What is our primary use case?
We typically solve for any use cases that falls under different business functions within our company. That includes finance, supply chains, services, IT by itself, and a little bit of engineering.
How has it helped my organization?
UiPath has improved the way our organization functions. The flexibility with which the business processes get changed is great.
A lot of times we know the method of operation, however, certainly it'll not be the same after a few months, a few years, or longer. Our dependent script or whatever is in place (that is dependent on that business process) has to be adjusted. The flexibility with this tool has enabled us to adjust those workflows quickly and deploy them so that our business can continue using those applications or the workflows that we’ve been using before, even after changes to the underlying system.
What is most valuable?
The most valuable aspect has been the workflows. They have helped to deal with legacy applications. We have a lot of legacy applications in place, which we cannot get rid of. The processes around these legacy applications are something that cannot be automated in a typical way. The RPA is helping us to automate the business processes that have to work with legacy applications.
The ease of building automation using the solution is great as it is a low-code solution.
I'm able to create workflows. By the time I'm familiar with one process, I'll be able to automate the next one. This is the case especially with tools like task capture; I'm just working through the process. In the end, it will be a skeleton workflow and it can be used for deployment once we are done with the cleaning. It has reduced the development life cycle by about 30%. It’s done this by making use of the features that are enabled by task capture and certain other features within the Studio.
Overall, we’ve seen the solution has saved costs. That is our value realization.
Our first target metric is to bring up the number of processes that we can save. We have a formula to convert inter-dollar values in terms of the user experience we are benefitting from. That’s the user experience that is enabled by automation. Those are direct savings which can be calculated by multiplying the number of dollars that we have spent for one resource per hour. Indirect monetary benefits can also be calculated by looking into the user experience factors and adding them when we do the value realization. In the last four years, we would've saved $3 million.
The human error rate has been reduced. Initially, when we targeted some of the business use cases, they were straightforward. They were linear in nature and there the accuracy had the upper hand 99.9% of the time. The reason is that the process by itself is quite linear. It doesn't have multiple branches or exceptional routes that it has to take to complete a particular transaction. We have good accuracy, however, we have had challenges with the accuracy when the business processes get complex. If there is any human intervention or if the quality of the data is not proper, or if the user errors are low, that is where the accuracy rate used to be low. It's better now.
Due to the fact that all these are role-based bots, if there is something that is getting changed, the bot will fail. Down the line, I can see that, for linear processes, accuracy will be great. However, when it comes to some of the complex processes, that is where we have challenges that we are facing with accuracy and we are continuously fine-tuning the process in such a way so that the accuracy can get better. It's great we can continuously tweak.
The solution does free up employee time and allows for the employees to focus on higher-value work. We have a lot of examples within our organizations where they have to deal with some kind of manually intensive task, such as just reading something from the document and putting that into the financial system.
We normally take up the customization portion that comes directly from customers. Those kinds of customizations have to be updated back into the financial system in order to make sure that they are appropriate. These updates take a while as they have to do with talking to the customer, understanding what changes are needed for a given order, or based on specific correspondence from the customers. With automation, employees can focus on talking to the customer to understand what changes they have to incorporate. And they can offload all the data entry tasks to a robot. This way, they can focus on how they can engage more with users to understand the pain points faced by the customer rather than spending time taking all those inputs and then doing the data entry job. They can be more client-facing.
I’m not sure exactly how much time is saved with automation. I could say that we have around 150 purchase processes that we have automated. We don't trace back how it has replaced a team or member of a team. We always go with the number of hours saved. We go the route of checking and saying “okay, so we have done this, but it needs a constant involvement from them in order to make sure that someone is owning the process.” We still own only the work.
We have started to use the solution's AI functionality in our automation. We started it recently and we have finished the proof of concept on document understanding, which involves AI, of course.
In terms of AI automation, we will be leveraging this tool for all business functions. There is no limit with any of the business folk that we talk to. Whatever the process is, as long as we feel that it is feasible to automate, and there is a value in automating it, or as long as we feel that we are automating the right processes, we will just take that up into our pipeline.
AI does help us handle complex and involved processes. We include a lot of use cases where the sole core RPA capability would not suffice as a purely role-based automation. We often encounter a lot of use cases where they say, "Hey, this is something where there is no logic in doing it." If there’s analysis or natural language processing, et cetera, we are making use of AI. However, the process isn’t in use yet. We’re just starting.
We have used UiPath’s Academy courses and we are also encouraging our implementation partners to refer to those materials so that they can be approved.
It’s kept us up to speed with the solution. We refer to the Academy daily. Of course, we get help from UiPath whenever we face any hiccups; we normally ask them questions and they're able to sort it out for us. That said, the materials are great for trying to sort out issues or problems on our own.
What needs improvement?
It's been four years of practice and we've matured with the traditional RPA candidates. We have a strong foundation with what we have showcased to our business folks, and we are good with the healthy background that we are building. However, when it comes to the roadmap of what's next, that is where we are not clear. While we get the concepts, bringing them to reality is looking to be quite a challenge. We are unsure as to if UiPath can actually bring our vision to life.
UiPath is very clear in defining items such as this is what the high automation needs, this is what the process planning needs. We are getting clarity into those concepts and we are able to explain that and take that back into leadership to get other approvals. They are able to understand what UiPath is talking about within these different concepts. Really, it's just figuring out whether we have the right arrangement at this point and if UiPath can get us there.
For how long have I used the solution?
I've been using the solution for four years.
What do I think about the stability of the solution?
I'm impressed with the version that we have today. 2019.10 is a version that is quite stable, compared to how we did with 2018. A lot of pieces that are enabled as part of the new version, 2019, are stabilized. We have zero downtime with the tool.
What do I think about the scalability of the solution?
Today, ith X number of bots, if we want to reuse the same solution, and if there is an appetite for consuming those kinds of robotic solutions, of course, buying more bots is going to solve the problem.
If you think in terms of scaling this platform by itself or the other business functions, that is where the discovery piece will come into play. We have to constantly talk to businesses to understand where the opportunities are to scale in the correct manner.
Scalability is possible in terms of reusing existing automations. It's related to the number of bots that we are going to purchase. When it comes to the number of business processes that we are automating, during the discovery process, twe have to engage with our customers and constantly follow up with them. When we understand more about how they're doing business, we're able to locate the kinds of tools that are going to help them.
Currently, we have eight bots in production and 150 processes are automated. I’m not sure how many users are actually on UiPath currently.
We always follow up with our business to build our pipeline. That goes hand in hand with the implementation. We off-load all automation ideas and requirements to the pipeline, to our implementation partners, so that they will be able to implement our vision.
How are customer service and support?
Traditional support for the RPA is great. In terms of the help that we are getting, if we end up with some issues, running operational issues, it could be better if they can propose some fixes. It's not that automation is going to solve every other problem that the underlying system is having. However, we expect some kind of expertise from the tech support when we face issues that are related to the system. We need to understand if there's an ERP error, if it has to do with the underlying system, or if automation has to solve the issue. Often, technical support will say "Okay, so this is your error, go and solve it." Yet, due to the fact that support has seen more issues like this, they should have more insights and they need to be able to share those inputs in a way that is going to help us.
Which solution did I use previously and why did I switch?
While I didn't use a different solution previously, the finance team has used Blue Prism before. They implemented Blue Prism and they engaged Blue Prism to automate the processes that they have added for automation. Now, we have aligned on a single platform. It is UiPath now, however, they initially had around 50 processes that they automated using Blue Prism.
We proposed UiPath as the one solution based on Gartner ratings.
How was the initial setup?
The initial setup was quite straightforward as we understand how RPA works and we understand how UiPath is going to help, how UiPath is a tool to help us to automate things. It's quite straightforward in terms of that. Whenever we are doing some kind of initiative, like document understanding or data capture, it is quite straightforward.
However, with process planning, we didn't understand the documentation right away. That is where we used to get help from UiPath.
What's my experience with pricing, setup cost, and licensing?
The pricing was great at the start, and so, down the line, we have been enhancing all these features. We are seeing that, as we are looking for opportunities to grow, the number of robots that we need to purchase and the software cost is going to go up.
UiPath has increased the cost. We feel that it's good, however, based on all the new features, which we are pursuing. That said, we expect that whatever robots that we have purchased or whatever the standard platform that we have from UiPath should continue with the pricing that they had earlier.
There will be an offset, however, when it comes to the existing platform like Orchestrator or robots, and we are expecting that the margin should be less.
Which other solutions did I evaluate?
Back in 2017, we evaluated three to four products. Blue Prism was already used by the finance team, however, we evaluated WorkFusion, UiPath, and Automation Anywhere.
Of those other three, we evaluated WorkFusion and UiPath extensively.
The two reasons that we went with UiPath were, one, the learning curve, and, two, the community edition of UiPath, which had everything we needed to dig into the solution. Whereas with the other companies, there wasn't that option. With Blue Prism, for example, we had to buy a license in order to check whether the tool was going to work for us. In 2017, we were not sure whether this was going to work or not. At that stage, UiPath was the only company that gave us the entire set of tools to try and it worked really well.
What other advice do I have?
We are customers and end-users.
While we're using the on-premises deployment, we are open to moving to the cloud. There are a lot of cloud solutions that we already use in our organization. However, with UiPath, we have stayed on-prem out of concern for security. We don't have clarity on if a cloud solution is going to work securely.
The other concern is around how we are augmenting the capabilities of core RPA. We know that process mining is going to help us, however, whether process mining is already added into the RPA, do we have any solid use cases that we can start with.
I'd rate the solution at an eight out of ten.
Which deployment model are you using for this solution?
On-premises
Disclosure: I am a real user, and this review is based on my own experience and opinions.