Usage and Pricing Rules for Contracts Signed Before March 1, 2025
The following information applies to you if you are a Fivetran customer that signed an annual contract before March 1, 2025.
NOTE: All customers who have signed up for Fivetran after March 1, 2025 benefit from our updated free re-sync detection mechanism. See our 2025 Pricing Updates documentation for more details.
Plans
Apart from the plans available after the 2025 pricing update, you may be using one of the deprecated plans:
- Starter - Access our entire library of SaaS, events, and files connectors. This plan is best for small teams not yet ready to connect a database source.
- Private Deployment - Deploy Fivetran's HVR as a standalone solution if you require highly secure connections. This usage-based subscription plan requires you to send the consumption data files (MAR) to Fivetran at least once a quarter.
NOTE: The Starter plan has a maximum user capacity of 10 users.
Plan renewal
If you have signed an annual contract with Fivetran before March 1, 2025, the usage and pricing updates will be automatically applied to your account. If you are on the Starter or Private Deployment plans, you will have to transition to another plan that is still available. See our Usage-Based Pricing documentation for a list of available plans.
How can I find out which plan I'm on?
To find out which plan you're on, do the following:
- Log in to your Fivetran dashboard.
- Go to Account Settings > Billing & Usage.
- Click the Plans tab to see your plan information.
Only users with the following user roles can access the Billing tab:
- Account Administrator
- Account Billing
Account-wide usage tiers
Your MAR is calculated at an account level. In other words, the more you use on an account level, the less you pay per millions of activated rows.
This differs from the new scaling, which is applied at the connection level. See our 2025 Pricing Updates for more information.
Re-syncs
The following re-syncs count towards free MAR:
- Manual re-syncs related to initial connection setup, debugging, and troubleshooting purposes are free. This includes table-level and connection-level re-syncs and historical re-syncs initiated from the Fivetran dashboard or using our REST API. You may trigger unlimited re-syncs for these purposes each month. However, customers cannot trigger re-syncs to skirt their obligations under their signed Agreement with Fivetran.
- Re-syncs triggered by Fivetran, such as for maintenance or incident resolution, count towards free MAR. We only perform a re-sync when it is necessary.
- Automatic table re-syncs triggered by database connectors as part of their sync strategies or due to schema changes are free.
This is a more limited re-sync mechanism compared to the one in use after the March, 2025 update.
Re-imported tables
When a table is re-imported on the first day of the month, the entire table counts towards your MAR (free and paid). For all subsequent days that month, only new records count towards your MAR.
Any re-imported table that is included in an initial sync counts towards free MAR on the day of the initial sync. Re-imported tables are counted as incremental rows on the day after the initial sync.
For example, see how re-imported tables are classified before a free trial, during a free trial, and on the first day of the month after a free trial:
- Month 1, Day 1: An initial sync begins, and a re-imported table syncs 100 rows on this day. Moreover, the re-imported table will sync a new row each day. On this day, 100 rows are marked as counting towards free MAR.
- Month 1, Day 2: The initial sync has not been completed, and the re-imported table synced one additional row. In total, 101 rows are marked as counting towards free MAR.
- Month 1, Day 3: The initial sync has ended, and the free trial has started. On this day, 102 rows are re-classified as incremental.
- Month 1, Day 16: The free trial ends today. We logged 102 rows as incremental on Day 3, and one additional row each day since then, for a total of 115 rows.
- Month 1, Day 17: We update the incremental rows the day after the free trial ended, adding the 14 rows that the re-imported table synced every day since the free trial started. On this day, 116 rows are re-classified as incremental.
- Month 1, Day 31: On this day, 116 rows are still classified as incremental, however the re-imported table synced an additional 14 rows since Day 17.
- Month 2, Day 1: The re-imported table synced an additional 15 rows since Day 17. On this day, we mark 131 rows as counting towards MAR, leading to an increase in MAR.
NOTE: For re-imported table syncs that occur when no cursor column is provided, we consider the changes as incremental MAR.
History mode tables
Every time a record's value in a source table with history mode enabled changes, we insert a new row in the destination table. This new row counts towards MAR. Your MAR usage depends on the number of tables with history mode enabled and how frequently the data changes in your source system.
MAR from historical data in priority-first syncs
A connection's first sync counts towards free MAR. All other syncs that run on the priority sync feature, including syncs that fetch backdated/historical data, count towards paid MAR.
IMPORTANT: Connectors that support priority-first sync activate your connections' free trial period as soon as a priority-first sync is initiated and the system detects incremental data.
Automated schema migrations
When we automatically add a column to a connector schema as part of an automated schema migration, every row in the table that is backfilled with data counts as an active row. Rows that do not have backfilled data and have a null
value for the new column are not considered active.
Deletes in the source
If a connector supports the Capture Deletes feature, deletes in your source count towards your MAR. If you delete data from your source, we soft delete the corresponding data in your destination by setting the system column _fivetran_deleted
to TRUE
. As the delete corresponds to a row update, it counts towards your MAR.
XMIN system columns in PostgreSQL MAR increase
The value of the xmin
column has an upper limit of 4,294,967,295. After the database's total number of transactions exceeds this value, it "wraps around" to zero. In PostgreSQL versions 9.5 and later, after this event occurs, multiple rows may have the same xmin
value, and Fivetran may additionally sync older, unchanged "frozen" rows, resulting in more Monthly Active Rows (MAR) and higher cost. See our FAQ page Why Is the XMIN Wraparound Causing MAR Spikes? for more information on resolving this issue.