What is our primary use case?
We're specializing in law and local government, so that's our target market at the moment. Obviously it's usable everywhere, but that's our market, and we find that that's where there's a lot of opportunity for sale.
How has it helped my organization?
In one of our legal scenarios, we had a situation where they were trying to deduplicate their database. They worked out that the man-hour cost for deduplicating that database was 11 man-years. We got that down to 30 days with a saving of 3.89 million pounds in what they would've expected their cost of that process to be.
As a sales-based operation, we get a lot of inquiries through our own website. We were finding that we were extrapolating that information, putting it into an email, I think into a spreadsheet, putting it into Salesforce, whatever. We've now got our own robot, which we've built, which opens up the backend of Wix, which is our website building tool. It receives all those inbound inquiries. It then goes out and looks at the public domain. We'll use that email address, or use a name against LinkedIn, against Facebook, against Twitter, to find out any more information we can find about job titles, maybe phone numbers or email addresses if they're missing. Get all that data out, build up a real case for that client, then the robot would enter that data into Salesforce, it will send them a welcome email to say, "Thanks for you inquiry, please find attached the information you've requested." Bring them straight into Salesforce, set up a reminder, and then send me an email to say, "Follow up this client, they've downloaded some information from our website." We've automated that whole process, and we can then use that for events, and seminars, and webinars, everything else that we hold, because they're genuinely interested clients, and we get around that GDPR issue.
What is most valuable?
For us, it's the ability to offer as a subscription model. We looked at the other competitors to UiPath, and we found that the way of delivery that we offer is a subscription model. It's a very, very small upfront fee and a monthly cost, and UiPath gives us the flexibility to do that. Some of the other vendors don't do that because they sign you up for ten licenses over three years, and it's just not practical for us.
From a cost perspective, it's not really relevant. It's really about the processes they're trying to automate. A lot of people are doing data deduplication and data onboarding, so an attended robot is absolutely fine. Whereas when we're trying to do things like answering an email, an inbound email, set up a process, respond to that client, then a 24/7 unattended robot is the way to go. We use a mix of that with our client base.
What needs improvement?
We've had some serious issues with clients that are running in Citrix environments, and we've got a couple clients that have moved away from other competitors to UiPath and to us because they just cannot do that screen scraping technology. Yeah, we're finding a lot of it in a Citrix environment, and a Citrix environment, on an Azure cloud, on a virtual machine. So there's various steps, and UiPath's the only one we've found at the moment that will actually enable us to deliver that.
I think the AI question is being raised everywhere, and there certainly needs to be a lot more intelligence involved in that. The ability for the robots to start thinking for themselves is coming out in the later versions. There are very few limitations at the moment, because we tend to pick relatively mundane, tedious, repetitive tasks, and it's all about number crunching. Really, RPA globally is all about number crunching. But in the next six to 12 months, there'll be more intelligence added, and there'll be more stuff going on that brings it closer to the AI environment that everyone's spouting about at the moment.
What do I think about the stability of the solution?
We've been involved in automation for ten years or so, and UiPath for about two or three. We've had no major issues from a stability perspective.
What do I think about the scalability of the solution?
Very much so, and more so with the later revisions of the software. But we tend to start off with one project, one workflow, and then we're all about scale. But it's a land and expand approach for us.
How was the initial setup?
We always say that any idiot can build a robot. In reality, it's working out the workflow, working out what's involved, making that process lean that takes all the time. That's where we get involved. Actually setting the robot up itself is very simple. Some of the discussions we've had today talk about clients setting up their own robots. It is a relatively simple, drag and drop, point and click type setup for us.
What was our ROI?
We're averaging three to six months ROI, because we offer this subscription based model. There's no massive upfront cost. I think the biggest ROI we've seen is the one I mentioned with the law firm, where we estimated 3.9 million pounds of saving, and we actually realized 3.899 million pounds of saving. That was by far the biggest ROI, but as we start working with bigger councils and bigger law firms, those ROIs can only increase.
Not sure about percentages, but one of the anecdotes I always say to my clients is we've all managed to write our own name incorrectly on a web form, so if we cannot spell our own name correctly, what chance have we got of spelling the name of a client that may be from overseas? The reduction of errors is phenomenal, especially in that mundane, workaround environment that everyone seems to be in.
Which other solutions did I evaluate?
The model that we offer, as I say, is a subscription model. The idea is that we land and expand. We charge a very small fixed setup fee with a monthly subscription, and a minimum commitment to us is three months. We believe that we have to prove to the client that robotics is the way forward. If you go to one of the competitors, such as Blue Prism, they will insist you buy ten licenses, you sign up for a three year deal, you do a 50,000 pound proof of concept, and then at the end of the proof of concept, they build you a live, working robot. Our argument is that we've already built you a live, working robot to do that trial with real data, with real cases and scenarios, and with real workflows, and we just make it from a dumb environment to a live environment. Yeah, we absolutely compete with those on a very different level.
What other advice do I have?
We're very, very happy with the product. It does exactly what we want it to do, and it allows us to sell UiPath, or sell RPA in the way that we want to. We're not being dictated by the manufacturers to how we have to sell their product. We know our customers best, and we believe that our methodology is the way forward. That's the flexibility we get with UiPath.
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.