What is our primary use case?
I am an RPA developer and I work with UiPath in that capacity.
Our current use case involves the automation of a process involving healthcare-related data. This is confidential data that is received from the customer and inserted into Oracle forms. Reports are then generated from it and these reports are then used by the organization, which is in the healthcare domain, for their analysis.
The data being analyzed includes medical and treatment history. For example, with the current pandemic going on, there are all sorts of healthcare data that is related to it, including various types of treatments. When somebody walks into any clinic or hospital, all of the treatment is entered into a database and we get an extract of it. The analysis is used to get more details.
Another interesting use case, prior to this one, involved the documenting of invoices. We were working with approximately 250 different samples of purchase invoices, many in different formats. One might be a native PDF file, whereas another could be a scanned PDF, and yet another might be a simple handwritten invoice that was converted into a PDF based on a picture that was taken from a mobile device. We were receiving these invoices from our client and they wanted to extract data from them. It was accomplished by using the Document Understanding features in UiPath.
The other notable use case had to do with issuing refunds for purchases that were made on an e-commerce site. When a customer made an order and there was a problem that resulted in them wanting a refund, there were multiple ways that the client could request one. A refund application could be received by the customer care department in the form of a simple call, which was a verbal request, or as an email written by the customer, or as an automatically-generated email that was created based on filling out a form on the website.
Regardless of which of the three input methods is used, the refund request is gathered and sent to a mainframe application. At that point, the information is extracted from the mainframe and the refund is issued using another application.
The automation of these tasks using features such as artificial intelligence and document understanding has reduced our costs. For example, with the invoice processing use case, there was a team of between 20 and 25 agents who were doing it manually. Obviously, a team of that size has a large cost associated with it. Also, the volume was very high, which meant that the team was not able to deliver on all of the work. There are approximately 250 vendors sending invoices to our customer to process the data, which translates to about 1,000 documents being sent on a daily basis, to be processed by only 25 people. It was a huge task. With this level of volume, people tend to get frustrated.
We implemented the automation and the team size has now been reduced to only five or six people, and that is only required to monitor the bots. For example, they check to make sure that the data being fetched using document understanding is at par. We have set the minimum confidence of the documents being scanned at 90% and each day, a report is generated and sent to the customer. Overall, it was a very cost-saving implementation.
How has it helped my organization?
When it comes to building automations, it's pretty simple. UiPath has a lot of features, starting with Process Mining. For example, when a business person wants to start from scratch, they don't necessarily know which process to automate. They can use the Process Mining feature available in UiPath to get an answer as to whether a particular process is worth automating or not. Features like this are available to shorten the processes. Once a process is identified, we can go ahead and implement and develop the automation. Overall, it is very easy to develop automations because of the inbuilt features.
UiPath provides us with features to implement end-to-end automation, and it begins with the initial steps. They have Process Mining or task mining features to help companies that have an appetite for automation but don't know what they want to automate. Once a process analysis is completed, it can determine whether a process can be automated or not. At that point, we can proceed to the bot development.
Once the bot has been developed, there are different options, including third-party support, for deploying them. After a bot has been deployed, they have the hypercare facility to monitor and maintain bots that are in production. Taken together, all of these features facilitate end-to-end automation. This is very important to our work.
Consider that UiPath did not always have the Process Mining feature. In those days, in order to do process analysis, companies had to hire Subject Matter Experts (SMEs) for different domains at the beginning of a project to determine which processes could be automated. For example, if there were a requirement for a healthcare process then the company had to hire a healthcare SME. The SME would assist in process analysis and in the determination of whether it was something that could be automated. Nowadays, with the help of the Process Mining features, we no longer have to worry about who is doing the analysis.
From our standpoint, we have to do the recording and supply the inputs. With this, a report will be generated and it will tell us whether the process can be automated. If so, then the developers or an architect will come in and do the development. Following this, the DevOps team will come in for the deployment. Finally, the production support and monitoring are taken care of by the developers.
Using the cloud-based deployment helps reduce our on-premises footprint because we no longer have to take care of servers and local databases. There was a separate team taking care of the networking aspect, and another team taking care of the databases. For on-premises deployments, there are different teams that come into play, whereas with a cloud-based solution, we don't have to worry about anything. UiPath takes care of deployment and the databases, and everything is automatically handled. I can't estimate an exact percentage of how much our on-premises footprint has been reduced, but it has really made a significant impact in terms of substituting or replacing the on-premises facilities.
Attended automation has helped to scale our RPA benefits and in fact, in my almost five years of experience, I have not come across a scenario where the automation could be purely unattended. In the use case with the invoices where we used document understanding, there was still a minimally-sized team of five people in place after the implementation. They monitor the processes and if the criteria of 90% are not met then the invoice will be rejected and sent back to the vendor. In this case, attended automation is really helpful.
The Automation Cloud saves our IT department time because we don't have to worry about hiring different teams for different tasks. The infrastructure maintenance is handled by the vendor and all we have to do is pay for our licenses, then everything is ready.
Automation Cloud has helped to decrease our time to value. You don't have to worry about how the installation is to be done, you just have to pay the license fees and the team will do everything. They have packages, and you just have to give them the numbers according to what you need. Based on that, they will propose a solution and propose an infrastructure for you. Once that is agreed, they will do the installation within a very short time. After that is complete, your developers can start doing the development and within no time, you will start seeing the ROI.
The Automation Cloud has indeed decreased our overall costs.
It is really helpful for an organization to use the Automation Cloud when they do not have sufficient resources available themselves. In situations like this, the UiPath Automation Cloud plays a very important role. You just have to give them the number and they propose the solution including the infrastructure. You just need a few developers and everything else is taken care of by the Automation Cloud team.
We have a lot of clients who are looking at digital transformation, but they don't know what tool to use. There are a lot of vendors available in the market who provide products that are similar to UiPath, but UiPath has a lot of very strong marketing, a strong sales team, and at the same time, the product is very good. UiPath helps in doing quick installations, quicker deployments, and quicker turnarounds. Ultimately, it really helps with digital transformation for any organization.
Digital transformation requires an upgrade, but it is not very complex and not very costly. A company may have to be 25 to 30 million dollars to upgrade, which is a significant cost, but if it saves 100 million then in the end, the organization is saving money.
UiPath has helped to reduce human error to a great extent. When humans perform the same task over and over again, we tend to get bored and frustrated. Our excitement goes down day by day. However, when it comes to machines or bots, they don't get bored and they don't get frustrated having to do repetitive work. This leads to fewer errors, which has a positive impact on the business. There are fewer escalations and the customer is happier.
Our automations have helped to free up employees' time by reducing how long it takes them to complete their tasks. It leaves them time for learning or performing other jobs, and in general, it helps to reduce work hours. I estimate that percentage-wise, it saves my team between 30% and 35%. The total time saved is approximately 250 hours per month.
One of the ways that our employees benefited from having time freed through the use of automation is upgrading. A lot of people, in particular our technical staff, upgraded by learning new technologies. This ultimately led to helping the organization because they are able to work on different jobs.
Automation also leads to an improvement in employee happiness. If you ask them to do the same thing over and over again, on a daily basis, they get bored. When 30% of the time is saved by using automation, it frees up perhaps two hours a day for them to do different work or even recreational activities.
What is most valuable?
UiPath contains a lot of very useful features.
The artificial intelligence and document understanding features are very good and have been helpful when it comes to processing handwritten or unstructured, especially when there is a high volume and it is repetitive.
Using the artificial intelligence and machine learning features had a drastic impact on our use case. In the scenario where we used document understanding, we were able to successfully automate the use case without artificial intelligence. However, there was not a sufficient improvement. The success rate was between 78% and 80%. We did not use these features originally because they come at an extra cost and we did not have approval. When we were given approval and acquired the licenses, we re-implemented the automation and included the inbuilt AI and ML models. Our success rate went up to between 92% and 93%, which is significant.
What needs improvement?
The Orchestrator is in drastic need of improvement. Three or four years ago, the Orchestrator UI was very simple and comfortable to use but now, they have added so many features that it is difficult for new users to quickly understand it. The latest implementation that we did was so complicated that nobody was able to understand it, aside from the developers. Generally speaking, UiPath needs to improve the interface, and the overall user assistance should be boosted.
The developers are the core users of the product and it's important to make the product easy for them to use. The easier it is, the more people start using the tool. Then, as there are more developers available, more organizations start to hear about RPA tools and Uipath.
For how long have I used the solution?
I have been working with UiPath for approximately four and a half years, across three different companies.
What do I think about the stability of the solution?
I have found that one or two times out of ten, UiPath Studio hangs and needs to be restarted. These types of technical glitches do happen but not often.
What do I think about the scalability of the solution?
In the organizations that I have worked in so far, the UiPath features and the licensing available meant that we didn't have to worry about the infrastructure setup or scaling. You can start with a small number of licenses and robots, with little infrastructure requirements, and gradually, as you need more robots and more powerful infrastructure, you can scale up. It has great scalability.
Even if somebody is using the community edition, we are able to build a PoC for them. In fact, using this approach gives us a pretty good idea of how the automation will proceed. Based on this, we get an idea of what the infrastructure will look like. Then, if everything fits into the customer's budget or request, we will go with the enterprise edition and implement the solution. Throughout the process, scalability is not an issue.
In my department, we have between 70 and 80 people working with UiPath. Out of these, 80% are developers, perhaps between 10% and 15% are business analysts, and the remaining people are solution architects or tech leads.
We use UiPath extensively because it helps with quick implementation and a short turnaround time. Deployments are very quick and easy to do. Given the current pandemic situation, I don't see the demand for UiPath or any other RPA tool going down.
How are customer service and support?
I have been in contact with technical support many times and they are very good.
The issues that we typically face are related to Studio and sometimes licensing. The support has a fast turnaround time and if the organization is large, with a lot of licenses, then they have dedicated support for them. It is not necessary to deal with customer care, for example. Instead, dedicated technical support is available and they are really good.
Which solution did I use previously and why did I switch?
The companies I have been with did not use another RPA tool prior to UiPath.
How was the initial setup?
As a developer, I have not been directly involved in a UiPath installation or implementation. It's generally the responsibility of the implementation team that every organization has, and developers don't have the authority to simply download and install or implement UiPath. That said, I have given my input in the past.
Every company has a QA team and once they identify and finalize which automation software they want to install, the implementation team takes care of it. In my experience, the developers are given an instance of it to use after this.
It used to be a very simple installation, although, as I have seen UiPath change over time, it has become very complex. They have now Studio versus Studio X. They also have the Automation Cloud, which is different from an on-premises deployment. Somebody who is new to UiPath will not be able to do anything.
In terms of how long it takes for deployment, it depends on what process you want to automate. A very simple process can be automated, end-to-end from process analysis to deployment, in about a week. Complex automations have been taking us between six and eight months to complete. It all depends on what kind of automation you are doing.
In general, the implementation strategy is simple. Again, it depends on what kind of process we are automating. If the process is very complex and it involves a very high volume of data to be processed and stored, then it will require greater network bandwidth and more database storage. It may also need a fast server, with regards to processing speed.
What about the implementation team?
All of the implementations that I have seen have involved an in-house team. There are different departments involved, such as the networking team, the database team, or the IT department. Also, for the latest implementation that we did, involving Automation Cloud, the UiPath team helped us.
What was our ROI?
UiPath has reduced our overall costs. For example, if you are replacing 25 to 30 humans, where instead you only have to pay for licenses, the expenses are much less.
The company and our clients see a return on investment through savings in time and costs.
What's my experience with pricing, setup cost, and licensing?
There are different licenses available, including variations for an on-premises versus a cloud-based deployment. For example, an on-premises installation is very easy and not very costly, and is suitable for a client with a simple optimization that does not have cloud-based requirements. Generally, companies do not have to worry about infrastructure because of the different options that are available.
Different licenses are used to activate features that the organization wants to use. For example, they offer licenses for UiPath Studio X, attended automation, unattended automation, document understanding, and so forth. They also have customized licensing and packages available. For example, if a customer requires 50 attended and 150 unattended licenses, but it does not fit into an existing package, then UiPath will create one for them.
They have an on-demand, floating license available for companies to save on costs. In this case, if a company wants 40 licenses and is only using 10 at a time then they only have to pay for the ones that they are using, with the remaining 40 triggered only when requested.
Which other solutions did I evaluate?
There are other automation tools on the market, such as Automation Anywhere and Blue Prism. Of these, UiPath is very strong and this is because of the UiPath academy for training, the UiPath certifications, and other things that impact the RPA developers.
The company that I am with did not evaluate other products before choosing UiPath.
What other advice do I have?
Over the time that I have used it, the number of resources available in UiPath has grown to a very high level. For example, with the advent of Automation Cloud, the organization no longer has to worry about protecting against on-premises failures.
I have used the UiPath Apps feature but have not yet put it into production. I have completed a PoC and done some self-learning, but I haven't yet come across a scenario or process where we have to implement it.
We completed a proof of concept where our client was looking for customized input forms for their client. In that case, if I had to write the code for this customer's input form, I would need to hire additional resources at an extra cost. It might be, for example, code that has to be written in Java. In UiPath Apps, we were able to create a customized form for our client within a span of two to three days. It would be able to handle their various survey-related inputs. This feature is very helpful because it is much quicker than writing customized code for the client.
The Apps feature did not affect my IT department's workload but it did reduce the work required by my team. This was important because our client was not originally seeking to use UiPath for this part of their project. Instead, they had confidence in us and asked how it could be done, and whether Uipath could accomplish it. When we completed the input form using Apps, it definitely helped in reducing my organization's workload and in turn, reduced the overall cost to the client.
In summary, the features are very good and there are plenty of tricks and tips available, depending on what it is that you're trying to do.
I would rate this solution a nine out of ten.
Which deployment model are you using for this solution?
Private Cloud
If public cloud, private cloud, or hybrid cloud, which cloud provider do you use?
Microsoft Azure
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.