Sometimes your Snowflake query takes 2 or 9 minutes, and now it is taking around 45 minutes. How do you troubleshoot this performance issue?

I would start by examining the query profile in Snowflake. First, check the execution plan for any stages that are taking disproportionate time, such as data scans or joins. Look for missing or outdated clustering keys that could cause full table scans. Verify that the warehouse size is appropriate; if the query is CPU‑bound, scaling up may help. Inspect the number of concurrent queries and see if contention is causing delays. Check for any recent schema changes or data growth that might have increased the amount of data processed. If the query uses a materialized view, ensure it is refreshed. Finally, consider rewriting the query to use more selective predicates, push filters earlier, or replace expensive joins with semi‑joins. Document each change and measure the impact to isolate the root cause.