What Counts as a Data Source in DataBlueprint?
Each unique connected system is one source — regardless of how many tables, objects, or files it exposes.
Pricing pages love the word source without ever defining it. Here is exactly how DataBlueprint counts.
The rule
Each unique connected system equals one source.
- One Salesforce org = one source.
- One Snowflake warehouse = one source — regardless of how many schemas or tables it exposes.
- One QuickBooks file = one source.
- One Google Drive folder of CSVs = one source.
Why this matters
Most tools meter by table, object, or row. That punishes you for having a normal warehouse. DataBlueprint meters by system, because the integration cost — credentials, IP allow-listing, permissions, refresh scheduling — happens once per system, not once per table.
What about multiple environments?
A production Salesforce and a sandbox Salesforce count as two sources. Two Snowflake accounts count as two. The rule follows the credential boundary, not the brand.
Read-only, always
Whatever you connect, the connection is read-only. DataBlueprint cannot write back to the source. The Knowledge Graph is built from what we read; nothing is pushed in the other direction.
How limits work
Each tier has a source cap. If you exceed the cap, the extra connections go read-only until you upgrade or remove a source. You never lose the historical context — the graph keeps the data it has already learned.
See the pricing page for current source caps by tier.
Frequently Asked Questions
What about multiple environments?
A production Salesforce and a sandbox Salesforce count as two sources. Two Snowflake accounts count as two. The rule follows the credential boundary, not the brand.