How can I design an experiment to show the benefits of writing while under Snapshot isolation?

We've all read [the documentation for Snapshot isolation](https://learn.microsoft.com/en-us/sql/relational-databases/sql-server-transaction-locking-and-row-versioning-guide?view=sql-server-ver16#modify-data-without-optimized-locking) and know about Update Conflict Detection and when you should _theoretically_ use Snapshot isolation for writes. However, I have never found anyone who does their writes under Snapshot isolation. Paul White has [a post on how it can go wrong](https://www.sql.kiwi/2014/06/the-snapshot-isolation-level/) , but I have never seen anyone discuss what it looks like when it goes _right_. I haven't found it in any textbooks, blogs, or production servers. **How can I design an experiment to show the benefits of writing while under Snapshot isolation?** I am particularly interested in comparing write performance. However, what should I vary to test when writing under Snapshot is a good idea? Furthermore, what is a fair comparison to it? Read Committed, RCSI, or something more extreme like Serializable? I am **not** asking about Read Committed Snapshot or [using Snapshot for reads](https://dba.stackexchange.com/questions/346376/why-not-use-snapshot-isolation-for-everything-read-only)· ; They're both awesome.