How Experion Technologies Is Connecting AI Agents Across the Investment Lifecycle

[00:00:04] - [Speaker 0]
What if the biggest opportunity for AI in financial services isn't about choosing better stock, but actually helping people make better decisions and faster? And behind every pension, investment fund, and savings account sits an industry processing enormous volumes of information every day. And today, I've arranged for not one, but two guests to come and join me and explain how AI is helping turn all that information into action and also understand why data doesn't have to be perfect before you begin and how organizations can modernize without losing the human judgment that remains at the heart of every investment decision. But enough scene setting for me. Let me introduce you to my guest right now.

[00:00:56] - [Speaker 0]
So thank you for joining me on the podcast today. Can you tell everyone listening a little about who you are and what you do?

[00:01:04] - [Speaker 1]
Yes. I'm Markus Reiterman. I'm a member of the Experion Technologies Advisory Board, and I have worked in the institutional global asset management industry as global chief operating officer for over three decades.

[00:01:17] - [Speaker 0]
And I have not one but two guests joining me today. Suraj, could you tell everyone listening a little about you two?

[00:01:22] - [Speaker 2]
Hey, Neil. I'm Suraj Ali Mohammad. I'm the global head of data and AI at Experian. For over twenty years, I've helped organizations turn their data into value, which I'm sure as we'll get into, it's probably harder than what it sounds. Recent finalists at some of the awards, DataIQ, British Data Awards, and asset management is a space that's actually close to heart.

[00:01:44] - [Speaker 2]
So because the problems are real, data is rich, stakes are high. So looking forward to the conversation today.

[00:01:49] - [Speaker 0]
Me too. And just to set the scene, each day, I try and demystify an area and how it is impacted by technology, often an area that listeners might not associate with technology. But I would say, Markus, that most people listening will have a pension or some kind of savings being managed somewhere. So in plain terms, what is an asset manager actually doing day to day, and and why should they care about whether AI is transforming it or not?

[00:02:17] - [Speaker 1]
It's AI definitely, impacts everybody both in private and professional life, but there there are over a thousand no. There are thousands of asset managers of differing sizes and investment capabilities, across the globe. Just to give you a little bit of context, you know, the top 500 asset managers, for instance, manage over $150,000,000,000,000 on behalf of you and I and Siraj, other private and retail clients, pension funds, insurance companies, sovereign wealth funds, and charities. So it's a massive industry. And of course, in a in a massive industry, talent and technology are absolutely essential attributes of success.

[00:02:56] - [Speaker 1]
So therefore, utilizing technology solutions, lots of solutions, not just artificial intelligence, in order to augment human intelligence is mission critical for any asset manager. And, you know, an asset manager collects and analyzes vast amounts of information of structured data, numbers, profit and loss information, and unstructured data, sentiments which they get from social media and other sources. And they collect it from countless different sources and companies, including the Internet and indeed specialized data services providers. So that research is used on a daily basis to make investment decisions, buy, sell, or hold a certain position, and indeed also to assess investment risks as well as projected investment returns. So it's an integral part AI technology of any company that manages money either on their behalf and or on behalf of a third party.

[00:04:01] - [Speaker 1]
To your earlier question, Neil, you know, anyone basically should have a some saving and pension. I hope, sincerely hope. So using technology to review the data collected will no doubt create more time for fund managers and research analysts to focus on investment decisions rather than having to undertake repetitive data analysis tasks. That reduces operating costs and makes their fees more competitive for the investors, so you and I, and and others, and also positions them to create new products which might be of benefit in the future. So technology and talent are at the heart of an asset manager like, anywhere else in the world.

[00:04:46] - [Speaker 0]
And just to bring that to life a little, let's walk through a journey of a typical investment from, let's say, an idea to a settled trade. Where does the most time and the most effort get wasted today, Marcus?

[00:04:59] - [Speaker 1]
Hugely simplified, a typical investment process entails a number of core steps. So it all starts with research. So what I just described, they go out and look at companies they might want to invest in or might want to disinvest in. They then come up with the so called stock selection. So which shares or bonds they wanna buy, sell, or hold.

[00:05:21] - [Speaker 1]
And that is also then put into the context of what we the industry calls asset allocation. So in a in in, you know, out of 10,000, how much should be in shares, in bonds, in cash, or other investment instruments. And that then leads to portfolio construction. So they basically build a fund or a solution for a particular client, and that entails risk analysis, and then ultimately, once they have made a decision, buy, sell, or hold, trading occurs. So they buy, then they settle, they have to then do reporting, they do performance analysis, they report back to the client, they report internally and do, you know, regulators, etc.

[00:06:00] - [Speaker 1]
So a typical investment process starts with research, portfolio construction, portfolio analysis on risks, and then trading and execution. Now invariably, looking at operating costs of a typical asset manager, as I sort of opined earlier on, people and technology typically constitute the largest operating expenditure, no surprise. So people costs in an asset manager are typically tilted towards the investment professional, because they attract bluntly the higher salaries and bonuses than other parts of an asset manager. That's just the reality. And therefore, using technology to optimize the the fund manager's time and other human resources associated with it makes total sense.

[00:06:51] - [Speaker 1]
So this is why it is essential that technology is part of that journey of investment management.

[00:06:59] - [Speaker 2]
Yeah. And I think I think biggest opportunity too. Right? So when you think of it from a all the different systems that Markus kind of outlined there, as you go through that various journey, it is there is plenty of opportunity there for data to play a big part in that technology people problem that Mark has outlined, and enable businesses as they go through their journey and, you know, make the most of the opportunity.

[00:07:24] - [Speaker 0]
100% with you. And I think there also is a a series of disconnected problems from research, portfolio, construction, trading, back office. I'm curious from what you're both seeing here. Is there is there a single root cause that's connecting all of them?

[00:07:39] - [Speaker 1]
I mean, my my view, Siroj, might have a different one is is there's not a single cause, frankly, because everything is is connected, actually. There is there there are I have not met an asset manager in thirty plus years who doesn't have legacy systems, so systems which have been bought, introduced, developed, and continuously advanced, you know, for years, basically. There are also manual interventions, not everything of the process are just typical, well, the typical process I described earlier on, is automated, so there's manual intervention. And then of course, client demands change, you know, they want new products, they want new reports, there's new regulation, etcetera, and therefore, there isn't an answer in just having more AI tools or different technologies. The answer is that asset managers have to develop different architecture to basically have AI agents, as we call them, share a common foundation and operate alongside alongside humans.

[00:08:39] - [Speaker 1]
That's quite a difficult, undertaking because, again, you talked about disconnected problems, but they're all connected because it's the people, it's the processes, and it's the technology which have to align, seamlessly, in most cases, but that's quite a difficult undertaking. I don't know what you think, Siroj, on on this particular topic.

[00:08:57] - [Speaker 2]
Absolutely spot on there because as as we say that people are at the core of what drives businesses to success, make or break almost. And one of the great things with people is we are all put in situations where we've got to, you know, be on the spot and make a decision. And often, that decision to innovate might result in a new tool or a new platform that you wanna bring on board, which might not always sort of factor in everything else that you have been doing so far. Right? Because you wanna make a business decision at that point in time.

[00:09:28] - [Speaker 2]
Eventually, when you do hundreds of these as you go through your journey, you end up with some of these disconnected profile. And and I totally agree on the architecture piece, Marcus, because I wish we all had the time and energy to go away and kind of optimize the architecture as we go through, which isn't the case. Right? So by definition, we are building silos or disconnected pieces as we go along. Right?

[00:09:51] - [Speaker 2]
And at some point, we've all got to, you know, keep the house in order, make sure that we're all aligned and that we are not spending twice the amount, you know, multiple subscriptions or facilities that you could do with one platform. But, yeah, totally aligned on that.

[00:10:05] - [Speaker 0]
And if a member of a fund is is listening today and find themselves asking or thinking, why does any of this matter to me? What would you say to that person listening?

[00:10:17] - [Speaker 1]
Well, my my answer is that, again, technology is a critical component of any company's current and future success. And as an investor, I wanna be assured that my fund manager or asset manager can respond quickly to unforeseen or foreseen circumstances. You know, particularly at the moment where we have geopolitical challenges across the world, you know, markets are volatile and an investment manager has to be fast, focused, and look at, as I mentioned early on, almost yodobytes of data. Therefore, it is in my interest, the investor, to be assured that the fund manager can respond quickly, optimizes the human capital, you know, keeping costs for me down, so they because they have to remain at competitive levels. And also, more generally, you know, I want my fund manager to respond to my new demands, speed up products, you know, new reports I would like, electronic access to my portfolio twenty four seven.

[00:11:22] - [Speaker 1]
And and therefore, it is again in my interest that the fund manager has access to deeper data pools from more data services fast.

[00:11:32] - [Speaker 0]
And, Suraj, Marcus has, described a world of connected silos and disconnected tools there. And I know you work with some of these firms on the solution side. And to ensure we don't leave anyone behind here, would you mind explaining what composable AI actually means with without all the jargon there?

[00:11:51] - [Speaker 2]
Sure. I love this question. Because whilst Composable AI sounds like a jargon, the idea is actually quite simple. Yeah. For example, think of think of Uber, for example.

[00:12:01] - [Speaker 2]
Right? We all have it on our phones. They didn't build their own maps or their own payment services or its own messaging. What they do is actually use that sort of they compose existing components into providing you, the user, you and me, a seamless experience. Right?

[00:12:17] - [Speaker 2]
Each piece does its job, but they all share a sort of a common layer underneath, which is why they work together. Right? So so when you add a new feature, you're actually not rebuilding the whole thing as it goes by. So composable AI applies that sort of same logic to business. So instead of sort of one giant AI solution that tries to do everything or a separate AI sort of built for every problem from scratch, you're actually building a library of AI components.

[00:12:48] - [Speaker 2]
Markus referred to them agents effectively. Right? So there are agents that one that reads and summarizes documents, another one that kind of monitors a portfolio for risk, another one for checking compliance, for example, and and maybe yet another one that assembles a report. So the the key there is actually to bring a common foundation, the same data layer, so to say, the same governance. Right?

[00:13:11] - [Speaker 2]
So the same audit trails so so you know what's happening or where it's coming from. And because they all share that, the outputs connect. So the document agent's findings often feed into the portfolio agent. The risk agent then kind of can trigger it based on a compliance or and you're not automating tasks in isolation, but you're connect connecting the whole process end to end. Right?

[00:13:32] - [Speaker 2]
The key thing though here is actually the economics. That's the punch line. Because once you've built and tested an agent, adapting it for the next use case is actually an adaptation as opposed to building it from scratch. So you've got you've got that sort of 60 to 70% covered already rather than starting over. And that's kind of what turns AI from a series of expensive one off experiments, which I'm sure a lot of us are already doing, into a sort of an actual enterprise strategy.

[00:14:01] - [Speaker 2]
I was talking to this about some someone to someone recently. Think of it as kind of building with Lego instead of script sculpting from clay. Right? So if you had Lego bricks, you actually have the pieces that are pre built, in some cases, doors and windows even, because you all you know, when you're building a Lego house, all you're actually doing is actually reusable components clicked into places, and it all works on the same platform. You know, there are no two pieces that won't stick together types, and you're actually building it together into a system.

[00:14:32] - [Speaker 2]
So, that kinda makes sense.

[00:14:34] - [Speaker 0]
Yes. It does. Absolutely perfect. I love it. And if we go to the next stage, from there, can you walk me through what this actually looks like across the asset management value chain from research all the way through to the back office?

[00:14:47] - [Speaker 0]
And, also, what FS, Fabrics Experience Solutions approach, how how that, works at each stage. And our apologies, I've asked probably about three, four questions there. But if you could walk me through from beginning to end.

[00:15:00] - [Speaker 2]
Sure. No problem. So FS Fabrics, it's let me go stage by stage and sort of maybe keep it grounded as we go through. We'll start with research. Right?

[00:15:10] - [Speaker 2]
So for the research element, we have a bunch of reusable agents that kind of ingest data from any sources. Markus referred to a few of these coming from different databases, could be from the web. So it could be internal or external. It could be in any format. It could be Excel files.

[00:15:25] - [Speaker 2]
It could be web pages. It could be CSV files or text files that comes in. And so the research agents kind of run tasks that are like thematic screening and quantitative research in sort of seconds. Right? The example I'd kinda give is, I think, Marcus, it's probably fair to say a firm typically would get anywhere between 200 to 500 broker reports a week, for example.

[00:15:47] - [Speaker 2]
Right? And if you do the math, each of them taking about roughly, say, forty five minutes to read properly, teams skimming. If you wanted to do a lot more of these, teams would end up skimming them through them and kind of insights would probably get missed. Right? But when you put AI into tasks, and if you give AI to the teams who are actually doing this manually, they are now able to read one of these in maybe less than half a minute.

[00:16:11] - [Speaker 2]
Right? Because for about 300 reports a week, that's like 200 to two fifty analysts are saved or freed, and which can be redirected into that act that thinking that actually needs a human to do the thinking. Right? So you're taking off the heavy lifting piece of that research and putting time back into the analyst's hands that they can do real work on. Right?

[00:16:34] - [Speaker 2]
So that's that's the research side. Now, the second stage is actually more things like portfolio construction, for example. Now, the research agent has done all the hard work. The output from that agent flows straight in. The agent then translates these investment signals into what the optimizer understands.

[00:16:50] - [Speaker 2]
It would run through a series of stress tests and flag where the portfolio has drifted enough to act. Right? So a review cycle that then, if you think of something that would have taken a day or more, probably now takes hours and runs continuously and can and not once a week type stream, because you can do that a lot more. On the execution side, you could have something like a pre trade agent, for example, that recommends the best way to place a large order, which venue would you do that with, what algorithm, what size. And you could probably do all of that before the trade, not really after.

[00:17:25] - [Speaker 2]
Right? And so execution quality can vary maybe 10 to 15%, which at a fund scale is real basis points back to investors, right, in terms of returns. The other next element would be performance and risk. So daily AI attribution instead of an a monthly report, continuous risk monitoring that predicts a breach before it happens, maybe mandate compliance that kind of filters out the false alarms so that teams see only real issues. Those are areas that we can help with performance and risk elements.

[00:17:57] - [Speaker 2]
And finally, this is sort of the back office element. So a settlement agent can flag trades which are at risk of falling before the settlement day, so cutting failings by about maybe 35% roughly, which matters enormously in the sort of t plus one world. Right? And so regulatory reports that can assemble automatically from a governed or a traceable pipeline, running maybe 75% faster, these are all sort of benefits that you get out of these agents running individually and able to run much more quicker. So if you think of sort of that agentic system that we talked about, the fabrics, composable AI that brings all these different capabilities together, that's the payoff.

[00:18:39] - [Speaker 2]
Because they all share the same data layer, insight can travel between agents very fast. Right? So, compared to what probably is happening in a lot of organizations today, that patterns that you see in back office settlement patterns can be transfer transferred back into the front office risk and portfolio decisions fairly quickly. Right? In a siloed world, that signal never travels.

[00:19:02] - [Speaker 2]
But whereas in a in what we're talking through in an FS Fabrics or a composable system, it does and does automatically. Right?

[00:19:10] - [Speaker 1]
And I think I would highlight as well, Siroj, I totally agree, is also speed is the overriding benefit, really. It's not just the cohesiveness of your LEGO pieces or modulars, modular approaches, but it's actually having access to information which is correlated, both for reactive and proactive decisions ultimately. And just to expand that very briefly, you know, these days, if there is a major issue in a gold mine in Papua New Guinea, you will actually, a fund manager, will know this information very quickly, within minutes, whereas, you know, perhaps five, ten years ago, it would have taken, you know, two or three days until it gets into the news, etcetera. And that's sort of the other side of what you just described with Fabrics. That is, you know, the external changes are so fast and furious at times that actually the internal capabilities of an asset manager have to be enhanced, and one of the solutions very clearly is fabrics.

[00:20:12] - [Speaker 0]
And that's just and I suspect something that you have both seen that can derail AI projects inside large organizations is the data problem. Our data isn't ready. It's probably something you've both heard so many times, but how real is that barrier? And what do you do at Experian to tackle this?

[00:20:31] - [Speaker 2]
Yeah. I think I I think that's a real issue. Right? But but I also think it is probably the most overused excuse for inaction. Right?

[00:20:41] - [Speaker 2]
I say that having worked with many organizations on exactly this. Right? Because the truth is almost no one has perfect data. Right? Waiting for that sort of perfect dataset before you start with AI is a strategy that guarantees that you'll never start.

[00:20:57] - [Speaker 2]
Right? So the right question that you really wanna ask is is not whether your data is perfect, it's more about is it good enough to get real value in this domain right now. Right? So there is that sort of, you know, is good enough and that timing quite relevant. The Fabrics platform that we talked about earlier includes a sort of data readiness pipeline, and that layer kinda sits between sort of your organization's existing data and the AI agents that we explained.

[00:21:27] - [Speaker 2]
Right? So it handles everything from ingestion from multiple systems. It cleans it, validates the data, builds what is called a a RAG ready knowledge layer. Right? So RAG being retrieval augmented generation, which means that sort of when you ask AI a question, it answers by looking up specific verified sources rather than guessing what the answer could be, which I know AI can be really good at.

[00:21:53] - [Speaker 2]
So every output that you actually get back can be traced back to the exact document or the data point that you have in the system. Right? And you can always see that evidence and which is what often regulators will be looking for when you when you really need to deep dive. In terms of where to start, I would say never boil the ocean. Right?

[00:22:11] - [Speaker 2]
Because we we can we often begin with a sort of a two to four discovery piece to find maybe two or three use cases that you can start to clearly measure the sort of impact off. Right? So, we deploy something in a few weeks, not months, and that is by design. So if there are use cases that we think are too complex, you can't prove it immediately, we would really challenge to say, can we do that, a smaller one, that you can sort of validate in a shorter period of time to gain that sort of confidence. Right?

[00:22:41] - [Speaker 2]
And we do that with sort of pre built components. That will then 100% provide that sort of confidence by proving that value and will enable you to get that sort of next wave of fund funding that you'd require for the remaining use cases as you go through. At Experian, we we have been positioning our AI offerings with with something called ARC. And part of ARC is a tool called Rethink, which enables businesses to identify what those use cases are and enlist them with a set of ROI definitions of what you get for investing time into that particular use case. Right?

[00:23:21] - [Speaker 2]
And once you have built those use cases, have got the ROI definitions, it's much more easier for you to win the business over as you go through. The analogy that I sort of often leave people with when you talk about data, it's it's more like learning a new language. Right? So you can't you're inevitably gonna make mistakes as you do that, learn a new language. You don't wait until you've mastered all the grammar and the the the poetry and the literacy around it.

[00:23:50] - [Speaker 2]
You actually start speaking. You start with probably a few phrases. You probably use them. Someone will correct you. You'll get through better with practice.

[00:23:59] - [Speaker 2]
Waiting for the perfect conditions means you'll you'll never begin. So my advice is start with what you have. Let's see what you can make. Make mistakes, but don't get stuck in that sort of prototype cycle too. Right?

[00:24:11] - [Speaker 2]
Think about what you can take back to the business and win win some success and win their confidence so that you can start to deliver a lot more of these AI as you go through.

[00:24:20] - [Speaker 0]
And last question for both of you here. If we have somebody listening today that manages money for a living or works in financial services, and they're wondering whether they should be paying attention or what they should be paying attention to, what what message would you offer to those people listening? And, Marcus, we'll go to you first.

[00:24:40] - [Speaker 1]
Well, I sort of like the quote of Shakespeare if he basically would have lived in the twenty first century, and that is to use or not to use AI. This is not the question. So let me elaborate on that, very briefly. I mean, we we, at Experian and I as a market practitioner, observe the rise of enterprise wide, AI operating models, and and the new commercial reality of quantum computing is an entirely different subject, of course. We also observed the AI paradox, which basically means a lot of spending and, most of the major impacts are still elusive, so we're very early on in this journey.

[00:25:17] - [Speaker 1]
We have a lot of enthusiasm in the industry and, of course, of every provider, be that big or small, but, you know, the impacts are still relatively elusive. And the companies that are truly innovating with AI are doing something very different from their peers, namely, they are developing AI capabilities to reshape their products, the services they provide to the clients, the core business processes, organizational systems, etcetera. In my view, however, their advantage does not just come from the tech they use, their advantage comes from how and how fast they apply technology to solving real business problems at scale. One of the challenges for any senior manager in an asset manager is to actually identify which business process, investment management process, they want to automate, augment, or some set in some cases as well. So every tech and AI transformation is also a people and cultural transformation, because it's not just tech which solves everything.

[00:26:21] - [Speaker 1]
I mean, you have to bring the people along, your staff, and in many cases bluntly also senior management. And every fund manager, basically has to realize that this is not a tech transformation in in in someone's firm, when it relates to AI transformation, but it's also a people and cultural transformation, which starts from from top down. And therefore, the transformative power, of this technology, in my view, has to be unlocked in ways that strengthen, not supplant human judgment, trust, and fiduciary responsibility. And therefore, you know, one of the challenges of AI is that a tech led innovation and an operational integrity must advance hand in hand, because there are risks by using artificial intelligence, are widely reported, and we don't have time to to talk about today. But that's balance in between innovation and integrity is something our industry, investment management industry, has yet to grasp in full.

[00:27:26] - [Speaker 0]
And, Suraj, to bring us home, anything that you would advise and leave everyone listening with?

[00:27:31] - [Speaker 2]
I mean, Marcus has covered it all, to be honest. But I think that culture point, I couldn't, labor on it enough, to be honest. Because as a data and AI practitioner and people being enabled with the clods and chatty pities of the world, solving a technical problem, I would say, is almost getting really, really easy these days. Right? So don't really need the technical background to make a start or make an impact.

[00:27:56] - [Speaker 2]
The culture and and how you nurture that within the organization, that's gonna leave a bigger impact in terms of how how impactful you are or efficient you are through the whole process. So if you're in the sort of asset management for or financial services space and wondering whether you need to pay attention to this, I go back to my earlier statement. Right? So start small, but start. Right?

[00:28:23] - [Speaker 2]
Your competition is already there. They're probably looking at ways of, you know, those who are already investing or have been working on AI and data science for years. They have been probably investing significantly to get ahead of the market. You don't need a perfect data estate like we talked about. You don't need an eighteen month program to get started.

[00:28:42] - [Speaker 2]
You just need one or maybe two use cases that's clear value and you start from there. Right? And think of it as we refer to as, you know, thinking freeing up that sort of thinking space in your organization. Take the data gathering or the sort of copy pasting or reformatting that these highly paid individuals within your organization that they're wasting their time on, take that away from them and so that let them spend their time on what they are really good at. Right?

[00:29:10] - [Speaker 2]
So that they can use that judgment and the decisions where you really need them to do that. So that's not a cosplay. It's actually just you're being more efficient or more intelligent in the way how you use your people resources. Right? In terms of vendors, there are so many players, and Markus touched on this as well, that actually bring in AI into their own offering.

[00:29:31] - [Speaker 2]
So all the tools and technologies that you use there, they're also playing the AI game. So you have a lot more AI tools to hand. And, obviously, bringing something into that sort of common shared data layer that can talk to each other. We talked about FSRabrics, etcetera. That's where I think your head should be at.

[00:29:49] - [Speaker 2]
So if you're an organization that's a bit more mature or getting mature with that, I'd say relook at how you enable for the future, not just answer the questions of today. Right? But that's it.

[00:30:01] - [Speaker 0]
Alright. It's a powerful moment to end on. So, Marcus and Suraj, thank you both so much for joining me today. I'll be adding links to everything that we've talked about, links for Experian, etcetera, and both of your LinkedIns. I would encourage people to keep following you and with the work that you're doing.

[00:30:18] - [Speaker 0]
And I, for one, really enjoyed the conversation and appreciate both of you today for sharing your insights and experiences with everyone listening. So thanks again for joining me.

[00:30:28] - [Speaker 2]
Thank you so much. Pleasure.

[00:30:30] - [Speaker 0]
I think today's conversation showed that the future of financial services isn't gonna be shaped by AI alone. Success actually comes from combining trusted data, thoughtful architecture, and the experience of your people who know how to ask the right question. And I think Marcus and Suraj made an important point today that applies far beyond asset management, and that is organizations that start with practical business problems and then build confidence one step at a time. These are the ones that will move faster than those just waiting around for the perfect conditions. But I'd love to hear your thoughts.

[00:31:08] - [Speaker 0]
Is your organization treating AI as just another tool, or is it genuinely rethinking how people, data, and technology work together? Let me know. I wanna hear all the good, bad, and ugly stories out there. So techtalksnetwork.com. I wanna hear your stories, your insights.

[00:31:27] - [Speaker 0]
Do you need AI agents that you can trust? Well, with an AI data layer providing real time connection within your data platforms, you can trust your agents to provide accurate solutions. So scale your business by trusting your agentic AI, accurately getting the work done for you. Trust its capabilities with Denodo, and you can do that by simply visiting denodo.com to learn more. But I'm afraid that's it for today, so I'll be back again tomorrow with another guest.

[00:31:58] - [Speaker 0]
But thanks for listening as always, and I'll speak to you again very soon. Bye for now.