That is one of the drawbacks of Cassandra. I feel compelled to make the obligatory declaration that this is a contrived example, but even in the real world you would have examples like this that require updating multiple tables. It’s a symptom of using a Db system that is so cheap on the writes and so expensive on the reads. But what I’ve found is that storing data like users and emails, which could potentially be updated often, is not the ideal use case for Cassandra. It’s strength is really in immutable data that is written often and read less often.