Are you looking for a way to speed up queries in Snowflake without altering the size of the warehouse?

One way you might achieve this is by taking the largest table in your query and applying the necessary filters before you start using it in your query. This means that the aforementioned largest table that contained hundreds of millions of records before the filters were applied, suddenly only contains a few million records.

Now all the operations you were going to be doing with this initially HUGE table only have to applied to a relatively small subset of rows of the table. This speeds things up tremendously.