Replication frequencies, which dictate how often Stitch attempts to replicate data into your data warehouse, are defined when initially setting up an integration. For the majority of integrations, you get to decide how and when your data is replicated.
It all comes down to this: smaller frequencies could potentially lead to more data usage. The more often the frequency, the more continuous the replication, resulting in more data consumption.
Currently the frequency setting applies to all tables in a connection currently being synced to Stitch. Therefore, when deciding what frequency to use, it's especially important to consider the replication methods you set for your tables. If you select a smaller frequency, tables using Full Table Replication could quickly eat up your monthly data allowance.
Consider this: we have a table containing 2,000 rows during the initial sync of our database. After the first sync completes, we set the table to Full Table Replication and the replication frequency to 5 minutes. If all replication attempts are successful, in one hour we'd use 24,000 rows; in one day, a whopping 576,000! At the end of one week, we'd be close to using up the 5 million rows included in the Free tier. This is the result of a single table - if we were syncing other data, we'd have to change our tier or pause the connection to avoid overages.
In addition to how your integration is configured, evaluate your data needs. Do you constantly need your data to be replicated? Can tasks still be completed with data that's replicated less often? If so, consider using a smaller frequency in conjunction with an Incremental Replication method.
The same idea holds true for SaaS integrations - the faster the frequency, the potential for more data consumption. It's important to note that the majority (but not all) of SaaS integrations are automatically set to incrementally replicate.
When setting the frequency for a SaaS integration, we recommend you also consider how many tables will be replicated. Integrations with potentially high table counts will lead to increased data consumption if set to replicate on a more frequent basis.
To get an idea of what's in each integration, check out the Expected SaaS data section.
When initially setting up an integration, you'll be asked to select a replication frequency. However, this setting can be changed at any time by going into the Integration Details page and clicking the Integration Settings button.
Select the desired frequency and click Save Integration. Note that may take a few moments to save your changes.