What is our primary use case?
My primary use case is automation. I worked in multiple companies with the same product on the same profile, and most of them were automation. The actual business use case would vary from company to company, and project to project as well.
What is most valuable?
One feature that I personally found valuable was the orchestrator. It is a pretty mature platform as of now, and it was three to four years back when I started to first use it. It has matured quite well. They had a major change a couple of years back. Our company transitioned from an older approach to a newer modern approach that they deployed. The orchestrator platform was very well-suited to the new approach - as was the development studio. It's really easy to use and intuitive. That has matured quite well as far as I can tell. These two are what I liked the most about the product.
UiPath’s ease of use and quick deployment times were great as the cloud orchestrator, which did not need much of a setup.
To build automation using UiPath is fairly simple. The studio is quite easy to use. Even now, with the community edition, it’s great. If we want to learn to start or try out something, we do not have to wait for licenses or anything else. That said, we can also get an enterprise trial. If we want to do something, learn something, even during our personal time, we can just download it. They also provide a free orchestrator version as well, so it becomes quite easy to learn and develop.
The building, deployment, and manual deployment processes, for small-scale projects, are very easy. If we need to build something, we just publish it, and it generates the NuGet package. It's very easy to deploy there.
The materials and the training courses are all pretty well-structured to get started with.
UiPath Academy courses have assisted in the process of getting our team up to speed. The basics were there even when I started out. I was not initially an RPA developer. I was into server operations before this. The UiPath Academy training really helped a lot with the initial courses, where they give you a tour of the platform and each and every activity. For audiences who are not much into software development, these courses can guide them towards that. The building blocks got us up to speed. They have very good courses there.
Regarding the Academy, it is a great learning platform for basic tasks. However, for more complex information, I turn to UiPath Forum. Sometimes I need some Python or C# scripts or am building custom libraries there. That gets shifted onto different platforms like Stack Overflow. We Google other platforms as well for the other types of queries.
UiPath Forum is a pretty good place in terms of the user community. Most of the queries that are posted generally get answered. Sometimes, even for smaller issues, we do not go directly into UiPath support and we first try to resolve the issues via what we find in the UiPath Community. Overall, it’s a pretty good place to solve our issues, and the community as of now is pretty active.
We saved time in our IT department since we started to use this solution. UiPath handles infrastructure for the orchestrator and its maintenance. There's a pretty good amount of time saved as we had initially had a server on-prem deployment as well. However, it became cumbersome to deploy multiple databases and they have some Elasticsearch requirements and security updates that need to be regularly maintained and in sync with UiPath. Due to this infrastructure overhead, our time could be consumed maintaining everything. Without them handling the infrastructure, we'd be maintaining instead of building automations and deploying them. We realized that an automation cloud would be a better option which is why we switched.
UiPath reduced human error. That said, we do not track errors in the process. It's a good metric to track as well, however, we currently do not track it.
It reduced employees’ time on certain tasks. The main purpose of automation is to save us the number of hours that the project will take. There are many other parameters, however, the time saved is one of the big ones.
What needs improvement?
A weakness with Academy is that, with the current updates that they have, the newer updates, the courses are not up to speed. Nowadays, Academy does not feel that intuitive and does not give sufficient information about the product to the customer.
In our current use cases, we do not need much user interaction. One shortcoming with UiPath Apps is that it cannot directly integrate with the orchestrator platform itself. For example, if we need to fetch any assets, values, or cue data from the orchestrator itself, it's basically a web platform. Even if we develop apps and want to do something based on that data, they do not have direct integration with the orchestrator. We need a separate bot, which will then interact with the orchestrator and then pass it back.
If we need ten to fifteen users who might simultaneously use apps, and we want to run a process for each of these users, then we might need fifteen licenses to do that. That is something that has been holding us back from using it, as it does not have direct orchestrator access. We need a separate bot to get the data to perform some actions.
Scaling and licensing need major improvement. I know that they have released something called Serverless Cloud Robots, where the bot machines do not need infrastructure. However, we do not generally want to run the bots or the data in the UiPath cloud as well. There are some hiccups that do happen when we run bots on another machine. That said, it might be a good feature and we can scale up and scale down more effectively.
For how long have I used the solution?
I've used the solution for nearly four years now.
What do I think about the stability of the solution?
The stability has been quite good for a couple of quarters now. We had some issues two or three quarters ago, where there was a downtime of around thirty minutes which impacted assessment. After that, for the last couple of quarters, we have not seen any issues regarding the platform itself. It is pretty stable.
What do I think about the scalability of the solution?
The product allows scaling up, however, when we deploy and run the bot, the bot does not actually run on the orchestrator. The orchestrator is just a management platform. When we need to scale up bots, what we would need to do is spin up new virtual machines. We need those virtual machines in order to scale up. This, along with the licensing affects the ability to scale.
If someone does not have a license, it cannot scale up. When comparing it to something like AWS offers, or any other cloud service, where you can rapidly scale up and rapidly scale down based upon our demands, that is currently not possible with the UiPath. We do not get to easily scale up. We need to plan in advance as to when we run our automation, what time we need to offset the loads, and which automation gets priority at that time. That becomes a bit of an issue.
As of now, scaling is a bit cumbersome, whether we are scaling up or down, and the licensing also revolves around scaling.
We use both attended and unattended automation. In the case of attended automation, scaling is a bit tricky. We need to consider licenses. Very few need the same automation to run on one hundred machines. If we do, then we would need to find a way to manage these one hundred licenses as well. It again comes down to cost.
Our thought process is that whenever possible and where we need to scale, we try to avoid human interaction, and we try to convert our bad pieces into attended automation. If some automation requires a manual login due to regulatory compliance or maybe due to capture, what we do is have these login paths that we take in as unattended input. Then where it’s possible, we would run it as an unattended mode and maybe then pass on the output to the attended mode.
How are customer service and support?
We have support from UiPath, however, I'm not sure what model we are on.
The support is not as responsible as we would expect. It's not bad, but it's not good.
The response time, the overall solutions that they provide, and the workarounds are okay. It's a mix of everything. We've had somewhat of a mixed experience with them.
How would you rate customer service and support?
How was the initial setup?
UiPath handles infrastructure, maintenance, and updates for the Automation Cloud - the infrastructure being the orchestrator infrastructure. The robot infrastructure is handled by us.
I was involved in the initial deployment of UiPath in our organization's deployment process. The deployment process is pretty straightforward with automation cloud data, so we did not need to worry about that. The administration is pretty straightforward as well. They have all of these access models, folders, and groupings. It's very easy, even for a new user that needs access to a particular box to run. It's also easy to maintain.
In terms of deployment on the cloud, there is no overhead. The administration process is simple. Maybe it took us around two to three days of initial setup. Most of the time was spent brainstorming on how we would need to structure our use case. That was what took the majority of the time. Once we decided on that, it was pretty easy. It can be done in one day. The process is also ongoing as the requirements change and the roles change, and it always requires some sort of maintenance, taking out users, taking in new users, et cetera. However, that's pretty easy.
In terms of the deployment of individual bots, it is pretty easy. The manual deployment is also pretty simple. We deploy it from the studio. We get a bundle package and we upload it at the studio level as well as the orchestrator, and it's done. This is a straightforward model. We do have a CI/CD pipeline setup for enterprises where we avoid manual deployments. In that case, we do not use UiPath CI/CD. We do use Git and pipelines to push our packages directly to UiPath.
What was our ROI?
I can't speak to if there has been an ROI or not.
What's my experience with pricing, setup cost, and licensing?
The licensing could be more flexible. They might have a different enterprise cost strategy for each of the licenses. The license is rigid in that you cannot generally scale up. To scale up, we have to have a license procured before we can run a bot there.
Which other solutions did I evaluate?
I am not involved in the comparison between Uipath and its competitors, such as Blue Prism, Automation Anywhere, or any other platform.
What other advice do I have?
We do not really have clients in healthcare, however, my previous company has mainly evolved into what I can say is a pharmaceutical supply chain. The company is an ordering platform for pharmaceuticals. I'm not sure whether I can consider that as a healthcare pharmaceutical or supply chain use case.
I had done some trial POCs around the UiPath Apps feature. We did some basic trials within our team, however, we do not have any end-users who actively use UiPath Apps.
We're still at an early stage in terms of using AI in our automation via UiPath. We tried out some POCs, and I'm also just getting training on that as well. We do not have any production use cases right now that go into full AI or ML.
In general, they have a good ecosystem of developers. It would be easier to set up and use it. However, if a new company has heavier workloads and needs scaling capability based on time, they’ll need to calculate their requirements. For example, if I process 1,000 to 2,000 transactions per day and I need ten robots to do this and it's fixed every day, then it's fine. That said, if I have varying workloads, where the workload is the last week of the month and the workload is very high, maybe I’ll need twenty or thirty bots to accommodate this workload, while, for the rest of the month, I’ll just need around five bots. That's twenty-five bots that I’ll need to purchase, with many idle most of the time. That is one issue that needs to be planned correctly during the initial stage.
Overall, I would rate UiPath eight out of ten.
Which deployment model are you using for this solution?
Public Cloud
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.