Developers who would never ship a method they had not read will happily ship a query they have never explained. The plan is not exotic; it is a tree of decisions with row estimates attached.

Read the estimates first

Compare estimated rows with actual rows. When they differ by orders of magnitude the planner is working from bad statistics and every choice downstream is suspect. That single comparison explains most surprises.

• A sequential scan on a large table is not automatically wrong; on a query returning most of the table it is correct.
• A nested loop over a bad estimate is how a fast query becomes a timeout after a data migration.
• Sorts that spill to disk are visible in the plan and invisible in the code.

Capture the plan for your important queries in a test. When it changes shape you want to hear about it from CI, not from a customer.