Can anyone accurately explain rules execution limits?

[Douglas Bain]

If you interchange information between a Jira Service manager project and a work manager project you need to use rules on both sides.  The UI for creating tickets for end users in work manager is too basic so Service Manager must be used. We have Premium Service Manager users and either free or standard work manager users.  We cannot tell if our software setup will fail due to reaching a limit.    Seems idiotic. 

[Prince Nyeche]

I believe the explanation on this doc is pretty straight forward but just to summarize. On a standard subscription you have 500 monthly execution in all rules for any multiple project configuration and on Premium subscription you have 1000 execution per user (per application) execution on the same multiple project. If you do the math, depending on the number of users, you could have a higher rate in execution as a combined total of every premium user.

Now, if on either plan you can have unlimited execution based on a single project. i.e as long as the rule is working on only 1 project and not cross-project, the rule execution will be unlimited for that single project. The moment your rule works on another project, your monthly execution limit is deducted. e.g JSM project executing actions on a JWM project in essence that's cross-project execution. This also applies if you're using a JSM project A executing action on a JSM project A and B. In essence, the rule is working on more than one project.

To avoid such conundrums, it's better to know your limitation based on the number of available execution and use cross project execution only when necessary.