Previously I wrote Will Robotics (RPA) Replace ACL?
The short answer is no, and I describe the reasons in that post.
But that doesn’t mean someone won’t try.
Shortly after I wrote my original robotics post, I encountered robotics vs. ACL, part 2.
A long time ago, in a file system far, far away…
A couple of years ago, I had a manual process that I performed every day, and got tired of it.
So I automated most of it with some ACL scripts.
Every morning I’d get an email telling me whether everything was okay, or whether I had a problem.
If a problem occurred, the email listed the files that were causing the problem. When I got the email, I worked with the auditors who caused the problem to fix it (what the problem is and how to fix it isn’t important to the point of this post).
Years ago I thought about automating it totally, but determined it wasn’t worth the effort.
Also, automating it had some side effects–it would train other auditors to become a little careless, and I didn’t want that.
Finally, the problem didn’t occur that often. So I automated the big piece and left the little, occasional piece that the auditors had to fix as a manual task. I called the new, mostly-automated process Project One (it was my first scripted and scheduled ACL project).
The thing hummed along just fine for a couple years.
Over time, the auditors learned not be careless because they knew I’d be talking to them if they screwed up.
What they didn’t realize is that ACL tattled on them (shhh!), and it took no effort on my part.
Let’s Do Robotics!
Then our management decided to do a robotics process. So they started to look for something to automate.
They remembered that my Project One was NOT totally automated, and asked me if that wouldn’t be a good project to use as a pilot.
When I asked them who would write the code for the robot, I was told management would hire a vendor.
When I asked them who would work with the vendor team to help them understand the problem, the solution, and solve all the technical issues they would encounter (as well as work with IT to get security access, change configurations, approvals, etc.), I was told “You, of course. You’re our technical expert.”
When I asked them why they didn’t select a higher value process that contributed actual value to the business (vs. value to me and our department, which is what my Project One did), they said that they only had budget for a 10-week project.
I suggested several other worthy processes to automate. They still liked Project One.
Some Sharper Questions
So I did one of the things I do best: ask tough questions.
“So let me make sure I understand what you’re suggesting,” I said. “You want to do the following:”
- Pay $50,000 to a vendor to totally automate something that brings the company, as a whole, no business value?
- Automate a solution for a problem that occurs, at most, once a month? That takes less than 10 minutes a month to fix? That reminds auditors to be careful with their data?
- Take 10 weeks of my time to assist a vendor doing meaningless work–time which I then cannot spend assisting auditors with analytics or automating other audit processes?
- Use robotics software we don’t know and can’t support to automate a process that I could automate in ACL, test, and validate in six hours? (I’ve learned a bit over the years)
The reply I received: “Those are all good points, but we need a simple project that we can get done in 10 weeks. We need a good pilot project, and this is it.
My Quick Solution
When logic didn’t work, I had only one other choice. And I don’t mind taking risks.
I went back to my desk and automated the entire process. Testing, validation, and updating the documentation took a little over 6 hours.
The next day I told my management that Project One was not longer a good candidate because it was automated.
They agreed it didn’t make sense to automate something that was already automated.
So now they are looking for another process to automate.
You can’t make this kind of stuff up.
I kid you not.
If you shook your head constantly while reading this post, you might also relate to Audit Management Sometimes Sucks