What is our primary use case?
Most of our use cases are related to business, like reconciliation and reporting. Therein, they have some internal applications to automate SAP automation and Salesforce Automation. Our most recent use case is related to documents, like the invoices coming from customers. We have to extract that data from invoices via different formats, e.g., some are digital formats and some are scanned formats. So, we have to extract the data, which we are doing with the help of UiPath.
We are using both attended and unattended automation. For 90 percent of our use cases, we are using UiPath for unattended automation.
I use UiPath almost every day. When I finish developing one process, there is a new process to develop. If a process is complex, it almost takes six to eight weeks to develop it, then you have to deploy it for monitoring. After that, the next process comes up.
How has it helped my organization?
UiPath helps based on how it is configured. In our case, there are so many transactions coming in, it is not possible for a human to complete them in nine hours (working hours). So, we went with unattended bots, which we mostly run at night. We start the bots after the working hours of humans, so when the operations team returns in the morning, their work is complete.
From a developer perspective, the process is smooth and easy. You can find a solution on Google easily. You can develop your own code. From a technical perspective, it is 100 percent.
What is most valuable?
UiPath Studio is great. It has all the activities. You don't have to write anything. Even after that, if you feel that you have to do something for yourself, then you can write your on-premises code in it and develop your own framework. Everything is there. You just have to use UiPath Studio.
if you are a business user, even if you don't have a technical team, you can install the second version of the Community edition, which is StudioX. This is specifically made for business people who don't have a lot of ideas about technicalities. This is a great feature.
What needs improvement?
If websites are made in a recent programming format, it is very easy to automate them with the help of UiPath. However, if that technology is based on legacy applications, then it is very fragile and hard to do that. So, we have to choose the technology first, and if the technology is new, then you can easily automate it with other applications or the help of an RPA tool.
The process can be complex if an application is a legacy application or the data is unstructured. The analysis of some bots is lacking. If a customer asks for analysis of a particular bot, you cannot just get the data from the UiPath and give it to the customer. It is not that easy. I would rate this process as a seven or eight (out of 10).
The Document Understanding feature should be more developed and advanced. For example, you have to make a template with their ML model. Currently, we can't use our own ML model, and we have to use the UiPath ML model. UiPath has only a few ML models right now. They should come up with more ML models or make it easier for us to use our own ML model. While they are working on this, I have felt much difficulty in extracting data during our last process for Document Understanding. We had to go with the Python language. So, I think they are lacking in this feature.
What do I think about the stability of the solution?
After deploying the bots in production, processes are very stable, unless something happens with the machine. You don't have to monitor a process every time. So, I am very impressed and satisfied with things.
We are looking to update to version 20.
What do I think about the scalability of the solution?
This solution is very much scalable. If you are working in a small or large organization, it doesn't matter. It is very much scalable, up to anything.
We have a team of around 100 to 120 people in RPA automation, in which 60 or 70 users have the developer license.
Which solution did I use previously and why did I switch?
I am very much impressed and satisfied with the UiPath solution. Earlier, I used the Automation Anywhere solution, and it is sort of messy and complex. You have to pull everything from a single workflow. Then, I moved to UiPath, and everything was very sorted. If you really like coding, you can do that. It gives you a real developer type feel.
How was the initial setup?
Developing and deploying robots with UiPath is very straightforward. It hardly takes five minutes to deploy a process on Orchestrator.
What was our ROI?
The last process that UiPath covered saved the work of two people.
If you have repetitive tasks, you can apply UiPath and have your people trained to do other work.
What's my experience with pricing, setup cost, and licensing?
If you want to start doing RPA, I think you should definitely go with UiPath because it has the Community edition. You can just install it and check whether or not your process works fine with UiPath. It will be an attended bot, but you can form an idea whether your bot can easily be automated.
Which other solutions did I evaluate?
We have one use case related to Microsoft Word Automation. Word Automation is not compatible with other tools. However, in UiPath, we were able to write our own code to automate and format the Word document, which is why UiPath Studio is the most valuable feature. We are also using the Automation Anywhere. But, in Automation Anywhere, we cannot write our own code. So, we can't automate Word Automation from Automation Anywhere as well.
Automation Anywhere has come out with its own new version of 2019, which is as effective as UiPath. However, when I have tried to run or deploy the bots, it still lacks in features. For example, in UiPath, AML activities are coming up very frequently. Whereas, in Automation Anywhere, these features are lacking, which is why I go with UiPath. Also, Citrix automation is very good with UiPath. You feel like you are able to detect the elements and images.
UiPath gives you REFramework, which is absolutely amazing for business use cases. Automation Anywhere also lacks this feature. With Automation Anywhere, you need to make your own framework, and if you are making your own framework, then the look becomes messy. If someone is trying to understand it, then they have to spend more time on the framework to understand it.
I use the IQ Bot of Automation Anywhere, and even after training a hundred documents, it's not picking up or extracting the data from documents. I feel like I have to train the model again, which is not the case with Document Understanding. If you properly make the templates in Document Understanding of UiPath, it gives you full text values. So, it is more advanced and suitable for me.
What other advice do I have?
We use UiPath Assistant to run processes about 10 percent of the time. Most of the time, we are using Orchestrator.
UiPath releases new features every 15 days to a month. They have already come up with AI and machine learning.
If scanned documents are coming in for some of the work, we are also using Python language for this.
If you already have a technical team, then you can ask them to look into UiPath Academy. If they have basic knowledge of programming or coding, then even in seven days, they can easily learn UiPath and start applying it in their organization. You don't need to hire outside developers.
Overall, the solution is a nine (out of 10).
Which deployment model are you using for this solution?
On-premises
Disclosure: PeerSpot contacted the reviewer to collect the review and to validate authenticity. The reviewer was referred by the vendor, but the review is not subject to editing or approval by the vendor.