 Live from Las Vegas, it's theCUBE, covering UiPath Forward Americas 2019, brought to you by UiPath. Welcome back to Las Vegas, everybody. This is day two of theCUBE's coverage of UiPath Forward Three. This is the third year of the North American Conference and the second year theCUBE has covered this. Anthony Abatista is here. Cube Alami was on last year from Deloitte. He's a principal there. Anthony, good to see you again. Great to be here. It's a great day. Yeah, so it is. I mean, we've seen the growth of RPA, generally UiPath, the whole automation. We're starting to talk about intelligent automation. AI has its wings and is starting to soar. But give us the update from a year ago. We talked about accelerating last year. I think it was, you had some really good statements around, look at, yeah, going fast is good, but you want to accelerate the right things. Speeding up bad processes, paving the cow path, as I sometimes call it, is really not the way to go. But what's new? So I do think there's still some issues around getting programs to scale and thinking about automation at scale, which has been a major theme here at the conference, is still in front of us. People are still figuring out how to climb up that curve. But what I think is new is when we thought about automation before, it was an or statement. Was it humans or automation? Is Abat going to replace the human? And I really think we've, Deloitte's always had a campaign about eye to AI that we kicked off a couple of years ago and said, yeah, how do we have automation and humans interact with each other? And I don't just mean attended bots, but how do we actually start to use automation as sort of the glue to hang together a much more rich experience to start to put the AI components in there? So that leads us to the age of width, which is how do we use the technology along with humans to change their role. And there have been some great talks, one of my partners earlier that was here with Walmart, his client, and they talked about how they're redefining the HR processes at Walmart. And that was, I thought, a really good presentation because they changed the workers' work, they didn't replace workers. So how was this concept of the age of width, how was that different than attended bots? Can you maybe talk about a possible use case or example? So if we think about a call center, you know, if we know who's coming in, we used to just look them up and say, hey, do we know who's calling? Now we can say, do we know who's calling? Do they have a history with us? We can use data, and that's another part of the width is data plus analytics with automation. And we can say, well, what else do we know about this person? Do they have a history of calling us? Do they have an open ticket? Have they had some issues or complaints in the past that we can deal with or get in front of? And basically start to put the intelligence in the front end, and that could be unattended, right? That could just be some screen pops around automation. But we start to introduce natural language, we start to introduce some advanced analytics, and that would be a simple way of enhancing that process. So let me double click on that. So normally what you would get is you're in the other end of the line of the call center, and it's like, hold on, I'm just reading the notes. Right. And, you know, they're scanning these notes. It's like an eye test, you know, and they can't ever get to. See, it's faster for you to just explain. And let me tell you what, you know, what I'm imagining is in different experience where this is happening in near real time, you're either getting pop-ups, or, you know, some other messaging, is that what you experience, and how real is this today? This is real, and, you know, I always like to say, automating is easy if you just take the process, repave the call path, but it's very real, because the natural language components, they work up front now, you can ask some questions, you can start to do pre-searches on which materials might help with that type of question. You also can train the process over time. So data over time, you know, what's the call satisfaction? Did you actually complete what it was the call got, you know, started about, and how quickly did you do that? So you can train these models and start to use machine learning to actually improve that experience even further. So I think that's the, again, back to the whiff, it's adding these components. I like talking to folks with a consulting background, because, you know, when you talk into the vendor community, they get very excited about, you know, the ROI and how, you know, lack of disruption for to install some software, right? And so that's one of the advantages, I guess, of RPA is you can pop it into an existing process, good or bad, and get going right away. We've seen this, you know, time and time again in the industry, when you have a big, you know, to force people to change, you know, it's slow. When you can show immediate ROI, you start to see these rocket ships. At the same time, as a consultant, you really want to have a bigger impact on a business. You don't want to just repeat, you know, automate bad processes. So it's, you know, how do you work with clients to sort of manage that insatiable desire for quick ROI and then the transformative components that, you know, could maybe defend you from disruption or allow you to be an incumbent disruptor? So I think what's interesting is transformation, use the word, we're really good at transformation programs. So starting to say how do we think of automation first, as we do a traditional transformation program is very near and dear to us now. And instead of saying, hey, we're going to bolt the ERP system in and then figure out if we can get some improvement by automating later, we're saying, you know what, let's sort of go backwards. Maybe it's a little old fashioned, but what does this whole process look like? And can we put automation in launch, not as a process improvement after launch? So I think when we think of these transformation programs, bigger ERP programs, we're really in there doing it. Automation's now on the front end of that program rather than afterthought, like reporting used to be. Right. So I mean, you guys have to be technology agnostic in your business. I mean, we happen to be at the UI path conference, but there are, you know, RPA, generally UI path specifically, it's not a panacea for all problems. I mean, we've talked about AI, we've talked about other automation process, automation capabilities, you've got existing systems, all this stuff has to work together. So, and people always say technology lasts, people process first, you guys live that. So how are you seeing automation evolve in terms of the adoption, how people are dealing with existing systems and some of the other technologies that you're having to bring together? So I think the first thing is the technology has to work. It has to be bulletproof and resilient if you're going to put it in these processes and make it part of your work life or serving clients or that sort of thing. So first it needs to be bulletproof. That's becoming a given. Second, I like to think it's well architected, more and more as you bring in AI or other advanced components. You need to be ready to have a changing ecosystem. You know, the best document processing right now might not be the best in six months. So starting to think of your automation solution as that technical glue and doesn't allow you to swap out your trade components as you refine processes going forward or something new hits the market. So now we're working in an ecosystem and I think for the RPA vendors that are having great success in the market, like UiPath, they sort of, they give you that platform and they give you the off-ramps and on-ramps to integrate the other technologies. And I think, like I said, I think that's table stakes in addition to being bulletproof, but the next piece of that is how do we get various people involved in the value proposition of creating automation? So various tools and studios, some for the business user that might not be as technical, maybe self-design a bot at a process description level and then maybe a more technical workbench for the technical bot builder. So I'm starting to see that in the product suite and some of the announcements here this week, how do we tailor the tools to different users and engage them in that process from one end to the other? So you mentioned scaling before, what are the blockers? What's the challenges of scaling? Why does it seem to be so hard? It's clearly an area of focus here at this event. So I think first of all, the technology is still new to some areas. There's still back and forth whether it's the business or IT-led initiative. I think there are some scars and wounds over the last few years of automation where people might've gotten started on the wrong foot. So there's even some redos to learn from. So I think people are looking for the business case. They're getting more comfortable with it. So the job sizes, the deal sizes are getting bigger for the RPA vendors and for us. But I think it's just an evolution. And like I said, there are a lot of stub toes early on in automation. What are some of the big mistakes that you've seen people make? People thinking that it's only a business tool or only a technology tool or technology to the point that they get started on something that becomes either a real technology problem or a real business problem. Maybe you build a bot out in the business and you attach it to your ERP system and you cause performance problems or you have security problems and then it becomes a real IT problem. I've also seen the reverse where an IT group will start and say, hey, let's do some automation. And they push it into some departments. It might not have a fully-baked business case. It might not have good support and it becomes a technology science project rather than delivering any real value. I've been trying to think about analogies and analogous ascendancies in software. I use service now a little bit, but that was kind of a heavy lift. It started in IT. It was very clear IT. You're seeing this massive rapid growth of UiPath, fastest growing, probably the fastest growing software segment in history. And it's striking to me that we're just now starting to see cloud come into the play here. If we just, UiPath, I think announced this week, it's got this new SaaS capability, which you would think would be born in the cloud, but people have explained sort of why that is. Do you have concerns about the pace of growth and a company like UiPath and its competitors, their ability to sort of keep up and continue to deliver quality? I mean, a big part of what you guys do is sort of risk management as well. So how do you manage that risk? So I think what you look for, if you're going to be an alliance partner, if you're going to work together and proceed things together, first you have to have the basics. It has to be bulletproof. It has to work. When you hit bumps in the road, you have to have escalation paths that make sense. And there's growing pains in any firm or any company that grows as quickly as UiPath. On the other hand, the question is, are your cultures aligned? Do you know how to fix problems? How do you put your customers first? And I think that's what we look at in the alliance, which is how do we, how do we partner with people that have similar DNA about customers first and you put everything else aside, roll your sleeves up and do the right thing. So that's what we look for in the alliance like this. We always talk about the buzzwords of digital transformation, which, you know, at conferences like this, it is kind of buzzy. But when you talk to customers, they're actually going through digital transformations. And then a couple of years ago, they started experimenting. They bought one of everything and they'd run things in parallel with legacy systems, but now they're starting to place their bets. They say, hey, actually we've got some use cases that are working, we're going to double down on the stuff that we think works, RPA in some cases fits there. We're going to unplug some of the legacy stuff and try to deal with our technical debt. But I guess my question is, where do you see RPA fitting in to that whole digital transformation matrix? I like to think of a matrix where you've got different sets of services and you've got different industries that are tapping, you know, all data-centric, that are tapping these new capabilities and formulating new businesses, new industries. That's how you see this disruption happening. And then the incumbents saying, hey, we've got assets too, we're going to tap that same matrix. And whether it's open source software or cloud or, you know, new security paradigms or data and analytics. So where do you see RPA fitting into that matrix? So I think at the glue level, at the architecture level, it can be the orchestrator of that experience of taking a variety of technologies and integrating them, providing, again, on-ramps and off-ramps, doing what a human can do, looking at screens, you know, analyzing content. So it can be the glue that orchestrates those processes, orchestrates maybe some of the, some of what used to be a void between legacy systems and new systems. And RPA helps take all that away or level the playing field on that. So X is another set of eyes and ears for process integration or technology integration. And I think that's what is probably its best place now. Are there good process tools there? Can we get, you know, community development's a big discussion, you know, right now? I think some people have been successful at it. But, you know, it requires a lot of care and feeding and planning to have, you know, your community stay on the rails or stay between the curbs and do useful things. So I think we're in the beginning of how far can we go with community development? But again, I think the technology is really the glue. So community development in terms of best practice, sharing. Yeah, and users, you know, developing their own bots. You know, what are the guardrails? Does the process, their automating matter? Does it introduce risk? Is it going to perform? You know, how do you make sure your bots are uneval that people are creating? It's a pretty powerful technology. Is there IP in there that you don't want to, we talked about this last year, that you don't want to necessarily share with others? So, now your role, you used to have focused specifically in financial services, now you're more horizontal, but how does Deloitte look at this opportunity? Is it an automation practice? Is it, are you cut across all industries with automation or is it more sort of broader than that? So my colleague here runs the offering, which is do we have the people, the training, the tools, the delivery centers, in the know-how to go out and do this kind of work and we've scaled tremendously in the automation space. The second part is how do we look to the adjacencies? So we work very closely with our colleagues in AI and ML, and we say, how do we go do the next generation of this out of the gate? How do we experiment? How do we say, do you want fries with that as we do some of this work? But then we look for the industry intersection and that's where a firm like Deloitte, we've got deep, deep industry expertise. We say, what are those intersections where we can go make something happen, we can work with our partners, our alliance partners in making something happen at an industry-specific level or can we go solve a specific problem? So I think that's what we bring, that's unique, but we do it both ways. Yeah, I want to ask you a question, it's kind of off the topic here, but I was talking about that matrix before and again, I'm envisioning technology, horizontal technologies and then vertical industries and it used to be for decades, if you were in financial services, you were pretty much stuck in financial services, you had a value chain that was specific to financial services and you knew it inside and out, whether it was product development or marketing or sales distribution, whatever it was. And same thing for automobiles and manufacturing and education on and on and on and you developed these industry areas of expertise and you had domain experts within there and you guys have been built up a global powerhouse doing that. But you're seeing as digital, as companies become digital, what's the difference between a business and a digital business? It's how they use data, data's at the core and you're now seeing organizations, companies, tech companies specifically, traverse different industries. You're seeing Amazon in content, you're seeing Apple in financial services, other companies getting into healthcare. How is that, first of all you see that and what do you think was driving that and how does that affect your business? Are your clients asking you to help you traverse new industries, get into new industries or defend against others, these big tech companies trying to, with a dual disruption agenda, trying to take them over? And at the center of all that you mentioned data though, but the center of that is who are the ultimate customers and what experience do they want and how do they want that experience integrated? So it's not channel by channel anymore, it's which pieces fit together and how do I want to buy things? How do I want to be serviced? You're getting whole crossed economies around what the consumer wants enabled by technology. I think the other thing that plays into that is you start thinking of the internet of things and how connected people are and how do you use, monetize, and integrate data about particular people and how they want to be served to make that a better experience. I think the consumer ultimately is driving a lot of that in technologies enabling it. Yeah, and as you think about that picture again, I like to use the metaphor of a matrix. I mean, I see RPA as just one piece of that. There's so many others, you mentioned IoT, we talk about AI all the time, we talk about blockchain, it's how you put those different capabilities together and apply them to your business that really makes the difference. Not that, I mean, RPA right now feels very tactical, but it's part of a much more strategic agenda. Absolutely, and again, it could be the glue and the ecosystem of emerging technologies. And again, I do see it as the eyes and ears, the fact that what you get out of the box from a regular RPA vendor can really integrate some really painful things, like looking at spreadsheets and looking at guys with green visors, looking at columns of numbers, it's really good at that stuff as a base task. Yeah, nothing wrong with tactical and quick ROI. So Anthony, thanks very much for coming on theCUBE, really appreciate your time. Thank you, great to be here. All right, you're welcome. All right, keep right there, everybody will be back with our next guest, day two from UI Path Forward in Las Vegas, you're watching theCUBE.