Automation Anywhere (AA) Review

Automation has significantly reduced our FTE count. There is not much guidance for implementing DevOps, like continuous development.


What is our primary use case?

We have automated some fund management and fund reconciliation processes, which has been done successfully. The tool has been performing well. 

How has it helped my organization?

In our fund management and fund reconciliation process, we had a lot of headcount. Automation has significantly reduced our FTE count, investing that IT personnel into other work. Also, the quality of the work has slightly improved.

What is most valuable?

Our development time has been reduced because the solution is very user-friendly. You don't have to know a lot of things. It has an IDE with drag and drops, in addition to workflow development. This gives developers an edge to write code quickly.

I am a developer with nine years of experience in full-stack development. I have found this tool strikingly different, compared to the other tools, in terms of ease to use. Anybody can be taught it, and it has a very small learning curve.

What needs improvement?

Automation Anywhere has some limitation with its integrations. You can build your logic with MetaBots and .NET components, e.g., if something is not gathered by Automation Anywhere or built into functionality, you can write a MetaBot. A MetaBot is like an external component. You can write it through C# or .NET code, and make a DLL, and import it. This is not open to all data types or EXE functionalities which can be written through C#. You have to define your variable types, matching them with Automation Anywhere existing variable types. Because of the knowledge required, I cannot let another person, who doesn't know Automation Anywhere, build my external DLL code.

They are already addressing some concerns that I have been having, like having cascaded code blocks, which is very helpful for developers to use when finding and debunking codes. Until now, we haven't had collapsible cascaded code sections in Automation Anywhere. This is not typical, as any IDE will usually have it, on any language.

The current interface for business users is not intuitive, as it is a code-based view. A business user wanting to develop a process may find the interface a bit elusive. Automation Anywhere is now working on a process flow view with a drop and drop, which should be better for business users.

There is not much guidance for implementing DevOps, like continuous development. Nowadays, the technology has come for the deployment part, it should be part of Continuous Integration and Continuous Development (CI/CD). 

With Jenkins or TeamCity, there is a job, and I want to see them have integration with Automation Anywhere, so we can directly code check-in, run, test, and deploy. Now, we are doing this manually for deployment, but I want to see a DevOps job instead.

For how long have I used the solution?

The product has been in production for six months.

What do I think about the stability of the solution?

We had couple of Windows' tools that we automated. The stability is remarkably good. 

In some cases with RPA web development, where we had to integrate with some websites, there were some problems in terms of reliability or stability. The issue is not constant. Different browsers, like IE, work well, where others, like Chrome, don't work well. With legacy websites, based on old technology, it also does not work well. 

What do I think about the scalability of the solution?

The scalability is fine. You can create small, repeatable tasks. You can expand through different bots. So, you don't have to recreate the same stuff. Automation Anywhere provides the facility so you can reuse components to make a scalable solution.

We are improving scalability on a daily basis, as RPA is new to us. Our code is now more scalable and reusable than we developed our basic structure. Going forward, if we have to create a new bot with any given scenario, it will take 20 to 40 percent less time to create. This is because we have already made the scalable, reusable components.

How are customer service and technical support?

In terms of accessibility, the Automation Anywhere technical support is the best. Out of all the RPA tools that I have used, their support forum is more agile and responsive. For any issue, we just have to create a ticket and a technical person will get assigned in no time. They are the best in terms of responsiveness, but I have mixed feelings about the quality of those responses. I feel sometimes to support their fast responses that they probably compromise their quality.

If you previously used a different solution, which one did you use and why did you switch?

We used to get a lot of emails from the original fund management users (the consumers) asking about their estimated stock prices: monthly and biweekly. This information comes through a lot of emails. We had to read those email, classify and categorize them, then go into a different portal to get some more data, and finally, send them a result in an Excel format.

What about the implementation team?

We used a consultant, and the deployment was very smooth. We have been maintaining three environments: development, UAT, and production. The deployment procedure is pretty simple. We just have to export from one to another environment where the dependency is just a package file. 

What was our ROI?

We are thinking of implementing Bot Insight and the CoE dashboard to calculate our ROI, since our bots are not analytics-based. We are just using our bots as a bot. They are not integrated with any analytics. I know that they have a CoE dashboard, where you can define parameters to automatically calculate ROI. At the moment, we are manually calculating our effort savings.

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

I don't agree with the IQ Bot licensing costs, which depends on the consumption of the document rather than per bot or user. This is confusing for me, because most of the cases for us in a possible future scenario, there would be a lot of invalid data to be scanned to get to real data. E.g., we would have to supply approximately 100 pages of scanned data to extract two pages of original data. In these type of cases, the IQ Bot might be a costly venture.

Which other solutions did I evaluate?

The company also looked at UiPath. From my understanding, Automation Anywhere was chosen for ease of use and licensing.

What other advice do I have?

The IQ Bot seems interesting in the terms of populating semi-structured data and gathering intelligence of its own. Though, I haven't implemented in my project yet.

I didn't know Automation Anywhere seven to eight months back. All my learning came from the Automation Anywhere University. Though, I have experience in other RPA tools, the courses are really helpful, as they helped me to gather the knowledge needed to implement and master the product. Right now, I'm applying the information to my domain and company.

I haven't used it in a Citrix environment.

I very excited to see the next version of this product.

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