of unique rows inserted, updated, or deleted across all syncs in a month.

Billing is based on MAR, not on the number of syncs.

You can reduce MAR and overall costs by:

• Excluding nonessential tables or columns from connectors
  
• Pausing unused connectors
  
• Reducing transformation model runs
  
• Evaluating whether a lower pricing tier better fits current needs
  
• Using the Billing & Usage dashboard to identify high-MAR connectors and optimize them

Quickstart data models are pre-built dbt models that enable data teams to transform raw data into production-ready tables without writing SQL or setting up a dbt project. With just a few clicks, they simplify and accelerate analytics, eliminating the need for third-party tools or manual coding. 

Key benefits include:

• Time savings: Data teams can instantly lay the groundwork for analytics without writing any code or building a dbt project
• Accelerated insights: Our managed approach frees your data team to focus on high-impact projects, allowing them to drive insights instead of managing pipelines or building foundational models.
• Operational efficiency: Integrated scheduling ensures real-time data by automatically triggering model runs after Fivetran connector syncs. This reduces unnecessary compute costs, eliminates manual logic building, and minimizes data latency.

Fivetran offers both single-source and multi-source Quickstart data models. Multisource data models, such as for social media ads, integrate and merge data from connections of a similar type. 

Single-source data models are enabled by default for eligible connections and incur the same charges as any transformations run through Fivetran. You can disable them at any time.

Some tables are linked through schema dependencies in your source schema. Fivetran automatically includes related parent or child tables to preserve data integrity and relational structure. These tables cannot be deselected individually.

Follow the Snowflake connector onboarding video for a step-by-step guide to connecting Fivetran to Snowflake.

The right choice depends on your use case:

• Data warehouses (for example, Snowflake, BigQuery, Redshift) are best for structured, analytics-ready data.‍
• Data lakes (for example, Amazon S3, Azure Data Lake) are well suited for large volumes of unstructured or semi-structured data, including machine learning workloads.

All Fivetran security and compliance documentation, including SOC 2, HIPAA, and GDPR certifications, is available on the Security & Trust Portal