What is our primary use case?
We're looking to discover routine processes that our users do and ultimately automate those processes and deploy them to robots.
We've really been looking at just about any of processes. We've initially used it for processes that update invoices or update data on our system. We've used it for processes to send emails such as notification emails. We've used it to run scripts and to run data updates, and as a scheduling tool as well.
How has it helped my organization?
We've already automated about 40 hours' worth of weekly savings. It has definitely helped us to implement some of the workflows that I've shared with you and to take that work away from our partners so they can focus more on other, more valuable work.
What is most valuable?
The thing that we find the most value in is the Process Discovery component and the ability to really see what users are doing on a day-to-day basis. We're starting to make sense of the processes that they're doing and getting visibility into what they're doing. Overall, the Process Discovery is very good. The product is very new to the market so there have been some issues here and there but Kryon support has been great in working through that. It has really helped us drive understanding of what's happening. It certainly helps us understand the work the users are doing, the like things they're doing, and it helps us filter some of that work.
What needs improvement?
The product requires some more time in development. It still has some bugs and some things to work out. They're constantly releasing. They need to continue working out the general issues of using the product at scale, first and foremost.
I've talked with some of their product teams about features that I would like to see. With the Process Discovery, they're collecting a lot of data and they're spending a lot of time in the algorithms of the data. But one of the things we've talked about very recently is being able to aggregate or summarize what a user is doing in a day. That summary wouldn't necessarily be in the context of a process, but in terms of the applications they're using, the timeline of what they're doing, the number of clicks per hour. That sort of information could be very valuable.
For how long have I used the solution?
We've been using Kryon for about for about four months or so. We are using version 19.4 of Process Discovery and 19.1 of the RPA solution.
What do I think about the stability of the solution?
The stability has been pretty decent. There are certainly opportunities in terms of stability, but with each subsequent release it gets more and more stable.
What do I think about the scalability of the solution?
The product is becoming more and more scalable. On the Process Discovery side, you can collect a lot of data very quickly, so there is opportunity there. But overall, it's working pretty nicely.
We want to deploy Process Discovery across the majority of our company, which consists of about 1,200 or so people, to identify what they are doing. Over the next year, we are likely going to be deploying it on at least a couple of hundred users' machines.
How are customer service and technical support?
Tech support is great. They've been really good and very responsive. We submit our tickets via our portal. We feel that we get really good support from them.
Our interactions with their customer service are great. Really great. We've had a variety of people off and on, not a specific person. We've probably worked with half-a-dozen people, and all of them have been able to help us pretty successfully. We do have an account rep who manages our account from a higher level, and she has been consistent with us and has been really great.
Which solution did I use previously and why did I switch?
We did not have a previous solution. We came across Kryon from an RPA consultant who wrote a book on RPA. He used it a while back and recommended it to us.
How was the initial setup?
The initial setup was pretty complex. The install is certainly not for the faint of heart. There are a lot of options and a lot of ways you can install. Given that we were installing in the cloud too, it took some time and it took some troubleshooting and work, but we got there.
It took somewhere in the neighborhood of a month for us to get the AWS environment up and running and then figure out all the nuances and dependencies of the install.
Our implementation strategy was to work side by side with Kryon and our network teams and get it installed. They gave us some initial pre-reqs and things to install and things to look at. We were one of the first to get everything operating in the AWS environment, so there were some specific things to us that we had to figure out on our own.
For the deployment, we had the equivalent of one dedicated person for a month. And we have the equivalent of two people maintaining it. They are RPA developers who work on setting up RPA jobs and configuring Process Discovery, etc. We have five users in total.
What about the implementation team?
We've done it all internally thus far.
What was our ROI?
We have absolutely seen return on investment. We felt that we had to get to between 40 and 80 hours of weekly savings to get our ROI and we're already almost there, and that's good news overall. So far it's saving us about $80,000 annually. We expect that number to grow.
In terms of business analyst hours, it hasn't saved time there yet, but it will over the long haul for sure. It has definitely increased operational efficiency and has reduced human error by a pretty great degree, for the things we've automated. It does what it needs to do pretty routinely and pretty flawlessly.
What's my experience with pricing, setup cost, and licensing?
The cost is fairly similar to its competitors' costs. It might have been a little bit more reasonable than the others, and that cost was with Process Discovery in addition.
There were no other licensing costs, but there are our hosting costs and our own AWS cost.
Which other solutions did I evaluate?
We evaluated Automation Anywhere, Blue Prism, and UiPath a little bit.
The biggest difference was that Kryon had Process Discovery and the ability to discover your processes across your base, and the others did not. That was a big differentiating factor for us. Automation Anywhere was starting to come out with its cloud deployment so users won't have to deploy everything internally. That was something that sticks. But in terms of the rest of the capabilities, they were relatively at parity.
What other advice do I have?
Spend time focused on standing up your center of excellence around the solution. Kryon gave us some great advice on that in the sales process regarding defining what we're trying to do with it. Get key sponsors and then get a key team that works through the implementation, together.
What we're learning from using the solution and what we're really still trying to figure out is what is a good candidate for an RPA process, versus the other ways we could automate, whether it's process elimination or deploying some of our core engineering teams to write different tools. We're still trying to figure out what the framework is for picking the best processes.
The processes that lend themselves to RPA are processes that are highly repetitive, high volume, low-judgment types of processes, where you can write down the process in a Word document, write down the logic, and then turn that into an RPA solution.
We haven't been able to make that leap yet, in terms of Kryon's full cycle of automation from the discovery of our processes to turning on the automation and scaling it up, but we see the potential there. We haven't quite gotten to that point yet, given how early on we are in our journey.
When it comes to using the solution, for both business users and developers, you do have to have some technical background to use it. It's unlikely that you would deploy it to somebody who doesn't have any background in it. But people who are somewhat technical have been able to use it pretty successfully.
I would rate Kryon at eight out of ten. It has a lot of potential and we see that potential. It needs a few more iterations for me to bump that score up higher. It has been a good experience so far and we're looking forward to the future.
Which deployment model are you using for this solution?
If public cloud, private cloud, or hybrid cloud, which cloud provider do you use?
Amazon Web Services (AWS)