Qlik Connect 2024 | Mike Capone | Qlik & Sarvant Singh | Penske

theCUBE.net

Clips
News
More from Qlik Connect 2024

Mike Capone

CEO

Qlik

Sarvant Singh

Vice President, Data & Emerging Digital Solutions

Penske

Qlik Connect recap: Where there’s data, there’s opportunity

As I recap Qlik’s Qlik Connect event held this past week in Orlando, I can’t help but start with Qlik Chief Executive Mike Capone’s comments in his keynote: Where there’s data, there’s opportunity.Data is the foundation of everything today, in AI and pretty much everything in between. Today, we are experiencing a new dawn of data and a new era of opportunity. With the rise of artificial intelligence in general and generative AI in particular, we have reached an inflection point that is changing everything all over again — and at the core of that transformation is data

Three insights you might have missed from theCUBE’s coverage of Qlik Connect

Comprehensive data solutions are becoming increasingly important as generative AI adoption grows. There is a broader platform shift ongoing, making accessibility of quality data crucially important.How to respond to this seismic transformation in data integration and analytics innovations was a central focus of Qlik Connect. The event made it clear that QlikTech International AB has a strong culture, but also that there’s no such thing as an “easy button,” according to Shelly Kramer, managing director and principal analyst at theCUBE Research, during the event.Shelly Kramer of theCUBE discusses what’s next for Qlik during Qlik Connect 2024.“Data integration is key

Qlik debuts new data integration and curation tools for generative AI systems

Data integration platform provider Qlik Tech International AB this week debuted a new product called the Qlik Talend Cloud, saying it’s designed to deliver high-quality, trusted data to artificial intelligence systems.It was announced alongside a new generative AI assistant that promises to help companies make better use of their unstructured data. The announcements came earlier this week during the company’s annual Qlik Connect event in Orlando. There, it also revealed strategic alliances with Amazon Web Services Inc. and Snowflake Inc.During Qlik Connect, the company was keen to stress the importance of good data and proper data management in enabling successful AI strategies

Penske Truck Leasing and Qlik seize the moment with generative AI for predictive operations

As a major provider of business intelligence and data integration software, QlikTech International AB is leveraging generative AI to capitalize on the surging interest in artificial intelligence.With a portfolio of intelligence-based products spanning more than 30 years, Qlik believes that the market has come to it.“We’ve been getting ready for this moment for years … this was the generative AI moment when we finally crossed the chasm and said, ‘OK, now we can actually make this stuff productive,’” said Mike Capone (pictured, left), chief executive officer of Qlik

play_circle_outline Qlik Connect event in Orlando, Florida featuring CEO, customers, and partners

play_circle_outline Importance of data quality and foundation in AI implementation

play_circle_outline Shift in thinking about AI investments and operationalization

play_circle_outline The Rise of Workflow and Data as Intellectual Property: Empowering Gen AI with Self-Service AI Development

play_circle_outline The emergence of new operating models for neural networks and knowledge graphs

Info

Mike Capone | Qlik & Sarvant Singh | Penske

Mike Capone

CEO Qlik

Sarvant Singh

Vice President, Data & Emerging Digital Solutions Penske

search

>> Good evening, Qlik fans and welcome back to Orlando, Florida. We are here for our last segment of Qlik Connect. It has been an awesome day. I feel smarter than when I woke up today. My name's Savannah Peterson. Delighted to be joined by John Furrier. John, what a show we've had.

>> It's been a great show. It's very action packed, but don't feel it. It's really well-designed show. Those stories popping on all cylinders. Generative AI is actually changing the game, and Qlik we have the CEO on and the customer and the partner, and we're going to hear. Because when you get your data right, AI is magical. And I think this is what we're going to start to see all new experiences coming online and then new expectations following. So it should be a great next five years and just the beginning, scratching the surface. So I'm super excited. Looking forward to the segment.

>> Yes, I know. It's like we saved the best for last this time. Hello, Mike and Sarvant. Nice. Thank you for being here. Nice to be join us.

Mike Capone

>> Pleasure.

Sarvant Singh

>> Absolutely. Pleasure being here.

>> Here. It's so exciting. Mike, I got to start with you. This is your show. Thank you for having us here. What is it like to see 2100 people that you know?

Mike Capone

>> Yeah, well look, it's my show, but as you see, and it takes a village to put on a show like this, but we couldn't be more pleased. I mean, you've been sitting there all day, so you see the energy, the excitement. We're oversubscribed. We have more people come than we expected, which is always a great thing.

>> Oh, that's amazing. I hadn't heard that little nugget yet today.

Mike Capone

>> Yeah. We're really, really happy.

>> That's awesome.

Mike Capone

>> And look, the feedback from the event, the keynotes and all the product stuff has been over the top terrific. So day one. So far so good.

>> So far so good. Sarvant, exciting day for you too, I imagine. What's it like? How's your show been?

Sarvant Singh

>> Absolutely. I think it was great. I think we started the day with some great keynotes and great updates on the products. I was very, very excited about looking at Qlik's AI strategy and I think I got some very interesting nuggets out of that. So it has been great.

>> Here you are a proper partner and you're still taking notes in the keynote. I love to hear-

Sarvant Singh

>> Absolutely.

>> It's a learning conference. Mike, I want you to lay out and I'd love to get the commentary from the front lines as well. What is the update on Qlik? Give us a quick minute, set up the context. A lot of new things in the fold. We've heard other execs on your team. I've been in the business for decades. Finally, this is the moment. We have all the puzzle pieces from a customer. So it's feeling like things are here, clear line of sight on all the key puzzle pieces for the foundation for data. Lay out the Qlik story and then we'll get the reaction.

Mike Capone

>> Okay, great. Yeah, look, I'll be quick because you brought it. It's this moment, we've been waiting for it. I was like looking at your earrings, Rubik's cubes finally in place. The pieces of the puzzle. We've been getting ready for this moment for years. This was this gen AI moment where we finally crossed the chasm and said, okay, now we can actually make this stuff productive, useful. And so this is what we've been building and the products that we brought together, our data integration technology, now our data governance technology, analytics, and then automation, that chain has come to fruition. That's what we showed the audience today. And clearly they liked it. They really liked it

>> In hollers, I have to say, I think it was one of the more vivacious keynotes I have heard in terms of just the pure real enthusiasm that your community has for the product.

Mike Capone

>> And because they've been waiting. We talked about this being an inflection point. It's like there's always been this promise of AI, but also this fear of AI. It's got all this potential, but got to be careful. And what we showed today was how you can deploy AI productively and responsibly. And that's the key.

>> And the thing is give you guys props. We've been talking on the queue for the past year around how the data market's going to look like a cloud market. If you remember the cloud, born in the cloud days where SaaS kind of hit the inflection point there. It was a magical moment. Here with data, you're hearing things on stage. First conference I've ever been to heard... And I go to all the conferences, I've never heard data supply chain on the keynote mainstream. So you're seeing signs that data not just lineage, but that's a security conversation, software supply chain, where's it come from? So you're starting to see data look a lot like cloud native, born in the data world. We're now in a whole nother level spot. You are in this world of the customer. This is kind of like a platform shift. We heard that from the analyst as well. We're seeing the same thing. Again, I've never heard data supply chain. That tells me there's a developer market. That tells me there's tracking around security, role of data. I mean is the first time I've heard that.

Mike Capone

>> Yeah, yeah. Well look, I mean you usually hear it in the context of security and things like that. And certainly protecting your data is good, but protecting bad data is no fun either, right?

>> Absolutely.

Mike Capone

>> So, as we start to reach for the promise of AI, this data quality discussion has really come to the forefront. And in my keynote today, the one big message I left people is you got to do the work. You cannot just plug in AI and expect it to happen. You got to do the work. And you take companies like Penske who've done that work, first laid the foundation, and now they're achieving the value from it.

>> What's your reaction to all this?

>> Yeah. I want to actually just lay the land since we got the vision on Qlik. How is Penske been approaching AI in general and then dive in there? Yeah.

Sarvant Singh

>> Yeah, absolutely. So as I mentioned in the keynote, we have been operationalizing predictive and prescriptive AI for almost a decade, but generative AI is a different ball game altogether. I think it is going to fundamentally transform how we build applications, how we expose data or insights to the end users. I think in last 50 years, majority of the business applications which were built, they were built to receive inputs from keyboard and mouse. And when touchscreen and mobile came, we saw how we all had to adjust to swipe, left swipe. I think generative AI is going to cause a much, much bigger disruption, much bigger shift. So in my opinion, I think we will see a majority of the business applications getting disrupted at some point of time, because generative AI is fundamentally changing not just the interaction with information, but the generation of information itself. So we are looking at it with a lot of excitement.

>> How has it changed your data practice? Because again, we're hearing a lot about foundations setting that data foundation doing the work, which is definitely totally true there. But then now I got, okay, I did things the old way. Now there's a new way. So we started to see the industry clear lines of sight between old way and new way. What are some of those new ways and old way? What goes away and what's replaced? Can you share your view of... Because something's got to change and this is disruptive. It's a disruptive enabler.

Sarvant Singh

>> Yeah. So let me share what will not go away. I think as Mike mentioned this morning, garbage in, garbage out. It's not going away. So we have to be very, very careful about data foundations. We have to get data right, data quality right. So that's very, very important. Now, what is going to change is the way we have been thinking about business and technology capabilities. I think AI is not a once and done thing. In fact, building a model or coming up with a model is just the start. It's not the end. There's a lot of care and feeding which needs to happen when we are operationalizing AI. So I think the way we think about AI investments, that has to fundamentally shift. And as far as Qlik is concerned, I mean we have significant investments in Qlik today. So if we can generate insights using gen AI capabilities right then and there, I think that is going to expedite AI value delivery for my business partners. And that is why I'm so excited about Qlik's AI strategy.

>> What kind of solutions do you... And I think don't want to misquote it, so I'm hoping you can repeat it. But I heard an interesting stat about keeping trucks that were going to break down off the road so that you didn't bump into that problem. Can you share that nugget with us?

Sarvant Singh

>> Yes, absolutely. So think about it like this. Before we had an AI solution, a truck is going on on the road and a failure may occur. And in the past there was just no way for us to proactively do something about it. Now with AI, we can predict when a truck is going to fail. And in number of cases we can actually intervene. We can call the driver or reach out to the customer and say, well, here is our next location, it's 50 miles away. How about you take a break while we take a look at the truck? And I'll quote an instance that in the day we went live at 6:00 AM there was a fault code which was generated by 8:30 AM. That truck was back on the road because we were able to quickly respond to that failure proactively. And last year alone, we prevented 90,000 trucks from on the side of the road.

>> 90,000

Sarvant Singh

>> 90,000. That's a big number.

Mike Capone

>> 90,000. Yeah.

>> 90,000 physical hardware edge devices essentially, is how we can look at them, loosely. Yeah.

Sarvant Singh

>> Absolutely.

>> Cost savings alone must be huge.

>> Exactly.

Sarvant Singh

>> Absolutely. Cost savings are significant. But on top of that, what's most important is the customer experience. Because we are avoiding costly and time-consuming downtimes for our customers.

>> Yeah. Delay .

Mike Capone

>> And customer satisfaction.

Sarvant Singh

>> Absolutely.

>> The killer app is time with AI.

Sarvant Singh

>> Absolutely.

>> That's our time.

>> Right. Well, and we talk about the hype's obviously having its moment right now, but 90,000 vehicles not breaking down is a real thing. That's not hype. That's dollars in the pocket and smiles on faces. Speaking of smiles, 93% customer retention for you guys.

Mike Capone

>> Yeah, that's right.

>> That is wild.

Mike Capone

>> It is wild. It is wild. It's a statistic we're super proud of.

>> I can't believe that. I mean I can believe it because I've met a lot of them today. It's been incredibly impressive, you included. But I think that really speaks to the passion and the value you're driving there. What is it like for you when you hear stories like Penske's experience?

Mike Capone

>> Yeah, look, I'm super lucky because I'm one of those people who loves coming to work every day. And it is the energy you get from customers and it's always been true. This is not something new to Mike Capone, the CEO of Qlik, who's been here six years. One of the reasons I joined Qlik was I talked to customers before I even joined. And what they all said is, "You guys have a great product." And again, this philosophy of partnerships, this philosophy of we go with you on your journey, we don't impose things on you. Super important. And that's what we try to preach. And when I hear customers stand up, not just say it, but stand up in front of thousands of people and say it proudly, what better could a CEO ask for? And as I mentioned, I was a CIO, I was in Sarvant's chair at one point and I knew the kind of companies I wanted to do business with, and they had a certain DNA and a certain culture, certain approach. And so when I joined and with my leadership team, we said, "Let's be that company. Let's be the company that customers want to do business with, not a necessary evil."

>> Talk about the roles inside your customer base, and if you can get some data points as well, share that? The data market, if you connect the dots, this continues, generative AI will hit, we know it's coming, everyone's preparing for it. Changes your game, Qlik's game from analytics to just all apps. I mean it's now everything. So the data's in all the applications, data is the product and there's software wrapped around it.

Mike Capone

>> That's right.

>> So if you look at it that way, Answers is just an app of a data foundation with middleware, AKA data and software. So if data is now programmable, data supply chain's got this cloud field too, I'm going in this direction. Data developers are coming. What's your rich reaction to that?

Mike Capone

>> Sarvant?

>> He's smiling.

Sarvant Singh

>> Yes. So data and we are going to see more and more digital products which are primarily data driven. That's exactly where it is headed. To your comment about Qlik Answers, where Gen AI is being made available as part of this platform, I think what we will see is embedded AI. I think we will see expediting of that trend because it's no longer a one app, which you need to go to do something. I think it is going to become pervasive, and it'll become part and parcel of each and every business application at some point of time.

>> And you mentioned earlier pre Gen AI, you already had the data practice. You're doing predictive prescriptive machine learning. It's unsupervised, supervised, that's hardcore. Okay, get that.

Sarvant Singh

>> Absolutely.

>> Now if you take that forward, what gets easier about leveraging that capability in gen AI? Is it the embedded AI? Is it the agents that are going to come out of it? If you help 90,000 trucks get better, what are you going to do for employees?

Sarvant Singh

>> I think it's all of that. So the first benefit we are going to see is to a knowledge worker. For example, the search is fundamentally changing because of generative AI. So we will have intelligent search. In the past, if you remember, even if you'll go to Google, you will search something, you'll get 10 pages and then you had to figure out what's meaningful, what's not. Now generative AI is going to change that. Now, what you will get is some very actionable information and you can act on it immediately. So I think Search is one area which is getting disrupted and it will enhance employee productivity. I think call centers are another area where we will see significant productivity improvements and efficiency improvements because of Gen AI. Because the response is much more human-like compared to what we used to see with either smaller models or traditional AI.

>> Yeah, I think we've seen a lot of evolution in that space. Speaking of, how are you continuing... You've taken notes in the keynote, you mentioned you're learning today.

Sarvant Singh

>> I was trying.

>> Got a lot going on at this point.

>> He's in the keynote, he's taking notes and he's in the keynote-

>> No, no. This is low-key, like on the cam .

>> He's the master, the teacher and the student at one time.

>> No, but that's great. With that in mind, how are you hoping to continue this momentum and the success that you've already seen implementing this at scale in the next year or two?

Sarvant Singh

>> Absolutely. So earlier we were more use case focused, but now our focus is on having a groundswell on AI. So self-service AI is a significant part of our data strategy. So we talked about 170 Qlik developers. Our focus now is on converting our analysts into AI developers. And that's where I think the significant positive impact is going to come from because AI is going to become pervasive as a result of that. So as I said, self-service AI is the next big opportunity for us. And I do believe that Qlik has a role there because majority of our data is already sitting in a Qlik engine.

>> I got to ask you guys something, if you don't mind commenting. As you said, we see the apps are already out there. The workload. Take your normal application. It's end-to-end. You guys talk about end-to-end solutions all the time, but now you've got an end-to-end application that's pre gen AI, it's scope, you know what it does has a utility and value. Now it's to make that gen AI enabled. The data model changes a bit. So lots going on under the covers. So now the workflow's known end to end with data that's proprietary, that is the IP of a company. So we're seeing the workflows and data are the new IP. One, do you believe that? And two, how do you retrofit that workload? It could be a banking app, it could be your app, it could be a new app?

Sarvant Singh

>> Yeah, so my impression and look, this field is still evolving, so there is a lot to be seen, but I don't think the business process is going to change a little bit. I think there is a potential that it may get disrupted or significantly transformed. And I think we still have to figure out that if using gen AI we can fundamentally transform a business process, how will we deliver the new experience? How will the end user experience going to shift? How will the customer experience going to shift? So the way application gets designed, I think that process itself may get transformed. And I think that's why-

>> And there'll be data hungry, right? So I mean AI is going to be very compute intensive. Obviously GPS XPS are coming out. Mike, do you agree with this, because the process might not change but how it's constructed might and the amount of data involved and how data interacts maybe changes?

Mike Capone

>> That's absolutely true. I mean it's tough to paint in broad generalizations. I think that's the danger of the whole gen AI craze, is people are making extreme statements. I think it's very case by case, very situational. I think we're going to be able to very quickly layer gen AI into existing data pipelines and applications. We're doing that today with Qlik's sense applications, or analytics application and get value right away. I think there are other cases where we're going to need to rebuild the data pipelines with AI in mind and figure out how we get our data ready and develop actually all new platforms and use cases and things like that. So I think it's going to be quite situational. The compute thing is really interesting. That does keep me awake at night. Like how much the power consumption. And we're going to have to as an industry find ways to optimize. And one thing I coach our customers on is the data is actually showing or the science is showing that more isn't always better. Sometimes everyone's like more data is better, more data is better AI, it's actually not true. More isn't always better. Sometimes better is better.

>> Better is better. Yeah, yeah, yeah.

Mike Capone

>> Having smaller better data sets training your algorithms can be better. And so I encourage customers just to think about that before they just start throwing everything that could sink into their models.

>> And you guys have been also talking about trust. I think one of the things that we learned today from your team is, risk management and product management are now disciplines that are at the nexus of all the situational decisions. Do you go specialty models, stay small. Do you want to take existing data and experiment, then double down? So these classic risk management decision... You're shaking your head, you see the same thing?

Sarvant Singh

>> Yeah, absolutely. I think what you were just describing is where the vision of AI mesh comes in. Where you are looking at different aspects of a decision and there are different models feeding into that. So definitely that's the next frontier.

Mike Capone

>> And you saw, I think you've even had them on earlier, our AI council. So one of the things that we've done is put together a council of extortion outside of corporate America, outside of global corporate work. And they're from academia, they're from government, geopolitical advisors. And it really is about getting that foundation. Because in order to get this right, you got to start from the beginning. And as you start building these things, build privacy, security ethics into your process and your modeling.

>> You're talking about the AI council segments, go check it out. It's worth watching for the folks watching here. One thing that jumped out at me, Dr. Michael Bornstein talked about was he does the deep mind. He also worked at Twitter Graph. He's big time neural network guys. So we were kind of riffing. We do kind of outland just things on the queue. I just say that preface, this is a crazy statement. But we've been speculating that there's an operating model emerging, not yet known. If you take neural networks and knowledge graphs together, what runs those? So the question that we're asking is what does that look like? And is there yet another operating model behind what we're scratching the surface now? If the graphs are proliferating, if there now data has to be highly available, is it highly available or highly high availability?

Mike Capone

>> You get that question?

Sarvant Singh

>> Can you repeat that question?

>> So in the storage world, highly available and highly availability have two definitional meetings. One's low latency at the edge. So data latency is an issue. You've got all kinds of synthetic data. So the speculation is what runs this? What operating system runs this?

Mike Capone

>> It might be one that isn't invented yet. Look, I'm not going to try to get even on the same plane as Michael in terms of these conversations. He is a genius and that's why we have him advising us.

>> I can imagine we could go down a big rabbit hole. We go down a very-

>> Let's not go there.

>> A very serious rabbit hole.

>> We'll save that for later.

Mike Capone

>> Par of being a good CEO is knowing when you start to getting over your head and you just change the subject.

>> Well, actually where I was going is kind along those lines. I want to ask you both this question, but Mike, I'm going to start with you. Everything's moving real fast. Everybody wants to be building things. I'm sure everyone wants solutions. How are you determining your roadmap? How are you deciding when to make these... I mean obviously made some very strategic and smart acquisitions earlier this year that have led us to where we are now, but I'm curious, what are those conversations like? How are you navigating it? What's your advice to other leaders doing that?

Mike Capone

>> Look, the number one asset we have is our customers here. We've got hundreds of one-on-ones with our customers right now. So while you see everybody checking out our solutions, we're secretly also interrogating all of our customers about what they want and where they see the world going. We have our executive advisory board, we have a set of premier customers that are actually really, really deep in their fields and they're advising us as well. We have the AI council, et cetera. Then we got great product managers who are constantly doing research. Things are moving fast. But the trick is actually not to be fooled by the flavor of the moment. Because if we just started building stuff a year ago when this Gen AI craze was happening, we'd just be plugging LMS and ChatGPT and everything and it wouldn't have created a lot of value, but we would've satisfied the hype. And so what we try to do is be agile and quick, but thoughtful.

>> And real problems.

Mike Capone

>> And real problems.

>> I like that. Sarvant, is it same for you? Yeah, go for it.

Sarvant Singh

>> So if I could, even though we decided not to go there, but I would like to address your question.

>> Yes.

>> We'll just buy you .

>> Let's go.

Sarvant Singh

>> So this is what happened. When AI capability first became available, people were like, "Oh, we need a data lake so that I can bring all the data together and then run AI on it." So there was a race to the right. But because of edge computing now there is a race to the left. And a lot of people are confused, "Am I racing to the right? Am I racing to the left?" And I think that's where the questions around latency, around decision making are going to be very, very important. And this decision making that, "Am I racing to the right or am I raising to the left?" Needs to be done on a case by case basis, and latency or decision making will play a very, very big role.

>> And what came out of the AI Council on that point was is that people don't yet know what to do if that scenario happens.

Sarvant Singh

>> That is correct.

>> They're comfortable with centralization, because they know what it's like, that an interrogated data set. That's an old school concept in data analytics. But okay, federated data, what do I do? I've never had the chance to do that. So these new generational problems are opportunities. So again, this is what we're trying to understand. And again, back to what the council said was, "Ignore the hype and focus on real problems."

Mike Capone

>> Focus on real problems.

>> You could have a North Star vision, but don't get over your skis. So that's my word. They didn't say that. But the same. Don't go too far.

>> Well, focus on getting trucks not to break down on the road.

Mike Capone

>> Yeah, that's right. That's right.

>> No, I think that's great. All right.

>> So talk about the relationship to you guys customer. What about Qlik? Talk about how you guys are innovating together? Because obviously you guys have a great partnership. What's the key to success? Is it the foundation? We'll talk about your scenario?

Sarvant Singh

>> I mean, key to success has been that anytime we have a challenge or we need a strategic conversation, Qlik's team is always available. So I think at the end of the day, it's all about just having a great relationship, having this human to human interaction. And I think the two organizations have been just very, very good at that. So I think that's what I would say. Mike?

Mike Capone

>> And it's customers willingness to speak up and give us feedback. It's never personal, it's always just business. And you got to make time for those conversations. And you also have to... When customers talk to you about things, they do expect you to listen and then respond and do something. So you have to keep credible.

>> Yeah. trust.

Mike Capone

>> Yeah, a lot of companies will say, "Oh yeah, we listen to our customers." But they really don't. It's like a one-way thing. But we try to have a governance model where we actually have this feedback loop where we're delivering. You don't get everything you want, but you do get a lot. And then when we can't do it, we just say, "Hey, we can't do it or we're going to do it this way." But it's very symbiotic and quite frankly, it's fine.

>> Why you got high retention rates. So what about the best practices? What advice would you give someone who wants to lean into the AI wave the right way, the generative AI wave? What's the best practice? What's the pro-tip from your group?

Sarvant Singh

>> Yeah, so a couple of things. One, focus on data. We have to get the data right. Because garbage and garbage out, as Mike said this morning, is absolutely true. So focus on data. That's number one. Second, taking a federated approach is definitely the best way. Because if you will centralize very heavily, you'll alienate some people. And if it is heavily distributed, then we are not going to get the economies of the scale. We are not leveraging the architecture. So I think it's finding that sweet spot, and I think the right answer is going to vary from company to company. I don't think there is a magic bullet, to be very honest. So depending upon the type of business you are in, the kind of customer pressures you deal with, I think the business capabilities need to drive the technical capabilities, the technical architecture around AI implementation. So that is what I would recommend, that keep your eye on the vision, the big goal, but how will exactly we execute? And I think that's very, very important.

>> That is important. All right. Last question for you, very insightful gentlemen. Because we'll use this for our promo for the next Qlik Connect next year. What do you hope to be able to say when you sit down with us in 2025 that you can't quite say yet today? I'll start with you, Mike.

Mike Capone

>> Look, I think we're at this inflection point. We talked about this earlier about, okay, you got to do the work. And what I'm going to say next year, I'm going to have Sarvant back on stage for sure. And a few other customers talking about, they heard us when we were on stage saying, you know what? We need to focus on the data quality, data work, and now we got the outcome. And that's going to be we did as well as these guys already did with their 90,000 trucks and preventative maintenance. And Sarvant is going to have a better story about what he did with his business.

>> And what do you think that story's going to be in a year?

Sarvant Singh

>> I think what Mike said makes perfect sense from my business's perspective. Because, if we will have some use cases where we have gen AI implemented within Qlik, and if we can productionize those, I think that's success. I think that would be a very good story.

>> I think it would be a good story, and we look forward to hopefully having it right here back on theCUBE and in our keynote next year. So Mike, thank you both so much for being here.

Sarvant Singh

>> Absolutely. Thank you.

>> This is great, and thanks for having such a wonderful community who has embraced this and so many fabulous guests just like you. It's been awesome. John, thanks for spending the day here with me in Orlando.

>> Great. Great team.

>> It really has been great. And thank all of you fabulous producers back there that I'm looking at for doing such a fantastic job. And everyone on the Qlik team who showed up today, we've had just a power pack lineup. I really do feel smarter than when I woke up this morning. And thank all of you, our fabulous CUBE community for tuning into Qlik Connect here in Orlando, Florida. My name's Savannah Peterson, you're watching theCUBE, the leading source for enterprise tech news.

Mike Capone | Qlik & Sarvant Singh | Penske

search

>> Good evening, Qlik fans and welcome back to Orlando, Florida. We are here for our last segment of Qlik Connect. It has been an awesome day. I feel smarter than when I woke up today. My name's Savannah Peterson. Delighted to be joined by John Furrier. John, what a show we've had.

>> It's been a great show. It's very action packed, but don't feel it. It's really well-designed show. Those stories popping on all cylinders. Generative AI is actually changing the game, and Qlik we have the CEO on and the customer and the partner, and we're going to hear. Because when you get your data right, AI is magical. And I think this is what we're going to start to see all new experiences coming online and then new expectations following. So it should be a great next five years and just the beginning, scratching the surface. So I'm super excited. Looking forward to the segment.

>> Yes, I know. It's like we saved the best for last this time. Hello, Mike and Sarvant. Nice. Thank you for being here. Nice to be join us.

Mike Capone

>> Pleasure.

Sarvant Singh

>> Absolutely. Pleasure being here.

>> Here. It's so exciting. Mike, I got to start with you. This is your show. Thank you for having us here. What is it like to see 2100 people that you know?

Mike Capone

>> Yeah, well look, it's my show, but as you see, and it takes a village to put on a show like this, but we couldn't be more pleased. I mean, you've been sitting there all day, so you see the energy, the excitement. We're oversubscribed. We have more people come than we expected, which is always a great thing.

>> Oh, that's amazing. I hadn't heard that little nugget yet today.

Mike Capone

>> Yeah. We're really, really happy.

>> That's awesome.

Mike Capone

>> And look, the feedback from the event, the keynotes and all the product stuff has been over the top terrific. So day one. So far so good.

>> So far so good. Sarvant, exciting day for you too, I imagine. What's it like? How's your show been?

Sarvant Singh

>> Absolutely. I think it was great. I think we started the day with some great keynotes and great updates on the products. I was very, very excited about looking at Qlik's AI strategy and I think I got some very interesting nuggets out of that. So it has been great.

>> Here you are a proper partner and you're still taking notes in the keynote. I love to hear-

Sarvant Singh

>> Absolutely.

>> It's a learning conference. Mike, I want you to lay out and I'd love to get the commentary from the front lines as well. What is the update on Qlik? Give us a quick minute, set up the context. A lot of new things in the fold. We've heard other execs on your team. I've been in the business for decades. Finally, this is the moment. We have all the puzzle pieces from a customer. So it's feeling like things are here, clear line of sight on all the key puzzle pieces for the foundation for data. Lay out the Qlik story and then we'll get the reaction.

Mike Capone

>> Okay, great. Yeah, look, I'll be quick because you brought it. It's this moment, we've been waiting for it. I was like looking at your earrings, Rubik's cubes finally in place. The pieces of the puzzle. We've been getting ready for this moment for years. This was this gen AI moment where we finally crossed the chasm and said, okay, now we can actually make this stuff productive, useful. And so this is what we've been building and the products that we brought together, our data integration technology, now our data governance technology, analytics, and then automation, that chain has come to fruition. That's what we showed the audience today. And clearly they liked it. They really liked it

>> In hollers, I have to say, I think it was one of the more vivacious keynotes I have heard in terms of just the pure real enthusiasm that your community has for the product.

Mike Capone

>> And because they've been waiting. We talked about this being an inflection point. It's like there's always been this promise of AI, but also this fear of AI. It's got all this potential, but got to be careful. And what we showed today was how you can deploy AI productively and responsibly. And that's the key.

>> And the thing is give you guys props. We've been talking on the queue for the past year around how the data market's going to look like a cloud market. If you remember the cloud, born in the cloud days where SaaS kind of hit the inflection point there. It was a magical moment. Here with data, you're hearing things on stage. First conference I've ever been to heard... And I go to all the conferences, I've never heard data supply chain on the keynote mainstream. So you're seeing signs that data not just lineage, but that's a security conversation, software supply chain, where's it come from? So you're starting to see data look a lot like cloud native, born in the data world. We're now in a whole nother level spot. You are in this world of the customer. This is kind of like a platform shift. We heard that from the analyst as well. We're seeing the same thing. Again, I've never heard data supply chain. That tells me there's a developer market. That tells me there's tracking around security, role of data. I mean is the first time I've heard that.

Mike Capone

>> Yeah, yeah. Well look, I mean you usually hear it in the context of security and things like that. And certainly protecting your data is good, but protecting bad data is no fun either, right?

>> Absolutely.

Mike Capone

>> So, as we start to reach for the promise of AI, this data quality discussion has really come to the forefront. And in my keynote today, the one big message I left people is you got to do the work. You cannot just plug in AI and expect it to happen. You got to do the work. And you take companies like Penske who've done that work, first laid the foundation, and now they're achieving the value from it.

>> What's your reaction to all this?

>> Yeah. I want to actually just lay the land since we got the vision on Qlik. How is Penske been approaching AI in general and then dive in there? Yeah.

Sarvant Singh

>> Yeah, absolutely. So as I mentioned in the keynote, we have been operationalizing predictive and prescriptive AI for almost a decade, but generative AI is a different ball game altogether. I think it is going to fundamentally transform how we build applications, how we expose data or insights to the end users. I think in last 50 years, majority of the business applications which were built, they were built to receive inputs from keyboard and mouse. And when touchscreen and mobile came, we saw how we all had to adjust to swipe, left swipe. I think generative AI is going to cause a much, much bigger disruption, much bigger shift. So in my opinion, I think we will see a majority of the business applications getting disrupted at some point of time, because generative AI is fundamentally changing not just the interaction with information, but the generation of information itself. So we are looking at it with a lot of excitement.

>> How has it changed your data practice? Because again, we're hearing a lot about foundations setting that data foundation doing the work, which is definitely totally true there. But then now I got, okay, I did things the old way. Now there's a new way. So we started to see the industry clear lines of sight between old way and new way. What are some of those new ways and old way? What goes away and what's replaced? Can you share your view of... Because something's got to change and this is disruptive. It's a disruptive enabler.

Sarvant Singh

>> Yeah. So let me share what will not go away. I think as Mike mentioned this morning, garbage in, garbage out. It's not going away. So we have to be very, very careful about data foundations. We have to get data right, data quality right. So that's very, very important. Now, what is going to change is the way we have been thinking about business and technology capabilities. I think AI is not a once and done thing. In fact, building a model or coming up with a model is just the start. It's not the end. There's a lot of care and feeding which needs to happen when we are operationalizing AI. So I think the way we think about AI investments, that has to fundamentally shift. And as far as Qlik is concerned, I mean we have significant investments in Qlik today. So if we can generate insights using gen AI capabilities right then and there, I think that is going to expedite AI value delivery for my business partners. And that is why I'm so excited about Qlik's AI strategy.

>> What kind of solutions do you... And I think don't want to misquote it, so I'm hoping you can repeat it. But I heard an interesting stat about keeping trucks that were going to break down off the road so that you didn't bump into that problem. Can you share that nugget with us?

Sarvant Singh

>> Yes, absolutely. So think about it like this. Before we had an AI solution, a truck is going on on the road and a failure may occur. And in the past there was just no way for us to proactively do something about it. Now with AI, we can predict when a truck is going to fail. And in number of cases we can actually intervene. We can call the driver or reach out to the customer and say, well, here is our next location, it's 50 miles away. How about you take a break while we take a look at the truck? And I'll quote an instance that in the day we went live at 6:00 AM there was a fault code which was generated by 8:30 AM. That truck was back on the road because we were able to quickly respond to that failure proactively. And last year alone, we prevented 90,000 trucks from on the side of the road.

>> 90,000

Sarvant Singh

>> 90,000. That's a big number.

Mike Capone

>> 90,000. Yeah.

>> 90,000 physical hardware edge devices essentially, is how we can look at them, loosely. Yeah.

Sarvant Singh

>> Absolutely.

>> Cost savings alone must be huge.

>> Exactly.

Sarvant Singh

>> Absolutely. Cost savings are significant. But on top of that, what's most important is the customer experience. Because we are avoiding costly and time-consuming downtimes for our customers.

>> Yeah. Delay .

Mike Capone

>> And customer satisfaction.

Sarvant Singh

>> Absolutely.

>> The killer app is time with AI.

Sarvant Singh

>> Absolutely.

>> That's our time.

>> Right. Well, and we talk about the hype's obviously having its moment right now, but 90,000 vehicles not breaking down is a real thing. That's not hype. That's dollars in the pocket and smiles on faces. Speaking of smiles, 93% customer retention for you guys.

Mike Capone

>> Yeah, that's right.

>> That is wild.

Mike Capone

>> It is wild. It is wild. It's a statistic we're super proud of.

>> I can't believe that. I mean I can believe it because I've met a lot of them today. It's been incredibly impressive, you included. But I think that really speaks to the passion and the value you're driving there. What is it like for you when you hear stories like Penske's experience?

Mike Capone

>> Yeah, look, I'm super lucky because I'm one of those people who loves coming to work every day. And it is the energy you get from customers and it's always been true. This is not something new to Mike Capone, the CEO of Qlik, who's been here six years. One of the reasons I joined Qlik was I talked to customers before I even joined. And what they all said is, "You guys have a great product." And again, this philosophy of partnerships, this philosophy of we go with you on your journey, we don't impose things on you. Super important. And that's what we try to preach. And when I hear customers stand up, not just say it, but stand up in front of thousands of people and say it proudly, what better could a CEO ask for? And as I mentioned, I was a CIO, I was in Sarvant's chair at one point and I knew the kind of companies I wanted to do business with, and they had a certain DNA and a certain culture, certain approach. And so when I joined and with my leadership team, we said, "Let's be that company. Let's be the company that customers want to do business with, not a necessary evil."

>> Talk about the roles inside your customer base, and if you can get some data points as well, share that? The data market, if you connect the dots, this continues, generative AI will hit, we know it's coming, everyone's preparing for it. Changes your game, Qlik's game from analytics to just all apps. I mean it's now everything. So the data's in all the applications, data is the product and there's software wrapped around it.

Mike Capone

>> That's right.

>> So if you look at it that way, Answers is just an app of a data foundation with middleware, AKA data and software. So if data is now programmable, data supply chain's got this cloud field too, I'm going in this direction. Data developers are coming. What's your rich reaction to that?

Mike Capone

>> Sarvant?

>> He's smiling.

Sarvant Singh

>> Yes. So data and we are going to see more and more digital products which are primarily data driven. That's exactly where it is headed. To your comment about Qlik Answers, where Gen AI is being made available as part of this platform, I think what we will see is embedded AI. I think we will see expediting of that trend because it's no longer a one app, which you need to go to do something. I think it is going to become pervasive, and it'll become part and parcel of each and every business application at some point of time.

>> And you mentioned earlier pre Gen AI, you already had the data practice. You're doing predictive prescriptive machine learning. It's unsupervised, supervised, that's hardcore. Okay, get that.

Sarvant Singh

>> Absolutely.

>> Now if you take that forward, what gets easier about leveraging that capability in gen AI? Is it the embedded AI? Is it the agents that are going to come out of it? If you help 90,000 trucks get better, what are you going to do for employees?

Sarvant Singh

>> I think it's all of that. So the first benefit we are going to see is to a knowledge worker. For example, the search is fundamentally changing because of generative AI. So we will have intelligent search. In the past, if you remember, even if you'll go to Google, you will search something, you'll get 10 pages and then you had to figure out what's meaningful, what's not. Now generative AI is going to change that. Now, what you will get is some very actionable information and you can act on it immediately. So I think Search is one area which is getting disrupted and it will enhance employee productivity. I think call centers are another area where we will see significant productivity improvements and efficiency improvements because of Gen AI. Because the response is much more human-like compared to what we used to see with either smaller models or traditional AI.

>> Yeah, I think we've seen a lot of evolution in that space. Speaking of, how are you continuing... You've taken notes in the keynote, you mentioned you're learning today.

Sarvant Singh

>> I was trying.

>> Got a lot going on at this point.

>> He's in the keynote, he's taking notes and he's in the keynote-

>> No, no. This is low-key, like on the cam .

>> He's the master, the teacher and the student at one time.

>> No, but that's great. With that in mind, how are you hoping to continue this momentum and the success that you've already seen implementing this at scale in the next year or two?

Sarvant Singh

>> Absolutely. So earlier we were more use case focused, but now our focus is on having a groundswell on AI. So self-service AI is a significant part of our data strategy. So we talked about 170 Qlik developers. Our focus now is on converting our analysts into AI developers. And that's where I think the significant positive impact is going to come from because AI is going to become pervasive as a result of that. So as I said, self-service AI is the next big opportunity for us. And I do believe that Qlik has a role there because majority of our data is already sitting in a Qlik engine.

>> I got to ask you guys something, if you don't mind commenting. As you said, we see the apps are already out there. The workload. Take your normal application. It's end-to-end. You guys talk about end-to-end solutions all the time, but now you've got an end-to-end application that's pre gen AI, it's scope, you know what it does has a utility and value. Now it's to make that gen AI enabled. The data model changes a bit. So lots going on under the covers. So now the workflow's known end to end with data that's proprietary, that is the IP of a company. So we're seeing the workflows and data are the new IP. One, do you believe that? And two, how do you retrofit that workload? It could be a banking app, it could be your app, it could be a new app?

Sarvant Singh

>> Yeah, so my impression and look, this field is still evolving, so there is a lot to be seen, but I don't think the business process is going to change a little bit. I think there is a potential that it may get disrupted or significantly transformed. And I think we still have to figure out that if using gen AI we can fundamentally transform a business process, how will we deliver the new experience? How will the end user experience going to shift? How will the customer experience going to shift? So the way application gets designed, I think that process itself may get transformed. And I think that's why-

>> And there'll be data hungry, right? So I mean AI is going to be very compute intensive. Obviously GPS XPS are coming out. Mike, do you agree with this, because the process might not change but how it's constructed might and the amount of data involved and how data interacts maybe changes?

Mike Capone

>> That's absolutely true. I mean it's tough to paint in broad generalizations. I think that's the danger of the whole gen AI craze, is people are making extreme statements. I think it's very case by case, very situational. I think we're going to be able to very quickly layer gen AI into existing data pipelines and applications. We're doing that today with Qlik's sense applications, or analytics application and get value right away. I think there are other cases where we're going to need to rebuild the data pipelines with AI in mind and figure out how we get our data ready and develop actually all new platforms and use cases and things like that. So I think it's going to be quite situational. The compute thing is really interesting. That does keep me awake at night. Like how much the power consumption. And we're going to have to as an industry find ways to optimize. And one thing I coach our customers on is the data is actually showing or the science is showing that more isn't always better. Sometimes everyone's like more data is better, more data is better AI, it's actually not true. More isn't always better. Sometimes better is better.

>> Better is better. Yeah, yeah, yeah.

Mike Capone

>> Having smaller better data sets training your algorithms can be better. And so I encourage customers just to think about that before they just start throwing everything that could sink into their models.

>> And you guys have been also talking about trust. I think one of the things that we learned today from your team is, risk management and product management are now disciplines that are at the nexus of all the situational decisions. Do you go specialty models, stay small. Do you want to take existing data and experiment, then double down? So these classic risk management decision... You're shaking your head, you see the same thing?

Sarvant Singh

>> Yeah, absolutely. I think what you were just describing is where the vision of AI mesh comes in. Where you are looking at different aspects of a decision and there are different models feeding into that. So definitely that's the next frontier.

Mike Capone

>> And you saw, I think you've even had them on earlier, our AI council. So one of the things that we've done is put together a council of extortion outside of corporate America, outside of global corporate work. And they're from academia, they're from government, geopolitical advisors. And it really is about getting that foundation. Because in order to get this right, you got to start from the beginning. And as you start building these things, build privacy, security ethics into your process and your modeling.

>> You're talking about the AI council segments, go check it out. It's worth watching for the folks watching here. One thing that jumped out at me, Dr. Michael Bornstein talked about was he does the deep mind. He also worked at Twitter Graph. He's big time neural network guys. So we were kind of riffing. We do kind of outland just things on the queue. I just say that preface, this is a crazy statement. But we've been speculating that there's an operating model emerging, not yet known. If you take neural networks and knowledge graphs together, what runs those? So the question that we're asking is what does that look like? And is there yet another operating model behind what we're scratching the surface now? If the graphs are proliferating, if there now data has to be highly available, is it highly available or highly high availability?

Mike Capone

>> You get that question?

Sarvant Singh

>> Can you repeat that question?

>> So in the storage world, highly available and highly availability have two definitional meetings. One's low latency at the edge. So data latency is an issue. You've got all kinds of synthetic data. So the speculation is what runs this? What operating system runs this?

Mike Capone

>> It might be one that isn't invented yet. Look, I'm not going to try to get even on the same plane as Michael in terms of these conversations. He is a genius and that's why we have him advising us.

>> I can imagine we could go down a big rabbit hole. We go down a very-

>> Let's not go there.

>> A very serious rabbit hole.

>> We'll save that for later.

Mike Capone

>> Par of being a good CEO is knowing when you start to getting over your head and you just change the subject.

>> Well, actually where I was going is kind along those lines. I want to ask you both this question, but Mike, I'm going to start with you. Everything's moving real fast. Everybody wants to be building things. I'm sure everyone wants solutions. How are you determining your roadmap? How are you deciding when to make these... I mean obviously made some very strategic and smart acquisitions earlier this year that have led us to where we are now, but I'm curious, what are those conversations like? How are you navigating it? What's your advice to other leaders doing that?

Mike Capone

>> Look, the number one asset we have is our customers here. We've got hundreds of one-on-ones with our customers right now. So while you see everybody checking out our solutions, we're secretly also interrogating all of our customers about what they want and where they see the world going. We have our executive advisory board, we have a set of premier customers that are actually really, really deep in their fields and they're advising us as well. We have the AI council, et cetera. Then we got great product managers who are constantly doing research. Things are moving fast. But the trick is actually not to be fooled by the flavor of the moment. Because if we just started building stuff a year ago when this Gen AI craze was happening, we'd just be plugging LMS and ChatGPT and everything and it wouldn't have created a lot of value, but we would've satisfied the hype. And so what we try to do is be agile and quick, but thoughtful.

>> And real problems.

Mike Capone

>> And real problems.

>> I like that. Sarvant, is it same for you? Yeah, go for it.

Sarvant Singh

>> So if I could, even though we decided not to go there, but I would like to address your question.

>> Yes.

>> We'll just buy you .

>> Let's go.

Sarvant Singh

>> So this is what happened. When AI capability first became available, people were like, "Oh, we need a data lake so that I can bring all the data together and then run AI on it." So there was a race to the right. But because of edge computing now there is a race to the left. And a lot of people are confused, "Am I racing to the right? Am I racing to the left?" And I think that's where the questions around latency, around decision making are going to be very, very important. And this decision making that, "Am I racing to the right or am I raising to the left?" Needs to be done on a case by case basis, and latency or decision making will play a very, very big role.

>> And what came out of the AI Council on that point was is that people don't yet know what to do if that scenario happens.

Sarvant Singh

>> That is correct.

>> They're comfortable with centralization, because they know what it's like, that an interrogated data set. That's an old school concept in data analytics. But okay, federated data, what do I do? I've never had the chance to do that. So these new generational problems are opportunities. So again, this is what we're trying to understand. And again, back to what the council said was, "Ignore the hype and focus on real problems."

Mike Capone

>> Focus on real problems.

>> You could have a North Star vision, but don't get over your skis. So that's my word. They didn't say that. But the same. Don't go too far.

>> Well, focus on getting trucks not to break down on the road.

Mike Capone

>> Yeah, that's right. That's right.

>> No, I think that's great. All right.

>> So talk about the relationship to you guys customer. What about Qlik? Talk about how you guys are innovating together? Because obviously you guys have a great partnership. What's the key to success? Is it the foundation? We'll talk about your scenario?

Sarvant Singh

>> I mean, key to success has been that anytime we have a challenge or we need a strategic conversation, Qlik's team is always available. So I think at the end of the day, it's all about just having a great relationship, having this human to human interaction. And I think the two organizations have been just very, very good at that. So I think that's what I would say. Mike?

Mike Capone

>> And it's customers willingness to speak up and give us feedback. It's never personal, it's always just business. And you got to make time for those conversations. And you also have to... When customers talk to you about things, they do expect you to listen and then respond and do something. So you have to keep credible.

>> Yeah. trust.

Mike Capone

>> Yeah, a lot of companies will say, "Oh yeah, we listen to our customers." But they really don't. It's like a one-way thing. But we try to have a governance model where we actually have this feedback loop where we're delivering. You don't get everything you want, but you do get a lot. And then when we can't do it, we just say, "Hey, we can't do it or we're going to do it this way." But it's very symbiotic and quite frankly, it's fine.

>> Why you got high retention rates. So what about the best practices? What advice would you give someone who wants to lean into the AI wave the right way, the generative AI wave? What's the best practice? What's the pro-tip from your group?

Sarvant Singh

>> Yeah, so a couple of things. One, focus on data. We have to get the data right. Because garbage and garbage out, as Mike said this morning, is absolutely true. So focus on data. That's number one. Second, taking a federated approach is definitely the best way. Because if you will centralize very heavily, you'll alienate some people. And if it is heavily distributed, then we are not going to get the economies of the scale. We are not leveraging the architecture. So I think it's finding that sweet spot, and I think the right answer is going to vary from company to company. I don't think there is a magic bullet, to be very honest. So depending upon the type of business you are in, the kind of customer pressures you deal with, I think the business capabilities need to drive the technical capabilities, the technical architecture around AI implementation. So that is what I would recommend, that keep your eye on the vision, the big goal, but how will exactly we execute? And I think that's very, very important.

>> That is important. All right. Last question for you, very insightful gentlemen. Because we'll use this for our promo for the next Qlik Connect next year. What do you hope to be able to say when you sit down with us in 2025 that you can't quite say yet today? I'll start with you, Mike.

Mike Capone

>> Look, I think we're at this inflection point. We talked about this earlier about, okay, you got to do the work. And what I'm going to say next year, I'm going to have Sarvant back on stage for sure. And a few other customers talking about, they heard us when we were on stage saying, you know what? We need to focus on the data quality, data work, and now we got the outcome. And that's going to be we did as well as these guys already did with their 90,000 trucks and preventative maintenance. And Sarvant is going to have a better story about what he did with his business.

>> And what do you think that story's going to be in a year?

Sarvant Singh

>> I think what Mike said makes perfect sense from my business's perspective. Because, if we will have some use cases where we have gen AI implemented within Qlik, and if we can productionize those, I think that's success. I think that would be a very good story.

>> I think it would be a good story, and we look forward to hopefully having it right here back on theCUBE and in our keynote next year. So Mike, thank you both so much for being here.

Sarvant Singh

>> Absolutely. Thank you.

>> This is great, and thanks for having such a wonderful community who has embraced this and so many fabulous guests just like you. It's been awesome. John, thanks for spending the day here with me in Orlando.

>> Great. Great team.

>> It really has been great. And thank all of you fabulous producers back there that I'm looking at for doing such a fantastic job. And everyone on the Qlik team who showed up today, we've had just a power pack lineup. I really do feel smarter than when I woke up this morning. And thank all of you, our fabulous CUBE community for tuning into Qlik Connect here in Orlando, Florida. My name's Savannah Peterson, you're watching theCUBE, the leading source for enterprise tech news.