I jz started using PM months back, and understand the simple API calls, flows that dealing with my local (single instance). But i am thinking to move the current integration dev flow from intensive code base to PM. Imagine we have 200 clients who are subscribing our services (e.g. “Get Business Contact”)
- Whenever we have new functions (API), we dont need to copy the same API/code to each single client
- 5% of the customer might as for unique workflow (which doesn’t apply to standard)
- Track and limit the API call control (in case someone abuse the service)
- Different schedule for the same process (using Monitor, some might run daily 5AM, some run for every 12 hrs)
What’s the best way to design the approach?
- 1 client/workspace? it sounds easy for expansion for 1 client, but there ain’t code/API sharing btwn clients
- 1 workspace, 1 client/collection? similar to point 1, no code/API sharing
- 1 workspace, 1 collection, multiple environment variables (e.g. McD-DEV, McD-UAT, McD-PRD, KFC-DEV, KFC-UAT, KFC-PRD, …) API/code are shared, but in “Flow”, for a shared flow with a minor variants (e.g. for “McD”), i will have to have an IF to branch out the unique requirement, which making the flow hard to maintain when client base is grow.
Thanks for your reading. Comments are mostly welcome!