08FAQFAQ
Marketing ops and analysts usually ask 4 things first: SQL, credits, Claude, and which tables ChatGPT sees. Then come 3 fixes: numbers that don't match Snowsight, a connection Snowflake refuses, and slow answers.
Do I need to know SQL?
No. Porter reads your table's columns and hands them to ChatGPT, so you ask for REVENUE by week and ChatGPT builds the query. Name the table and the date column, and write column names as they appear in Snowflake, so ChatGPT matches them to the fields Porter lists.
Does ChatGPT use Snowflake credits?
Yes. Every question runs on your Snowflake warehouse, and a suspended warehouse that wakes up bills at least 60 seconds, then per second (Snowflake's warehouse billing rules). Snowflake can reuse an identical query's result for 24 hours when the data hasn't changed (Snowflake's result cache).
Can I use Porter in Claude too?
Yes. Porter's MCP URL, the same one behind the Snowflake connector for ChatGPT, also works in Claude as a custom connector. Then Claude reads the same Snowflake tables through Porter, behind the same role. See Porter's Snowflake connector for Claude for the Claude setup and what it covers.
Can ChatGPT see every table in my account?
ChatGPT sees the tables that the role you give Porter can read. Porter lists all of those, and ChatGPT queries only the ones you ask about. Each table ChatGPT queries counts as one Porter account, so a narrow role keeps both the list and the bill short.
Troubleshooting
Why don't the numbers match Snowsight?
Two Snowflake settings can make ChatGPT's numbers different from Snowsight's. TIMESTAMP_LTZ columns show in the session time zone, which defaults to America/Los_Angeles (Snowflake's TIMEZONE parameter), so daily totals can shift. Unquoted column names are uppercase, so "Revenue" and REVENUE are different columns (Snowflake's identifier rules).
Why can't Porter connect to my account?
Check three Snowflake settings first. Snowflake disallows passwords for service users and asks human users with passwords for MFA (Snowflake's MFA and password rollout). A network policy's allowed list refuses every IP missing from it (Snowflake's network policies), and the role needs USAGE on a warehouse.
Why is a question slow?
Ask again once the warehouse is running, or point Porter at a warehouse with room to spare. A Snowflake question is slow while Porter waits for a suspended warehouse to resume, and the first queries after that run on a cold cache. A busy warehouse also queues statements beyond its concurrency limit.