#1 Best Workflow Automation Use-Case
Workflow Automation was the AI of last year. Everyone's now doing it, or it least say they're doing it.
Years ago, platforms like Zapier were there to connect multiple services together. If you can live with a delay, or have gobs of money, that's a good way to go. These days, most apps have some form of workflow built right in.
The Usual Suspects
What most people do with workflow automation are simple if-this-then-that scenarios. Some basic examples:
- If the status is change to "Send", send an email to the prospect
- If an email reply comes in, change the status to "Replied"
- If someone clicks on a button, create a new record in some other place
- If the status changes to "Sold", update the related sales order
- etc ...
Simple stuff.
Some go further, into building PDF's and reports, querying external systems, etc. Or doing large calculations and aggregations based on data and events.
But there's an elephant in the room.
Errors
The reality of asynchronous workflow automation is that you can't be guaranteed that all workflows are done the very second you're looking at a record or starting a process. With workflow processes that rely on external connections, there's a tiny chance that a call could fail and bad numbers start cascading through the system.
The bottom line is, what you're looking at on the screen, could be wrong 😱.
And sometimes the real-world consequences can be dire.
Which brings us back to the #1 most important use-case ...
Policing Data
If you're doing workflow automation, you absolutely must have flows that police your data. Just to double-check and validate that what you see is correct.
On a good day, this will seem like a waste: workflows duplicating calculations that have already been done.
But on a bad day - they'll be your saving grace.
For the same reason that programmers use tests to ensure their code is working, business owners need policing flows to ensure that their data is correct.
This is one of my favorite uses for InfoLobby's script fields. They run only when you look at a record and are never cached. So you can be guaranteed that their values are correct. Most of mine return nothing if all is OK, and a big fat red notice if something is not OK.