What is our primary use case?
Originally, we were using UiPath to draft documents and send emails on mass to where we had large communication exercises. We used a robot instead of a small army of paralegals to generate the documentation and draft up the emails where we had to communicate with 2,000 to 3,000 people. It was a little bit more involved than just doing a standard mail merge, but we were able to use UiPath to create a number of documents and email it to an individual customer all through a central email address.
Fast forward to where we are now, we have a few of these things focusing in on what we call post-completion activity, like the things you do after you sign a contract. So, it may be you're uploading it to the client's contract management system, applying stamps, or registering the contracts in an official register. The robot is able to do that for us post-completion. Those are our primary use cases at the moment. We're looking at more data integrity type stuff, like comparing our internal data sources against public record.
How has it helped my organization?
The ability to displace some activity that was traditionally on our paralegal team has improved my organization. We're an outsourced managed legal services provider. We're primarily a people-based business and UiPath displaces some activity off those paralegals and brings in automation. For me, it becomes an additional type of resource. The long and short of it is that we are able to move work that was traditionally done by people and would be charged at a rate, off to automation where we can bring the costs down. It enables us to reduce our running costs of our clients. A single bot running post-completion has saved as two FTEs.
UiPath has helped to speed up case resolutions in a couple of ways. It's focused on doing particular jobs, so it achieves the job faster. People don't need to complete an entire task end to end. They can stop at a certain point and the robot takes over. That allows a person to get through more work. Also, the fact that the robots are able to do this stuff overnight 24/7, means that we have more capacity to do stuff.
It gives us the ability to respond to clients. It gives us an option in how we're going to automate work for clients. It's hard to say if it has reduced the cost of our digital transformation because we deal with all the people. I suspect it hasn't done it internally, but suspect that has made some things cheaper for our clients. It enables us to deliver digital services cheaper for our clients.
UiPath had an effect on our legal staff. It takes work that people don't want to do at the moment, having to download the document, take some details off the top and the bottom of the document, apply a stamp, and then re-upload, it's not what our paralegals and new trainees want to be doing on a day-to-day basis. So we are able to shift activity nobody enjoys onto a robot and let them focus on the stuff they've been trained to do.
In terms of how much it has reduced the processing times, the task itself still takes as long but we've got a robot doing it instead of a human. I don't think that the impact isn't that dramatic on-site processing times. At the moment, humans are only involved in 80% of the transaction and 20% has been held by robots.
These automations have decreased errors but it's hard to quantify by how much. They've inserted 20,000 transactions a year. I have no doubt that the error rates improved. It's just a hard thing to quantify.
What is most valuable?
The most valuable aspect of UiPath is the fact that it's a low-code platform. Being able to use a low-code platform really lowers the barrier entry of introducing automation. Normally, you fill in a request to go to IT to get a development resource allocated, and then you spend six months trying to do a project. Because UiPath is a bit of a platform, you can quickly, within weeks, start to knock off automation and get it checked and then successfully deployed. The low-code development environment is key for us.
Now that we're scaling up and taking advantage of Cloud stuff, it's become a lot easier to use. When we started our journey, we just bought a couple of bots and had them sit around on machines. It was a bit chaotic and we thought that if we take advantage of their wider platform, the orchestrator environment, it would make life a lot easier because we have all the monitoring and management. We have access to that in one platform rather than having to watch the individual robots, which is where we started.
We're going to use the solution's AI-enhanced document understanding feature. It's something we're looking at to help us with invoices and incoming bills that come in. It's on the backlog. We haven't got to it yet.
UiPath enables me to free up capacity for people to work on new work because they are involved, they're less accessible. At the moment, the majority of our staff is focused towards the end time making sure that things are filed correctly. It's more about focusing resources rather than being more responsive.
What needs improvement?
As things become more and more data-heavy and accessing other people's products and managing things, like obtaining the data through APIs, it feels like there could be a lot more for them to do, to make interacting with data or manipulating simple things like text strings. You need quite a strong development background or a reasonable level of understanding to achieve that. I think that could be made a little bit more achievable.
For how long have I used the solution?
Three years ago we started using UiPath relatively small and we are looking to scale up significantly this year. Originally we started on-prem and as we're scaling up, we're going to move towards the cloud instance.
What do I think about the stability of the solution?
So far the stability has been good. With all of the low code platforms, it's actually more of a problem of what we've developed and deployed. It's how well we make our own software so that the platform will provide. It seems to stand up very, very well. I have not really had an issue. Anytime we have encountered a bug or whatnot, it's something we've introduced to ourselves, but thankfully there's a strong community where we can post questions and queries to get a response within a couple of hours.
What do I think about the scalability of the solution?
I don't see scalability as being too much of a challenge. If we need more capacity on the robots, we're able to buy more licenses and additional VMs on the machine. If we need to expand or scale, it's just about deploying more machines. I don't feel it's that complicated. I suspect there are some constraints on how you build your applications, but that's more of an internal decision rather than UiPath's.
There are three people who are responsible for how to put a source of business analysis as well as development. Their role is to work with SMEs or people within our business units to understand a customer's process, get them deployed, and work with them to do it. The stuff we've taken on tends to be the easier, quick wins. We have three internal developers who were able to break down processes. We're a large organization. We have a large IT function that helped us with virtual machines and data centers, etc. We're not directly involved with them.
It's very easy to build an automation and just let it run. One of the key lessons we learned is the fact that you have to keep an eye on these things and that things change in the environment. Passwords run out and expire, folders may move as people move things around the network and a robot is just as susceptible to everything else that our user is. In terms of when you're designing any solutions, you have to pay a little extra attention to things that may cause you problems in six months' time. The simple fact that a password that you were using has been reset or is expired or something else, could cause the robot to failover. While the robot can tell you it's got a problem, you still have maintenance effort to keep an eye on. There is a maintenance commitment that you need for everything that is put on it. You need to spend a bit of extra time detailing exactly how you are going to respond to those things. Just because it's easy to deploy stuff, it doesn't do away with the fact that you have to keep an eye on it.
There are three analysts who are respondents to book fixes, etc. We have people in the business who we work with to automate these processes. They take a level of responsibility and keeping an eye on anything we try to automate. They're the first line of support. If anything's going wrong or something they can keep an eye on it and then if it is a technical book fix or something that needs to be resolved, that's then escalated to one of those three developers.
How are customer service and technical support?
I only had one issue which was to do with the proxy setting when deploying some of the software. I raised the ticket on the website, got a response within half a day and it explained what I needed to do to fix it. That's my only experience of having to deal with them. I raised a ticket, I got the answer, and it worked. My experience so far has been fairly good.
Which solution did I use previously and why did I switch?
We decided to use UiPath three years ago when RPA was starting to become a bit of a buzzword. We took a look at it and realized that it would be a very, very good solution for the right project by allowing us to automate mass activity all at once. One of those projects came around and we had to communicate it to a large audience. The process once agreed upon and nailed down could be very heavily scripted. We looked at a way that we would communicate it to the 30,000 people, all with documents that are pretty much the same but with a slight variation. We knew we wanted to try an RPA solution. UiPath was a very strong contender in those days and it was easy to access. That's why we ended up with it. We're able to achieve something with a single bot. All of those things make the software easy to test out. And then from there, you're able to make a decision.
How was the initial setup?
The initial setup was really straightforward. In such a large law firm that has high data security obligations, we set these things up, appoint the orchestrator, and it just works. I have not encountered too many problems.
It doesn't feel like a heavyweight ERP system or some larger workflow tool. These things are deployed onto a desktop and they speak to a server. It's not heavy. It doesn't feel like a piece of software with a heavy footprint.
The deployment took a week. It took us longer to end-to-end to get the invoice approved.
We've taken advantage of the architecture. Our IT team set some ground rules about where the virtual machines need to be hosted and deployed, but it's not that heavyweight. We increase some standards with IT and then install the software on those machines. We're using the Cloud version so there's not a lot to worry about.
What about the implementation team?
We were able to do the deployment internally.
What was our ROI?
Some of our ROI is quite dramatic. We have to email lots of stuff out to different people and our projector will require this to have a team of six or seven staff working solidly for a couple of weeks. The robot was able to get it done after a couple of weeks of configuration. This thing was able to pay for itself in a matter of hours once it was done. One of our post-completion robots took a week or two to develop and get stable enough to deploy. It's able to offset seven or eight hours a day. If we target the use cases correctly, we are able to get a return on the automation we deploy.
What's my experience with pricing, setup cost, and licensing?
Take advantage of the Cloud-based implementation. You'll have to handle the Orchestrator licensing costs. It's obviously different for every organization. It's beneficial to get away from the on-premise installation. Also make sure that your business case justifies whatever the license cost is for an unattended or attended bot.
Show your business case and that the automation will help you to exceed the license cost. You want to look at things that are going to give you a return on investment in about six months' time. Take advantage of the Cloud-hosted version so as not to pay the cost for Orchestrator. Then for your bots, make sure you will see a six months ROI in terms of how much automation you've gotten and how much you can get the robot to get done.
Which other solutions did I evaluate?
We also looked at BluePrism and Automation Anywhere. We took a quick look over the top three solutions at the time. UiPath seemed to be one of the leaders in the area.
We partnered with an organization to help us deliver it. We got some consultants in and sorted out what they were comfortable with using and what they recommended. For us, it was the size of the platform. We were looking at Automation Anywhere or BluePrism. It just felt like it would be a bigger project to implement when in reality all we wanted was one robot to do one job for us at the start of the project. It was more about the barriers of entry to getting started.
What other advice do I have?
I would rate UiPath an eight out of ten. It feels like nothing deserves a 10, and I highly recommend every organization has a handle on RPA. There are still a huge amount of features we're still yet to explore.
Start Your UiPath Free Trial
Accelerate your digital transformation now with free access to the UiPath Platform