Which metric is most relevant for evaluating improvements after optimizing a slow query?

The most relevant metric for assessing improvements after optimizing a slow query is response time or latency — how long it takes to execute the query from the moment it is sent until the result is received.

Measuring execution time helps understand how much the optimization has reduced delays and sped up the system. It is also useful to look at throughput if queries are processed in parallel.

Example: if before optimization, the query took 2 seconds, and after — 500 milliseconds, then the optimization is successful.