A process is successful if p = c and m and r are zero, where the letters correspond to the amount of process tokens produced, consumed, missing and remaining :-)
Now, lame process mining joke aside, a process is successful if it is effective (it meets the overarching goal for which it was designed) and efficient (it consumes the minimal amount of process resources - time, talent, money etc - necessary to carry out the process activities).
One may further judge cost/benefit by expanding the two primary notions above, i.e. you can optimize for capacity (i.e. revenue), consumption (i.e. expense) or both.
"We can afford to focus on smaller and smaller defects and eliminate them at their root. That reduces cost, because things just work.”- Jeff Bezos
When a process is successful, no-one notices that it is a process. They just can do what they want to do. So they do more of it.
With customers that means more sales, with employees it means more tasks done in less time.
But the true measure isn’t in the numbers. It is in the happiness of those customers or employees (or both). For once the systems aren’t getting in their way.
Unfortunately nobody measures happiness.
Internally it shows in lower staff turnover, fewer internal arguments, fewer demands for more pay... a host of things which reduce everybody's workload.
Externally it shows in lower customer churn, less resistance to upsell, higher recommendation rates... a host of things which reduce marketing and sales costs.
So take the process improvement money from your HR and Marketing departments. I wish!
As many of the comments so far have implied: adoption is the key indicator. Of course, I'm not referring to adoption by users who aren't given a choice about using your BPM-driven solution. In that case, you want to look at workarounds and complaints to determine success (or lack thereof). But if the project has worked, as Dr. Samarin suggested, you will begin to receive request upon request for additional apps. PS, that's a good moment to ask for that raise.
I'm a little amazed that in most comments happy users are seen as more important than happy process customers. Sounds like a very internal and software view to me.
Without happy customers you don't even need users to execute your processes.
I'm with Emiel on this one, a good process equates to delighted customers and users who don't even realize how much complexity is involved behind the scenes. I recently revisited a customer who said that the process had automated 70% of the business rules and they were about to make enhancements to take that up to 90%. The new users didn't even realize how much automation was taking place behind the scenes because the training had been simplified, complex key strokes eliminated, manuals not longer required, switching from one system to the other, all gone. Result, faster resolution of customer queries and very happy customers, which is the number one goal of a customer services department.
It's the constant "battle" between efficiency and effectiveness, as it is also possible to see throughout all the answers here in the forum.
I mostly agree with opinions positioning a successful process as one that keeps customers happy and consequently bringing cash and more business into the house. On the other hand it's also true that too much focus on delivering and you might lose your sight from performance, employee engagement, and continuity of service. Keeping a process efficient, streamlined and with it's participants engaged with the work they have to do is also really important.
However, and as harsh as this might sound, if you don't have happy customers (or no customers at all) there's no point in keeping your employees or collaborators happy with great automated processes and ultra-friendly UI. Customer is still king. As Drucker once said: "the purpose of a business is to create (and satisfy) a customer"
Agreeing with what Bogdan said, a truly successful process is both efficient (great performance and engaged participants) and effective (great quality of delivery). But if I had to choose, it's better to be effective without efficient operations, than the other way around.