Automation Anywhere (AA) Review

Drag-and-drop and recording of mouse and keyboard controls make it easy to create reusable snippets of code

What is our primary use case?

Our primary use case for it was financial back-end operations. We dabble in Automation Anywhere. We don't use it.

How has it helped my organization?

We have an old ERP which required human interactions because of the lack of a system. For example, we were copying and pasting from an invoice to a spreadsheet, and then from the spreadsheet, we would run a macro that would plug it into our ERP. What we did is have the robotics read an invoice and plug it directly into our ERP, avoiding the extra steps, let alone a human doing it. We placed a robot and technology in place of a human. It saved about an hour-and-a-half a day. We measure in FTEs and we measured the savings as 20 percent of an FTE.

What is most valuable?

In Automation Anywhere we liked 

  • the drag-and-drop 
  • easily stitching the recording of mouse control and keyboard controls. 

They were not unique but they were user-friendly. 

For developers to pick up a new tool, you need something user-friendly where it's easy to create reusable snippets of code and use them in another process down the road. I would rate the ease of use at eight out of ten.

For how long have I used the solution?

Less than one year.

What do I think about the stability of the solution?

It was very stable. We didn't have any issues with it and, if there were, they wouldn't have been Automation Anywhere's problem. They most likely would have been our infrastructure.

What do I think about the scalability of the solution?

We didn't have to scale too far, so we didn't experience its scalability.

How are customer service and technical support?

We didn't use Automation Anywhere's technical support, even for deployment. We read and learned what we needed to know. We got on their support site, got their documentation, took the requirements, how to deploy, and what the right architecture was. We scaled based on what we thought we would use it for. My guys like to learn, they like to try, and I allow them to dig a hole and fall into it and then fix it later. They just didn't fall into a hole.

The documentation was good enough for us to read. Granted we're technical people, but it was good enough to read and take actions based on the content. It was really good.

How was the initial setup?

Setting up Automation Anywhere was fairly straightforward. I personally did it. I had my team behind me. They set up the IIS side and the SQL side. But for installation of the product, I did it myself because I wanted to know. I'd still like to learn, even though I don't do the work.

To install here, on-prem, at the server level, only took two people a couple of hours; perhaps a total of eight man-hours.

Our implementation strategy was, "Let's try it." We had an objective to save FTEs as well as to introduce technology to get around our lack of a decent ERP. It was a matter of picking among the big three, and some third-world country type of RPA as well, and seeing which one hit the button. We figured out what the requirements were, and we have a pretty hefty on-prem hosting, so we spun up some servers and installed it.

After deploying, our developers then took control. We had a team of three developers. In terms of maintenance, we usually set up our environment where we install updates monthly. That should take a couple of hours per server. I don't recall Automation Anywhere standing out as a "problem child." So maintenance on the infrastructure side might be about two hours a month.

What was our ROI?

We couldn't put the math together. When we decided to actually procure and get the quotes, they gave us free trials for a while and extended them for months. But when it came down to it, we couldn't do ROI because our company outsources to India. Our employees are in India, and in India, $250,000 goes a very long way. We just couldn't make up the money fast enough.

What's my experience with pricing, setup cost, and licensing?

Cost is the biggest area in need of improvement for Automation Anywhere. Annually it's $250,000. That's what deterred us right away. We stopped using it as a primary solution because of the costs.  We did not apply hardware to that, because we've already got the cost baked into our infrastructure. Otherwise, there would have been hardware costs on top of that. If we really took a full, all-in cost for Automation Anywhere, it would have been much higher. But we don't do it that way.

There are three big, heavy-hitters in RPA, with Automation Anywhere probably being the premier, followed by Blue Prism, and then UiPath. Of those three, Blue Prism and Automation Anywhere are very expensive, but the accomplishments are the same. UiPath is pretty affordable as a buy-in, with the accomplishments being the same.

Overall, each has its own uniqueness, strengths, and weaknesses, but when it comes to looking at it on the financial side, Automation Anywhere is probably one of the most expensive to have an all on-prem solution. We're all about on-prem. It was very expensive to stand that up.

We went with UiPath.

Which other solutions did I evaluate?

In terms of differences between Automation Anywhere, Blue Prism, and UiPath from a functional standpoint, there really weren't any. They all do the same thing. It's coding in .NET, coding in Java. They all have their strengths. Automation Anywhere stood out because of its high cost. 

When we put them side-by-side, everything we could accomplish in UiPath, we could accomplish in Automation Anywhere. Each one is making its own jumps. For example, when we were evaluating them, UiPath was making leaps in OCR and reading natural language, and Automation Anywhere was taking a back seat in that. But Automation Anywhere was advancing in its process improvements. Now, they're doing it the other way around, and Automation Anywhere is jumping forward ahead of UiPath. So they play this game, but either one is equally good.

What other advice do I have?

My advice would be "due diligence." Make sure you read, and make sure you engage the Automation Anywhere team and the support. We didn't, but we didn't have to. But do due diligence based on cost, and scale, and really what you're going to do.

RPA is the hot word right now. Everybody wants to do RPA. But what we did is just put it into our arsenal of other tools. It's not the golden bullet. It's not the one that is the end-all. It's just one of the tools in the arsenal that IT has. That's why we chose not to spend $250,000 and, instead, to spend much less. Sometimes RPA is the answer. At other times it's system integration, and at other times it's just raw development. That's what I even tell our customers. That's our toolkit. Our arsenal is developer's RPA, and we use a third-party integration tool as well.

Just the developers were using it in our organization. They tout it as it's user-friendly: Give it to a user and they can do it. But we didn't discover that at all. We couldn't just hand it to a user, so our developers were taking the processes and applying them with development code behind them. Automation Anywhere has "record-the-screen," but when our users were doing it, they would move an icon and it wasn't smart enough to find the icon that had moved.

In terms of extent of use, it was used daily for some of our daily processes. The finance process I talked about earlier is one example. We automated that and that robot ran on a daily basis. As for increasing our usage of RPA in general, we scaled up pretty quickly. Internally, we have four or five robots running 24 hours a day, seven days a week.

Disclosure: I am a real user, and this review is based on my own experience and opinions.
Add a Comment
Sign Up with Email