Qlik CEO on Building AI-Ready Data Foundations

[MUSIC] Hello and welcome to the Tech Disruptor's podcast here. We talk with CEOs and management teams about their views on Tech Disruption and how it is driving their decision making and strategy. I am Sunil Raj Kapal, software analyst at Bloomberg Intelligence, part of Bloomberg's research department. For those new to our podcast, Bloomberg's research department has 500 analysts and strategists working across all major world markets. Our coverage includes over 2000 equities and credits, as well as outlooks on more than 90 industries and 100 market indices, currencies and commodities. Today we are joined with Mike Kempon, CEO at click. Mike brings extensive experience leading high growth software organizations as CEO of click. He drives the company's strategy across data integration, analytics and AI. Mike, welcome to the podcast. Great to be back Sunil, good to see you again. Great. Maybe if we can start at a very simple level, can you give some insights on how click has shaped itself in the last one year? Of course, you did some acquisitions. And then where is it today? And maybe a high level overview of where you see the company in maybe three years from now? Sure, happy to do that. I can definitely talk about the last year, but it's important to recognize that we've actually been getting ready for this moment for the last eight years. We saw this moment where AI was going to become much more predominant in not just the Apollo world but in the corporate world. And so the whole journey we've been on, which is really building a full-entend data platform, acquiring a data integration technology company and its unity, acquiring talent, a data governance data quality company, then building out a platform and acquiring technology that can help with iceberg, for example. These are all things that were foundational to what is happening over the last year, which is now companies are starting to try to adopt AI at scale. And over the last year, we've been very fortunate because we've been in a great position now to help our customers really understand that you can't skip to the end with AI. I think in the beginning it was plugging in LLM and the world's going to be perfect. It's just not true. You have to do the foundational work. So over the last year, it's really been helping our customers with the foundations around data quality, data governance, high velocity, change, data capture. Being able to build a good foundation so that you can trust AI and want to start using it. And so that you can also now start to move into the genetic world where you're actually taking actions based on decisions that AI is helping you make, which is pretty scary. You better get that right. And the only way you get that right is the foundational work. So everything we do around data quality, data governance, data integration, as well as our analytics platform is all part of that. And then of course we have our own set of capabilities around agents and auto and now in automation that we can also talk about. So maybe before we dive into more on the subject of foundational elements that are required in the age of AI, maybe perhaps can you talk about how is your customer conversations changing in the last one year? And what are the parts of the offering or solutions that customers are really keen about? Because you brought an interesting aspect in terms of the having that foundations, right? And I think we are really excited about all that is happening in terms of code generation and how fast we can build applications. But I see that as a highly abstracted and top level framework, but there is a lot underneath that needs to go right. So just from that perspective, how has your conversations changed and the depth of those conversations? It's really shifted. So a couple of years ago in January, I first started becoming a topic. There was sort of this excitement mixed with fear. It was like, am I moving fast enough? Bords were putting huge pressure on CEOs to say, why aren't we doing more with AI? And it all seemed very easy. And again, it was like, let's just plug in a LLM. Let's just get something going. But now the conversations have shifted. And it really is that, you know, use the word, there's a little bit of a reckoning going on right now. Because it's not that easy, you know, despite what some of the players are saying. You can't just plug in and it works. And so the conversation has really shifted to what are the constraints around AI? And those constraints are, I think, first and all, you have the data sovereignty. Like, you know, there's regulation that's going along with that right now. There is cost. Right. What we're seeing is a real wake up in terms of how much customers are spending on these newer tools and not getting the return. Right. If you look, only, you know, during the stature, you know, some of the analyst terms of public stats, shocking stats at 86% of companies aren't realizing the value they expect from their AI investment. So that's why I say it's a reckoning. And then again, they're realizing that if you get a quality in your data governance, your ability to actually harness all your data isn't good. If your architecture, underpinning all of this isn't up to speed, then you're going to fail. So it really is, how do I manage AI inside of all these constraints and put around a framework and architecture and governance inside my company to make sure that we're getting the value of AI or getting it on the right way without running a valid regulations or upsetting the CFAL with big bills? Right. And also, given what has happened in the last maybe 15 days, in terms of the sentiments around the SaaS companies, how has your communication or maybe employee communication changed because I believe there would be a lot of questions coming out, just in generally in terms of how click is positioned today to take on the shift in the sentiments around SaaS platforms. I mean, I think it is generally, I think every SaaS or software company is now looked under the same lens. I think, yeah, that's where I just want to understand, how has that led to a shift in terms of your internal communications on how what you are communicating internally to your employees when they come up with all these kinds of questions related to where the industry is shifting? Well, look, the good news is we don't have to shift our communication much. What it really is is leaning into our strengths. So again, our core strength is around the quality of the tools we have around data integration data governance, their quality and those things haven't changed. In fact, it become more important. A second strength is our analytics engine, which is just different. We've been in business 30 years and the reason why I click is still super relevant in the, what we used to call business intelligence, we call BI and AI space, which is we have a very different platform, we did memory technology that finds patterns in data that nobody else can find. So many ways we've been doing this for a really long time, we are not a SQL based very things trained platform like some of our competitors. But the real, the real third thing that is really helping us right now is our independence and the fact that we're agnostic because by definition, the world is going to shift with a genetic to a world where data lock in is going to be very frowned upon. In vendors who hoard data onto their platforms should make it very difficult for customers to to get at data inside of their platforms or expect customers to do every possible thing from, you know, system of record, administrative transaction processing to reporting to agents. That's just not a model that's going to work, people aren't going to accept that. And that's where we really shine. We don't care where your data comes from, we don't care where your data goes to. We're completely agnostic, we like to say any source, any target, and we're completely open, we're completely ag driven. That's a huge source of comfort to our customers because quite honestly, the technology landscape is shifting way faster than procurement cycles. It used to be in Army 3 years to take a look. That's like every 3 weeks you're taking a look at what's going on. But the good is is whatever your ag modally, I platform of choices today tomorrow, whatever sources was you using, like we just, we go at the flow. And so that's where we're really doubling down with our employees explaining that to them why we're even more relevant now. And look, this whole thing about the SaaS apocalypse as they call it, it's overblown, right? SaaS companies aren't going to die. They're going to be, they have to be rewired though. There's tons of business knowledge layered into these systems, deep, deep domain knowledge. That doesn't go away. You're not going to run your whole business on a probabilistic engine. from AI. It's going to be a combination of things. And so that rewiring, which is, okay, how do I set myself up, where people are relying less on the the GUI user interface of my system and more on the quality of the data and in agents to to take actions. Understanding that the data in that system of record will probably be married up with data from a totally different system to create an agent to create a decision Matrix and take action. That's where the world is going. And what so else is Sassarun is better adapt but if they do, they're not going to be irrelevant. So what you're telling me is basically okay. AI as a top layer is definitely useful, but it still needs a much more deterministic layers underlying or to drive better performance. You need that deterministic layer. And that's where some of the offerings that you bring together in terms of your analytics platform really shines because of all the domain expertise and how how you have built your institutional knowledge in terms of all the processes, industry, sectors etc. Right. So I think I definitely think that there is a lot of valid reasons in thinking how you are approaching about it. But also, I think if you were to compare today, let's say maybe your own platform, where's this maybe running similar things using AI, where do you think it is? Because I can give you an example like let's say if I still like with some of the LLAMs like when we are running through, you still want to ensure the numbers that the LLAM is throwing up is absolutely correct because businesses need 99.99% accuracy. Right. So even that 1% slip is unacceptable. And I think LLAMs are not there yet but just want to get your sense of where they are today and what do you think might potentially shift or do we ever see them getting those at 99.99 without ever needing all those highly defined or structured way of things. Right. You see every answer you get from an engine known AI engine or a model comes back with all the cadiaz. We sometimes make mistakes and sometimes we make bad mistakes. Look, not in the very foreseeable future do I see these probabilistic engines turning into the level of certainty that you could just bet your business on it. You know, for example, you know, in Q4 this year, we signed a like a five year contract with the large global bank. And why? Why would they do that if the world is using technology changing? Well, because they rely on us for really important things like like the whole cash management and time money laundering or your client like you can't those are things you cannot make mistakes with right and you're not going to bet that AI engine. And you're saying, look, we need we need certainty and as long as we have certainty like these are the solutions are effects. I think you see a lot of that. AI, no question what you alluded to earlier. Co-generation. Great. Fast. Let's go. I think, you know, fully enabled man. I look like there's going to be a gentle solutions that work well that ought to be tasks and make decisions. You know, and are very, very effective. But it's hard to imagine that would just place, you know, the infrastructure and the quality of the solutions that are out there today that people that their entire business on. Right. One of the questions. Maybe I don't know how far you can answer is like how far are we in terms of getting to any TI or super intelligence. I don't think we're that close. You know, I think it's a fascinating topic. And you know, I listen to and read as many things as I can about this topic. But, you know, look, I mean, it's not, it's not implausible with the way compute continues to advance and what's happening. But I'm not, I must then awake at night. We're in about it just yet. All right. So just coming back to click. Can you talk about where we are now in terms of the, maybe where you were in terms of the customer momentum or the business momentum, maybe a year back when we had our last conversation and where you're today. And how are you thinking about the overall time market because now with, yeah, it potentially you could accelerate some of your product and project initiatives that was not possible. Previously, without the, I mean, adding enormous amount of headcount, but with AI like we talk about there is a, there is a massive shift or a meaningful shift in terms of the productivity that it brings in. So how are you thinking about your product roadmap and then the overall business roadmap? Almost just starting with, you know, the business outlook, the business outlook is quite strong. You know, if you add AI into what I'll call the analytics, tam, you know, that's a, you know, high teens growing addressable market, which is, which is crazy. Right. That's a receipt of love to be there. And then on the data data side, you can imagine fueling AI that's also, you know, a teens growing market. So I'm lots of opportunity out there. In we, you know, we continue to focus on both so we continue to focus on building our data and creation, data quality, all that governance that that cost we really need now our AI. And that is being fueled also by consolidation in the market, right. So you start to see some of our some, some competitors being swallowed up, you know, merge, etc. And that always ends up being tailored for us, right. Because it creates uncertainty. It creates. You know, different, different views of different vendors, let's just say. And so, you know, I think we're benefiting from that in art, you kind of our, it or the, you know, the integration world. And then an analytic side, it really is building on top of the asset that we've been building on, you know, our core analytics and continues to be real differentiated in the market. But just last week, you lost your own set of agentic solutions for customers of framework customers can build on. We're going to launch an agentic solution for data quality next quarter. So we're really trying to, you know, stay within our space, which we believe is really super relevant. What customers will trust us to do, but then leverage the newer capabilities and technologies and the power compute to bring real value to our customers. And yeah, like we're, you know, we're taking full advantage of AI capabilities out there. So, you know, things that code generation really helping us fuel our innovation engine. And without having to add, you know, a lot of net new headcount, and then we continue to be active on the M&A front. You know, so last year we, we apart a company called up solver. They're really good at your kind of iceberg technology that's happened to format, which is really helping our customers because they're gobbling the costs again. And you know, iceberg is a wonderful way to actually like manage, manage your costs. So, so really, you know, running on all cylinders right now. It's an exciting time. Great. So you're just about productivity gains, especially on the code generation side of the business. So how does this change your thinking in terms of the headcount because one of the, one of the risk that we see is definitely, you know, I mean much of not not per se the data data platforms or analytical platforms, which are not like a seed based models, they are fundamentally platforms. So, but then how should I mean, what is your sense of, I mean, if you see that 20% or 30% gains, how are you thinking in terms of your headcount strategy is that are we still building more headcount or are we seeing now, okay, we will not be building headcount, but we will be building on our new product initiatives and maybe rolling out new market strategies, et cetera. Or are you thinking of, okay, maybe that gives an opportunity for us to like look into the margins or how we, how we boost our margins, maybe do some kind of a realignment of the headcount. So how are you thinking in just in terms of the headcount strategy? Sure. Well, at first let me say this, I am incredibly skeptical when a company comes out and says we just laid off a thousand people because we deploy AI, we don't need them anymore. Very often those are companies that have a horse in the race in terms of AI solutions, right? I think, if you wake up one day, you only need a thousand people, I think there's probably other things going on in your business. I would say our strategy really is, look, we want to grow and we want to scale past business and we want to scale the top line faster, you know, than expenses and the AI and the productivity gains from that are an excellent mechanism to be able to do that. So, you know, starting with developer productivity for sure, being able to get solutions to market faster. I don't really view it as a, I'm not going to go out because I get, you know, 20 to 30 some prototyping improvement by using code generation to AI. I go lay off the third of my developers, I'm going to get products to market faster, know, and release solutions. solutions faster, you know, what you've seen from us, the rejected framework for release, what we're going to release of you to is benefiting from that tail and the productivity generated that AI is talking to generate. And then yes, as we automate certain things, you'll see us not grow headcount and through, you know, and maybe reduce headcount in certain areas, like, you know, some of the the GNA functions. But I want to reinvest that back in my business, my margins are fine. Like what I really want to do is make sure that we are fit to compete in this really fast evolving world of technology or send it around AI, but but also all the things around it, you know, that we the infrastructure to be support. When it comes to clicks offerings today, so what are the key? I mean, how would you segmented your offerings today, which are the growth engines or maybe say, as you accelerateers of the business, right, if you were to segment, you click as a three-part business, I don't know, whichever shape you want to do it. But if you were to segment it, how would you do that? And what are the priority business areas for you? Yeah, we really segmented it in two ways, although I have to tell you, you know, those are starting to merge together as the, you know, the becoming industry was all, you know, when you think about building AI data pipeline and then using that, you know, to feed, you know, the agents or the feed sort of an, an answers engine, you know, those things did the way we built it, they can all hang together nicely. But we do, for the moment, we do consider sort of what we call our data side of the house and their analytics side of the house. On the data side, it really is what the what we call click town cloud, which is the convergence of, you know, the, the click down integration technology with the really awesome assets we got, the we apart town, the couple years back, 222. And that was, and that, that manifests itself now in what we call click town cloud. And that really is the ability for customers to harness data from anywhere. So any source from mainframes to, you know, as 400s to SAP to TikTok, you know, you name it, we can at high velocity, without no load on those source systems be able to pull data. And then do all the things you need to do, you get to AI, you gobernate, transform it, you check the quality of it, we have a trust score for AI so we can have to inspect your data long way and, and give you guidance as to whether this data is fit for any model, is it, you know, this is a duplicate, this is the quality bad, et cetera. So from that, you know, and you can land data anywhere. So you want to land in the stilflight, your data breaks, your red chip, great, like we're super supportive of that. Again, we're very agnostic and customers really like that from us. So that's super one side of the house. And then you on the other side, you've got your core analytics business and our business is really 50 50 right now, look top line. And that that business could use to be really strong. But what we added to that is all the things you'd expect around, you know, modern AI capabilities. So, you know, we'll be probably going to have our own sort of, you know, gen AI capability. And we launched that last year, this year, what we've done, which we think is super unique in the market is we now have the ability to harness structure data and unstructured data. So data that comes from traditional like, you know, transaction process system and unstructured data and PDFs, some Microsoft Teams are again, the social media platform and create generative AI, you know, answer capabilities on top of that, which customers are really, really excited about. And then we launched our agentic framework so the ability to actually build agents on our platform. If you choose to do so, you can you can do that with on top of our data integration and analytics capability. So really is a full end and suite of offerings. But again, we're super careful about not saying, you know, you got to use everything for us. And we recognize a very heterogeneous environments. We try to meet them where they want to meet us and not force them into any specific paradigm. And then we're sure evolved for us. Also, when we talk about your customers and some of the business areas that you mentioned is one on the governance side and then the data side and then the analytic side of things, right? But if I have you seen a shift in terms of the competition layout because now, Infomatica has been taken over by Salesforce. And then, of course, CTA has had their own issues in terms of their predictive analytics, business momentum. So how are you thinking today? Has you stand in terms of the competition layout? And yeah, where is that going and which are the players that you often see when you are competing for a new contact? Sure. So yeah, it's a very interesting competitive landscape that we compete in, you know, because we do have the traditional say data, competitors of in that market is definitely consolidating right now. So you think about Infomac are going under Salesforce, you think about 5 Trant and DBT merging together. I know some, you know, there's some other companies that are, you know, probably looking to exit in the near future. That generally is good for us, because as you know, you'd like just back in the old days of click, you know, what we made all of our money, I was here, but it was a joyful time. And click when condos, you know, got bought by a by B.M. and business object, got bought by SAP like the innovation tends to slow. And they got gobbled up by kind of larger enterprise and they may not have the same level of focus. And click analytics. Now was the agree year for us. We're seeing similar things now is your assets get taken, taken out, you know, the focus goes away. This is all we do. Like 100% like we make every day data analytics AI, I like they don't that's all we talk about. We don't, you know, we don't do CRM systems. We don't do any of that. And we certainly don't do Bitcoin, you know, which is another interest of one of my competitors. We just all our investment goes into, you know, but the things that we've been talking about throughout this, this by just so, yeah, like, you know, so far this has been a fair world competitive environment for us. And you know, look, we, we, I would say on the analytics side, we often get paired, get prepared to say hello, we also got bought by Salesforce and power be I, you know, Microsoft tool, if you look at, you know, sort of the, the leaders in that space, it's three of us. But really, it's not, it's not apples for apples, you know, we, we, we are not a single database equals, SQL stores, analytics, will, you know, where an enterprise grade full platform analytics, you know, capability and memory, we saw really hard complex problems around supply chain around security, cybersecurity, et cetera. And that that's always going to satisfy part. But it's fun. Like I got to tell you, my life is not boring. It's not like I see the same two competitors every day. So when it comes to your customer segmentation, how do you go about it? I believe you, you offer to both highly regulated clientele, right? And then you have a much of these other players, like in other industries. So how do you segment today? And what does that mix look like? If I were to say ask your exposure to the highly regulated industries or client tell, where of course, there is a lot of emphasis and importance plays for all of these issues, like foundational data, governance, sovereignty of the data and accuracy of the data, where there is, I mean, little lapses that could be acceptable, right? So, it was the other industry. So how do you, how do you see your customer segmentation? Yeah, well, the good news is we have 40,000 customers, which would tell you that we represent pretty much every industry within our customers. Now, that said, eight of the 10 top global banks are click customers. We have a huge footprint in financial services. And yes, like we have killed to them because of our security capabilities and because also our ability to run hybrid, right? So a lot of a lot of vendors now are saying, you know, cloud or die. And you know, we're not doing that. Like we, we, we have a robust cloud offering our compasses is growing phenomenally. But some customers are ready for all the reasons you talked about, you know, their banks, you know, presence in China, right? We like, we got to work through that. And so we allow customers to move at their, their course in speed. There is some inevitability about cloud over the long run that you have to give customers a chance. And that helps us compete in these highly regulated environments. The fact that we are your roots are European and we saw a large presence there helps us, you know, because a lot of the data privacy data, you know, security requirements are coming out of Europe right now. We're very much in front of those, which also gives us an advantage in the regulated world. And we get consulted a lot by the various parties in the EU to be able to actually get to get input on like legislation to thinking about. So, um, so yeah, but you know, again, like so we, we appeal, I think we have an edge when we do when we work with regulated industries. But we also, you know, of our 40,000 customers are all over our directs. sales team definitely focuses up market, but then we have a large set of partners in channel, you know, challenges, figures of work for smaller companies. One of the things that is gaining prominence is these discussions on sovereign AI. So what has been your experience or interaction when it comes to that level of solution building and where do you think things are going? Because AI definitely, although brings in a lot of productivity, it also brings in a lot of risk and then the risk of bad actors potentially rises far ahead than what we have seen so far. So just from that perspective, I mean, what is the kind of discussions at a very high level or maybe at a sovereign AI level? It's a huge topic. Without Adaltsuniel. So we spend a lot of time talking to our customers away, frankly, we have a lot of big government contracts, you know, with many, many countries, not just the US, but many countries. And we spend a lot of time trying to stay in front of that topic. And again, the good is we reach sort of trusted partnership with many of these entities where they actually bring us in early to talk before they actually enact something to talk to us about how they're thinking about it. The reality is, is moving that way. So sorry, AI is going to be a thing. We've been told by multiple different government authorities that they're moving in action. Their diversification is also becoming a big thing, not being reliant on any one higher scale, you know, to run public platforms in particular, but even private enterprises. And so we're moving that way. You know, we're building out infrastructure, be able to support countries or regions that want, you know, sovereign hosting and don't want their data to leave. And then the real power of our platform is, you know, we've got best-in-class data-laying its capabilities, you know, so the ability for us to help customers understand where a day is coming from so that if you do something that you shouldn't have done, you took a piece of data out of some ERP system in Germany that somehow landed in. You asked me to show you that. That's a very powerful tool as completely start to stare down your potential fines from from regulators, you know, for not handling data appropriately. So I find that to be a real strength of ours and some of your customers trust us. One of the things you touched upon earlier is the debate around on-prem with us cloud and the some of them would say, heck, you either be on cloud or be dead kind of a phrase that you used, right? So from that perspective, but I mean, I'm just thinking through, if we were to think about all the AI-led productivity gains, etc. and the need for greater need for data security and greater control about your enterprise data or intelligence. So are you seeing some kind of a shift in terms of how those discussions are ongoing, are we seeing customers now saying, okay, maybe now that I have tools, now that I have productivity gains from AI, etc. Does it make sense for me to have a on-prem solution that is less reliant on cloud or just interacts a cloud in a limited way because I don't want to be passing on all my data to the cloud LLM providers because we don't be still don't know how it is going to be used in terms of training the models, etc. I think I'm sure you might have come across those kinds of questions. So how are you thinking about this and what is the debate around on-prem versus let's say cloud? When it comes to the discussion around, a cloud versus non-cloud, I think there's a lot of my thoughtfulness going into it right now. I don't think it's just like everything goes to the cloud. Now it's somewhat industry specific. I'm here in New York and look at my window. I can see probably three of the largest global banks, headquarters out my window here. And yeah, last week I had a conversation with one of them and they're like, we're not, we're keeping a lot of promise. We're going to be very thoughtful. It's not anti-cloud. So we're going to take advantage of the capabilities in the cloud and certainly a lot of AI is fueled by the advantage of the cloud, the horizontal compute scalability you've got there, obviously accessing some of the models and capabilities that they're, but they're doing it selectively versus wholesale. And what they appreciate is partners who will who will work with them on that and not just say everything has to be this way. And we're seeing that. And again, I think over time, as we solve some of this sourcing problem problems and sort through some of the shipping last beat, you'll see more movement to the cloud and people less afraid of it, but it's going to be a thoughtful process. It's not going to be all in nothing. So maybe at a strategic level, one question is, if you were to be thinking as a visionary here, thinking like 10 years out, right? So which of or what is the biggest shift we are going to see, maybe 10 years out. And just what are your high level thoughts on these subjects like where SaaS is today? Where will it go? And then also in terms of this AI lead momentum, how are you thinking through? I think what you're going to see the cut, let's put it, let's put it in this context. The companies that are going to be successful with AI are going to build an architecture that's going to allow them to reroute data, compute, and AI capabilities in a very agile fashion. The world is going to shift, models are going to change, systems of record are going to remain, but they're going to become more very business purpose focused, the little less around the AI and analytics. That's going to feed into this. And what you really need is this flexible architecture where you can pull data from any source anywhere that you have access to and make decisions quickly and then be able to make changes very, very quickly, not be locked in anyone paradigm. I think customers and companies can build that framework. They'll be able to take on whatever comes and maximize the AI opportunity. If they don't, if they allow themselves to be locked in to very structured environments, which used to be the desire of the ACIOs, I think they're going to suffer and not be able to take advantage of it. They might even perish if they don't keep up. Great. Mike, with that we come to the last session, which I call as the three straight arrows. Are you ready for those questions? Bring it. So my first question is when you wake up every day, what is the one metric that you watch out for in the business? Typically, there's still going to be new license bookies. That's all an old school software guy. That's what we measure growth. Great. My second question is about what would be your one advice to maybe anyone watching out in terms of the AI shift and the shift that is it is bringing on. Because we could all get carried away in terms of this momentum or high level abstracted messaging, but there could be a lot, a lot lost in terms of the underlying shift. From that perspective, what is your one message that you would think would be valuable here? The message that I've been passing along to all the CIOs, ME, all the C-level execs I talked to is don't panic. Everybody's having some struggles with this. When you hear stories about people who have massive success with AI, it's a pull these exaggerated. At the same time, you need to have a healthy fear of what's to come and realize that the best thing you do now is just understand you have to do the work. If you build the foundation, it'll sell your software success in the future. I would encourage everybody who is facing this to understand what your data architecture is, what your foundation is. Then if you get that all right, the model of the week will come in and you'll be able to take advantage of it. If you don't do that work, there you'll be left on. Great. My final question here will be what is the one book recommendation that you would give? Maybe it could be on AI or it could be on anything unrelated just from your last one year's read. Yeah, I've got it. Sure. If it's the one, I just read. It's actually I know not everybody is a big fan of management consulting. I think that a lot of value and this book, which is called Be Wired. which is the McKinsey guy to outcompeting the age of digital in AI. He's an amazing book. I've got a lot out of it, so I highly, I have to recommend it. It's a, it's a powerful read. Great, great. I think I know the author and probably I read one of his other books. I just cannot bring up on top of my mind, but then yeah, I think, yeah, it seems to have a lot of thoughts on all that is happening on AI. Great. Michael, thank you very much for coming in and bringing all your insights in terms of how the word is shifting around AI data analytics governance, right? So we covered a lot of topics. This was super interesting discussion and I think it was, it was very useful from my context. Thank you very much.