 And we are live here at Orlando, Sapphire now. This is theCUBE, SiliconANGLE.tv's continuous coverage of Sapphire, where we bring you all the information, the smartest guests. We extract knowledge and share it with you, our community. We're here with Sean Blevins, who's with Opera Solutions. Opera's a company doing some very interesting innovations around data, around big data. One of the hottest trends. You don't hear a lot of that from SAP. You know, last year we were talking about this on theCUBE. Maybe Bill McDermott mentioned it once in his keynote today, or this year about 10 or 15 times, so they're really starting to get it. Sean, welcome to theCUBE. Glad to be here. First time guest, and we were talking off camera a little bit, Opera. I know a little bit from our man, Jeff Kelly. He talks about you guys a lot, pretty hot company, doing some pretty innovative things. You are very familiar with Sapphire, very familiar with SAP, my co-host John Furrier. So, talk a little bit about Opera. Tell us, set it up, set up the discussion here. Who are you guys, and what are you all about? Alright, so Opera is an eight year old company. It has about 650 employees. What's interesting about it is it has 230 data scientists, so compare that to Google with about 300 scientists. Ours are all pointed to the commercial side of the business, and they do these very targeted industries and lines of business. So the whole premise of Opera is to industrialize big data predictive analytics in a way that can be used by the C-suite. As things become more complex, when you add more layers, more data warehouses, the CEO, the CFO, they've had to, at the CMO, they've had to seed all of that complexity down to IT. So they can't actually get just the core agenda items that they have met. And so what we've done in this company is we've built a series of signal hubs that take big data sources outside the four walls of the business, unstructured social information, Yelp, Facebook, Twitter, comments on the websites, government information, and we play that out by industry in a set of signal applications. So if you want to do customer attrition and fading, and you're a CMO, and you want to see exactly how to treat your customers, what products to treat that fading, how do you get a curriculum built where they buy and cross buy? There's one example of that where we used a customer, they deliver food to homes, and they come in with their hand held. So in the span of a year, we took two control groups, one who didn't get that treatment, and that big signal information coming in, and one who did. We drove $100 million of new business to the bottom line in recommendations. When you buy X, you haven't bought this in a while, you should buy Y, and have you thought about this. The other group churned out minus 16%. So that'll give you some idea when you're able to look at beyond just the nice little neat rows of data inside your business, but you're able to get to the outside and see how customers buy, what they prefer. If it's healthcare, it might be revenue leakage and things that are falling out of your business. You look at the line item invoice from a big data perspective, and you see this should have been for Tylenol and not three, so you missed that one, and Tylenol's in a hospital, or what, $1,000, right? There's a lot of revenue that leaks out, and so maybe they're not that much, right? But these solutions allow you to have an industrialized, repeatable play out of the box for big data and predictive analytics. And we announced yesterday, and you may want to talk more about this, we announced our partnership with SAP HANA so that, you know, if you're crunching all this information, before we'd have to batch it, we'd have to stage it across a bunch of servers, now we can literally put both the signal information, we can play more signals, more types of information, and the data science that our scientists produce, the algorithmic models, the blending of those models, all of that goes into a simple environment in HANA, and now we can collapse the time window from what used to take an overnight batch, and maybe even multiple days, to do that into moments, into minutes, or even seconds, depending on what signals and what kind of information you're trying to drive. Okay, so, I do want to talk about HANA, but before we get there, so you're taking this, so today, unless you get the wall to garden approach, right, very complex, and relatively inflexible, you're bringing in all this external data. There's a lot of exhaust, John calls it exhaust, there's a lot of noise in there, so you're saying you filter that down, and get to the golden nuggets, the needles in the haystack, is that right, is that the promise that you're delivering to the customers? That's exactly right, if you look at the business today, I'll give you an example, we were talking to a customer, and for the last several years, their revenue, based on demand forecasting, was stuck, right, 80%, and they have process optimization, they have all, they've done everything right internally, there's nothing that they could do any better from a best practice point of view. The challenge was, they can only see inside their four walls, right, so the options for that are, you buy a tool, you try and find a data scientist, and, or maybe you back a busload of consultants up, and you either train up your person on a tool, and you get this much lift, because the tool does statistical analysis of one kind or another, or you wait two years and you hire 20 people, and at some point after $30 million, then you get the lift you want, but, you know, you've had to wait for that. We're talking about weeks and months here, maybe 16 weeks, until that company can go from 80% accuracy to 90% accuracy, because we bring in all those sources outside the business. So it's very real, it's immediate, and I think the other point is that it's exponential, it's not some incremental 2% thing, it's 30, 40, 50 times what you would expect, because it's just the results are as big as the data. Sean, so last year we covered deeply with big data, and you're hitting a big business benefit, so it's not just like, oh, social media campaigns, this is real business model stuff. That's right. And you're talking about weeks versus, you know, years in ROI in terms of deployment of an actual solution. So it's pretty obvious. Your doors open up pretty fast, I imagine, when you make your pitch to the businesses, but how do you go to the market there? I mean, obviously, on the business model side, I mean, because you have to go right to the top, I imagine it's business line. That's exactly right. Take us through how you engage with a prospect, for example, and what's the value of the pitch. Okay, so in a nutshell, the first thing is it's non-disruptive, right? So we don't come in there and say you got to buy these things or store this in your infrastructure, you know, redo this, build yet another data warehouse. None of those approaches are our approach. We go directly to the C-suite and we say, okay, you're a chief procurement officer. You're running $100 billion a year of spend. We will guarantee that we got $30 to $50 million worth of spend we can put back on your bottom line. And the way we do that is we're very focused on non-disruptive as a service coming in. We know where those sources of information are. We know where those categories of spend are. So we focus not on putting the data in nice little refrigerators, right, but streaming it, right? So looking at it as panning for that gold that you mentioned, I think that's exactly right. And so we come in and talk to those C-suite about just those core agenda items that they need. And in the case of some of our customers, you know, we're six months into it, nine months into it, we've done it for someone else, particularly in the areas of spend, fraud, risk, marketing, all those kinds of areas, you know, we find that organizations need to get their media spend right. They need to deal with the friction. So if I hear it correctly, then you go into the verticals, you build out the methodology, and then it's leverageable, you can go to other customers. That's exactly right. Because even though it's different data, you still have external sources. It's exactly right. And the external sources are acquired through deals or just open APIs or how do you acquire the data? Oh, that's a great question. So obviously the first step is a privacy and NDA agreement, right? And then for the internal data, and then the external data, we just know where it is. And our scientists really come into two teams, the ones who build the predictive models and the data science machine learning piece, and those who know how to extract the signals. And so extracting the signals is like, you take 50 terabytes across 30 data sources, and that becomes a DNA strand of a terabyte of data. So we're looking for time series over history, behavior, data, and then derived data, basically. Precisely. You got it. That's exactly it. Awesome. I'll give you an example of how HANA drives the change around. And we're showing this in the demo floor over here at 228. If you want to go see it, we had a wealth management customer. This customer had a very specific need. When an analyst makes a recommendation, they had over, you know, almost $2 trillion in assets in inventory. They had 30,000 financial advisors, you know, millions of portfolios with all these different assets managed, unmanaged. Every one of those customers had different risk tolerance levels. They had different propensity to take action. They had the success rate of that analyst over time. So when he says buy this asset, and he makes that recommendation, our solution would go in, and it took about two and a half days. But that advisor would come in on that on that next Monday morning, and there would be five, not 500, but five recommendations that said, get out of stock A, and the machine is writing this language. There are no people involved, right? The machine language says get out of asset A, get into asset B, and you can go from a minus 3% in this portfolio to plus 5%. So the customer obviously loved that. The problem was, we could only do that every two, two and a half days. Honest lets us do it every 30 minutes. 40 minutes. That's huge. Maybe maybe I'm being, you know, maybe I'm being a sales guy. Maybe it's 60 minutes. But it sure is. It's multiple times per day. Yeah, so we're less than a day. Yeah, and we were driving, you know, ridiculous revenue before because people obviously we could go into portfolios and say, you know, here's why you should get out of asset A and B. Now it's an intraday thing. Let's talk about revenue. So, okay, growth side, how you guys self-funded. I'll say you must be throwing off a lot of cash flow on your end. You've ventured back. How old is the company and how do you guys grow? Companies eight years old. It is cash neutral at this point. We have received a first round series A investment, a minority investment from Silver Lake Sumeru and Excel KKR. They own, you know, they own a minority stake. And the I believe the company was valued somewhere around half a billion dollars. And so much is just taken. Oh, well, I mean, the private company. But did you announce the funding? We did announce the funding. Yes. 84 million. 84 million. No, I think we're about 100 mil this year. No, no, did you announce the did you announce the funding? We announced the 84 million funding. So they brought in 84 million funding and you're about 100 million in revenue. Exactly. Yeah, which is okay. So I gave you some of the idea of the valuation. I don't know why it's so so low. I mean, look at the what's going on. I think it's exposure innovation bubble. It's exposure. And I think the the the linchpin, the watershed event is having Hannah marriage made in heaven. They need that IP to let them go sell the C suite. We need the speed and performance. We and the collapsing of all the complexity. I mean, when you think about having to do this and staging terabytes of data in the cloud and doing it on premise, the ability to collapse that in a single environment and and take a scientific model. So if you go to the floor over here, our demo literally has k-means clustering and algorithms running in memory in Hannah. So we, you know, this is we have a built set of solutions. And then we can pick those up using the open standards in Hannah and just plug them right in. Well, we're big fans of what you're doing because one, we have our own little small scale analytics for our media. We have one data scientist. We use it. I do, but it's free. But what you're doing represents really the future around instrumentation of business and using data to drive real business value. So congratulations. And I love want to follow you guys for forever at this point. I think you'll be very, very successful as you roll out these verticals. The question I have is going forward, okay, in the customer environment, what's the preferred user interface for the customer? Is it dashboard driven? Is it obviously it's a great question? Are you platform as a service? What's your, how does it, how does you look there? You've put your finger on probably the secret to our success, right? Because we do that. You need a new class of interface where it's not, here's really the key. The current paradigm is I have a human being doing some unit of work and I'm going to let a machine do that work. What we've found is it's both, it's man and machine. The analogy that I've heard before in our company is you take a grand chess master, you put him up against Watson, it's 50-50, it could go either way. But you take an average chess player, just someone that knows how to play chess pretty well. You put them with a computer, he beats the grand wizard and the machine every time, right? Because it's human intuition plus what the machine sees. It's getting past our bias, getting past the way we want to see the data, the way we want it to happen. AI has always been grounded in training computers with some reasoning, meta-reasoning, if you will, in this case. And humans are the last mile, as I said. That's it, you got it. So the front end has to be different to say, for example, in our spin intelligence solution, you don't do that data munging or anything else. You walk in, right, and it has like a Facebook post that says, while you were asleep last night, I identified these six areas of spin. Let's go explore them together, because I think there's big money to be made here. So what's the UI? Is it dashboard based? Is it HTML5? Sometimes it's dashboard. It's HTML5. It's those kind of things. Sometimes it's been, you know, it's usually just a runs on iPad, ubiquitous dashboard, but can just as easily be embedded into other solutions, ran on mobile devices. But the API systems are... Yeah. So you're at a self-service? Exactly. We did partner with DataViz, the data visualization folks, so we got that library. And then we've continued to, our focus really going forward is to have front ends that allow you to have these kind of man-plus-machine conversations. I think this is something. Yeah. Is it something... I'll give you an example. Yeah, it's an Apple computer, Apple Fruit. You need to have someone do that. That's it, right, exactly. So at... Have you heard the story about the AIDS virus? That they took a machine and they tried to crack the AIDS virus, the actual gene for it? And they couldn't. And then they gave it to people and they couldn't. Well, they turned it into a game. And inside of six weeks, someone had actually, you know, they did a big game, lots of multiplayer's playing here, you know, folding. They used to fold it and six weeks later they'd crack the AIDS virus. And so it was the machine plus the people that made that happen. So how many customers do you guys have? Because that's really the core adoption, a bottleneck at this point. Just not enough people get it. So I'm sure that's a... That's a great question. I guess it's a sales inhibitor. But so how many customers are you currently working with? Well, the heritage of this company is interesting. Right now it's Fortune 50, very large wealth management, credit card issuers, retailers, and as you can imagine, this is a, you know, you walk into the C-suite, this is a competitive advantage. Right? So they're like, okay, well we, for the next six months we don't want you to talk to our competition. Right? So yeah, you got it, exactly. Add a couple zeros to the back of that one. Exactly. And so, but I guess the key here is we're starting with HANA, we believe that there's an opportunity to bring this to market at a certain volume. And do these, do these signal-based applications. And you're like 10 customers, 100,000? Oh, I would say at this point for different solutions we have 15 wealth management customers for our Mobius solution. We have about 130 for what we call our BIQ product, which is the spend analysis piece that SIP is now the executive dashboard for. You know, but it's, I would say it's somewhere around that. We're definitely at the stage right now where I think it's going to explode as people understand the paradigm shift. We're talking hundreds of customers, right? Yeah, about less than 200. Yeah, so let's talk about going forward, because you also are in a sweet spot. It's going to explode. You kind of, you're on that line and all of a sudden it kicks up and goes vertical when the growth comes in, when people kind of rock this. When you go in these conversations, what's the critical success, what's the critical, what is the biggest sales inhibitor for you? Put aside that they don't get it, but like when you go in and you say, okay, I can do X for you. Right. I can do kinds of instrumentation provide this kind of real value. Do they think like almost too good to be true? Is it like I want to do a proof of concept? Where do you kind of, great question, get that point there? Well, if it's, if it's an executive, if it's somebody who has walked into a boardroom and said I'll deliver this result, they're already drowning in information. They've got report after report, but they can't make decisions based on fact beyond what's in those reports. Right. And so a lot of times it's about, we'll take those information and the first step was we'll bring back that set of signals and we'll say, here's 45 ways of looking at your business you've never been able to see before. And that usually says, okay, now you tell us, you want us to play it out for revenue leak, you want us to play it out for marketing spin, for dynamic pricing, you tell us where you want us to start. And usually they have a very discrete set of, these are the goals for the company and we want to go that way. But it starts usually with getting their information and extracting that signal so that they can feel comfortable that we can do what we say we can do. Sean, we're getting the break signal here. We could also go on and do a whole day with you. We love your business. We love what you're doing. We think you're way ahead of the curve right now. Thank you. Congratulations with the HANA deal. In terms of performance, I think Dave, we should dig into that a little bit later, but breakthrough days. Sean, I love the enthusiasm. You were with SAP. I was. And you've popped up for this opportunity. SAP is a great company, a great place to be right now. Seven years of SAP running competitive sales and going up against our competitors. You heard lots of comments from Haas Saturday so there was a lot of fun times there. And I just think that this is a new class of impact to customers. And it was worth it was, you know, I think coming back here and seeing HANA working with our solutions is the perfect is the perfect marriage, the perfect way to go to market together. Well, congratulations. Good luck on your new entrepreneurial venture. You guys got some big, fat financing. Great customer base to build off of and great revenue run rate. Congratulations. Thank you, sir. Thanks for coming inside the Cube. Thank you, Sean. This is a special Cube coverage. A big day of meets business value. We'll be right back with more coverage from Sapphire now inside the Cube right after this short break.