Alteryx Designer vs MS SQL Server aggregation differences

I am working on converting a data pipeline that runs on Alteryx Designer into a MS SQL Server query, and I am noticing that I get different results when I use the GROUP BY step. For example, both processes read from the same table that has 2,318,794 rows, but when I use aggregate functions (GROUP BY, MAX(), COUNT() and SUM()) they Alteryx process returns 2,089,738 rowsand the SQL query returns 2,089,238 (a 500 row difference) and the difference increases the more I aggregate, to the point that by the end of the process I end with a 60k lines difference between what Alteryx returns and what my query returns.

Is there a difference on how Alteryx and MS SQL Server execute aggreation operations? Or how they handle NULL values when aggregating?

Any input or idea is appreciated, thank you.