This CFO Asked for What? How Carlos Olea Launched AI Transformation at Howard Hughes

People have asked me, even people inside the company, like, why should we be the ones that are breaking ground on this? Why should we be the ones investing in automating this when nobody else is? We're Howard Hughes. We're just going to do it. Hey folks, welcome back to Agents of Scale. This week we are joined by a fantastic guest. This is Carlos Olía. He is the CFO of Howard Hughes. Carlos, welcome to the show. Thanks mate, bless her. Carlos, so you have a fascinating background, but one of the things that you have been doing a lot of is just constantly thinking about efficiency and automation and how you can sort of get more out of your days. Where did this come from? It's something that I've had my entire life and for most of it I didn't really know what to do with it. I was an annoying kid looking at other people like whether it was my mom in the kitchen or me getting groceries out the car thinking I think I took one step too many. I can do it in fewer steps. Or if I had started on the left instead of the right it would have been faster. And my mind was always working like that. And I honestly, I struggled for a long time. I thought up for a long time, like, what do I do with this? It was almost like an annoyance, something that was just pinging me in the back of the head throughout my day. And then somehow, not thinking about that at all, but went to school, became an accountant, an auditor, and in many different things, I ended up finding out what to do with it. So now... I do the same but mostly with business processes and with back office operations. So finally have I used for that an annoying nagging feeling in the back of my head that I've had for my entire life. Do you have uh any interesting stories from when you were younger where you were just trying to shave off seconds here? I'll give you an example for me. If I'm driving home late at night, uh coming from somewhere, I will switch lanes from the highway to take the innermost curve every little step of the way just to shave off little bits of seconds here and there. It probably doesn't matter in the grand scheme of things, but my mind can't not do that. Yeah, I know what you mean. when I was in high school, I worked in construction for a summer. And I figured out, you know, we used to have this massively long extension cords to use our power tools in different parts of these warehouses that we were installing electricity on. And I figured out at the end of the day how to wrap them up in a more efficient, faster way. And my only reward was that from then on, I had to do it for everybody. I learned to sometimes give my ideas to myself. There you go. uh So one of the things you sort of have done is you spent like decades building up this like mental inventory of processes that sort of wish you could automate. Can you talk about like what that looked like? What processes in the back of your mind were like, I wish this was better. I wish was this was better. And like, did you actually keep track? It was only in your head or were you keeping track of it somewhere? How did this go? It started in my head when I was an auditor. wasn't my head because I really had no intent to put it down on Excel or a diagram or whatever and think about it more because as an auditor, I was not going to be able to change the way my clients operated. So I would just see things like, oh, this, their process to approve a sale has way too many steps. They could address the same risk in a different way and cut down the processing time and what have you. And before some people around my age will remember, before serving Soxley, you were able to, you actually gave sometimes your customers, your audit clients a written document that says, he's like other stuff that I saw that could make your operations better. After serving Soxley, that went away. You were no longer allowed to do that. It was not an admissible service. So I didn't, I didn't have no incentive to actually do anything with it. It was only when I moved to industry. that I was responsible for some of these processes. And I started thinking, how am going to apply these things? And I would run into sometimes technological limitations, or sometimes the technology existed, but I was not going to have the budget for it. So I started creating this backlog of typical back office operations, anything from how you deal with cash and get contracts approved, get vendors onboarded, fixed assets, like all the typical processes, HR and all sorts of things. And then that's when I did start creating an inventory and I had them categorized by things that I could fix now, but I don't have budget for things that I cannot fix now

even with all the money in the world because the technology didn't exist. And the big unlocks started to happen in the last three years when the technology started to be widely available and cheap, quite frankly, that I finally said, all right, it's time to bring up that list again and start tackling one by one. And I've just been having a lot of fun doing that. What was the big unlock for you where you started to think this new technology, AI, LLMs, actually could start to make a dent in this checklist of ideas you had? Well, I think it's mainly two things. One was that it truly breaks the monopoly that people that knew how to code used to have. And to open it to someone like me, because in the past, they would have to go to someone in IT and say, can we do this? Can we create an automation that does whatever? Right? And if they said no, I kind of had no recourse. I'm not an expert. I don't know how to code. That's not what I do. I'm an accountant. And so it would end there. And now, with this technology, what I realized was completely different. I could actually figure out how to do it. And I could then go to an IT professional for the program and say, I want you to do this. And even if they said, no, that cannot be done. No, it can't because I have the Python code. It was very freeing. Now I can actually do it. But then also the fact that it was accessible, like it's not millions and millions of dollars. Yes, some use cases require a lot of investment, but there are a lot that you can do with low hanging fruit that are very cheap and you just have to apply your imagination to solving a problem with the tools that you have. And largely you're going to find a better version, even if it's not the ultimately best version, because maybe you don't need a lot of budget for the ultimately best in the Urbana state, but you can find significant improvements with very little money as long as you're willing to challenge the preconceived notions and the way it's been done, which sometimes is not that easy. But in my experience, if you can do that, you end up in a better spot. Yeah. What was the first thing you automated? built? The first thing I built that em was end-to-end was really very simple. And it was just looking at different bits from different vendors and figuring out, doing the analysis to figure out which one had the best, it was really actually fun for me, but the best um value for money. Because that was super easy to build something that tells me which one's the cheapest. But I wanted to parse the details behind it to try to create like a waiter average sort of. sort of speak of to give me the best value for money and bring me the best three. And so I wasn't reading, you know, 25. I would just focus on the three that were giving me the best. And that was, there was an introduction where I said to myself, oh, this is really powerful. Like this is actually reading and understanding documents. If I, if, it does that well, then imagine what I can do with all sorts of contracts and leasing. And that's where it started to move to like bigger things that more that a heavy emitter reach. Yeah, it's interesting. One of the things I see with the folks who are most successful is they usually do start with something very simple. I find that folks that sort of say, I can't work, et cetera, they're usually trying to boil the ocean and they get frustrated too early. But the folks who find kind of just some small thing, they can start with that simple thing and then they can start to ripple out from there and solve bigger stuff over time. Yeah, I couldn't agree with that more. And I actually think, I mean, you can use it for something fun too. Just the fact that I've focused on something that some people would say, well, that's not interesting at all. It still showed me demonstrated the capabilities, right? But I know people who have used it to like create uh vacation plans. And I mean, whatever you want to do, you can extract lessons from whatever you're using it that you can then deploy elsewhere, as long as you're thinking about the lessons, right? You can say, it's really cool that Chad's GPT can give me recommendations in Rome and build a day by day schedule for me and you can end it there. Or you can look at that and think this took input from many multiple sources. It took constraints into consideration like time and money. And it gave me a well-reasoned answer. And you can say to yourself, I can use that for a lot of other things inside my company, not just for my vacation. So one of the things I'm most curious to hear right now, I think if you go talk to many CFOs, your peers in this position, you'll often hear folks saying, hey, I'm trying to measure the ROI of this.

I'm trying to justify the spend. And so while they're all over here trying to justify the spend and the deployment, you're actually out here automating the heck out of stuff. What do you think is different about your approach compared to what you see from your peers. Almost every day somebody tells me, you're a very different CFO. And I think that 80 % of the time is not meant as a compliment. appreciate that. Thank you. I think that is precisely that I'm thinking different, a little bit different than 80%. And that's why, but I think it's a foundation of all of this time that I spent identifying inefficiencies, right? It is hard to measure ROI. not going to lie. It is difficult. But at the end of the day, I see it in my GNA load. And so when people have asked me like, why are you so focused on this? Because, well, one, I like it. But from a pure CFO standpoint, I want to cut overhead. I want to run a very, very efficient and effective platform at a very, very low cost and give that money back in value to my shareholders. And so I see it as directly related to the remit of CFO to try to deploy technologies to run an efficient platforming. that's depending on who I'm talking to, the one that's straight CFO answer, that's why I do it. And people who don't do it, I think they're missing an opportunity because otherwise you're literally paying too much to run your platform. Why would you? I'm curious, did you have, most people have to go to the CFO to sort of convince them to do this stuff. You are the CFO. Like, did you have people that you were trying to get budget from, convince this to do this, or because you're in this role, how do you think about your role differently in sort of enabling Howard Hughes to drive this change? Yes, well, I mean, we are a public credit company. It's Howard Hughes Holdings is I had to go to the board. We started asking by very, very little money just to say, there's something here with this blockchain and AI stuff that has enterprise value and we want to explore it. And then when we started delivering on some use cases, then it started to switch and the board now is the one saying, you have to go faster. You got to hurry up. All of those are things that you haven't started. You got to start, you got to get it done. So it was, yes, it was the first, it was a risk. We have to use our political capital, so to speak, inside the company or credibility to say, I'm going to ask you for some money for something that we don't know quite frankly, if it's going to work, but we think it will. And we're going to take it very seriously. And then after that you deliver and you get all the credibility and you get the support. And now it's like, okay, now it's on you to really transform the way we do business. that's, that's thankfully where we are today. were some of those early wins that got the board to say, have to go faster. Yes, some of them were, we used to have uh operations in New York and we gamified a lot of our property and started driving traffic into the property because people wanted to be there and kind of play the little game that we have created for the time that we had it, it transformed into real money. And then the second one was a lot uh more uh back office-y. we started to show what we showed that we could abstract leases, which is something important in the real estate industry. Every time you sign a lease, you have to extract the important information that you need so that I know how much rent to bill you and when and who you are and which unit you're going to be in. And that's normally been a very, very manual process. And when we were able to show that, you can now automate it with a high degree of accuracy, actually more accurate than a human does it. That was a big switch. switch because it's a known problem in the real estate industry that everybody deals with. And when we showed that we could do it, and I think everybody took a breath and said, all right, these guys are for real. We got to start paying more attention. Maybe to paint a picture for folks who are in a similar shoes. What was required to get this done? What did you need from people, tools, skills, things like that to pull these off? Yeah, that's a really interesting question. I you need, I'll speak from my experience. I, I need it to make time because I'm still the CFO of the company. I still have to do all the other regular things that a CFO does. So I need it to become even more efficient with my time so that I could dedicate some hours of my days and weekends to ideate.

I think I really do think that your own imagination is the limiting factor. money comes to us. It's not the tools anymore. It's not the budget so much. It's your imagination. So carve out time to just think about how are you going to attack a problem and potentially reassign a process. Then I also think that you cannot be successful or perhaps it's a lot more complicated if you don't at least have a small set of people who are also thinking like that. You need that team that can create. Then you can use and take to everybody else and start changing hearts and minds, right? But if you don't have the people that don't look at you like you're insane and then say, no, actually that sounds like it could work. We should try to do it. Then you need the small tiger team. And I'm really, really grateful that we were able to assemble it here. And with that team, uh we've been able to start changing people's minds. I don't think it would have been possible again if you don't carve time to just imagine a different way of doing things, a better way of doing things. And then you need to have a core group of people that can share your vision about what you want to do. You know, with those are those folks from all from, you know, finance, accounting backgrounds, are those folks from a different set of backgrounds like? So we have people from traditional IT background that are putting aside their relational databases and their coding languages and are embracing this. But we also have people from startup environment that have said, this is interesting because you have Howard Hughes, large accelerated filing, public company, all the trappings of enterprise. We're supposed to be slow and boring and whatever. And some of them realized These guys are moving at startup speed with enterprise resources that are opportunities there. And so we were able to attract a couple of them and it had like an amazing impact on what we're Yeah. Are there, is there a certain like stack that you would recommend? Like say you've got another CFO that you're sort of sitting down and talking to and like, hey, here's some of the things you probably need to go get. Here's, need these like entrepreneurial like, you know, mad scientists that kind of need to be there. Here's some tools that if you kind of do this, like, you know, you're probably 80 % of the way there. I mean, if I were to start again from the people environment, you definitely need people with an entrepreneurial mind. And in some areas it's hard to find another. You can get people straight from the startup world that are very entrepreneurial by nature. In IT, not everybody's entrepreneurial. In accounting, it's hard to find entrepreneurship. But you need it. That has to be the vein that you all share. And you just look at it with your different... technical lenses, but you will have to share in my view, the entrepreneurial mindset to really, really get after it and act with the urgency of a startup. As far as tools, mean, that's a lot harder because as you know, better than me, I'm sure everything is improving so quickly that the one thing that I would say is like, don't, you can't go chasing every new development because then you never do anything. You start to build on something, pick your platform, make sure that it fits for the use case that you have. And unless something drastic changes, stick with it. Everybody else is going to come up with incremental improvements, but so is the one that you're using. So stick with it, deliver it, and then for the next use case, go through the process again. And for the next use case, the platform that you used before might not be the best one. That's fine. But we had a few starts and stops in the beginning. where we were chasing the next version of all of the elements that you know. And then we're like, guys, we're not achieving anything. We're just switching every five minutes. That doesn't work. So I would always recommend just do your research first. Start with one. If it works for your use case, just stick with it. Don't chase the next one. Let's stick on this uh culture like uh topic. you know, Howard Hughes has this like innovation legacy, you know, sort of attached to it. And I think that's something you take pretty seriously in your role. Like what were like, what are the things you do to instill that throughout the company, but also in a function to your point that doesn't always have a reputation for innovation. Yeah, well, the first one is, yes, you're right. I mean, I think very seriously, we'll work for Howard Hughes. And people have asked me, even in people inside the company, like, why should we be the ones that are breaking ground on this? Why should we be the ones investing in automating this when nobody else is?

Because we're Howard Hughes. Who else is going to do it? If Howard Hughes were alive, he'd probably be very disappointed that I haven't done it already. I carry that legacy with us. I personally, Do not accept the, is how it's done answer. It's one of the things that people who work with me know that it's just, this is not going to work. I'm not, I'm not interested in the way things have been done in the past. only interested in the best way of doing things. And, and yes, okay. This is coming out of the CFO office where I'm supposed to be more about the stable processes and stable procedures. And, and I am, but I want to have better process and better procedure. So. I'm not giving you a really good answer because to me again, it's not a lot of conflict in my mind. The more efficient I am, the better I am as a CFO. for me, it works. I know that it might not be the stereotypical approach, but for me it works. So it's obvious to you, what are the things that, like, there tactics or uh mechanisms that you use to get others to sort of see the world the same way as you? Yeah, well, I started by thinking like, do I win people over? And that's why I tackle first things that everybody hates to do. And that has worked out so far. So I think it looks like. the first most hated thing in your world that you're like, if I get this, everyone's going to be on board. Yeah, well, leader not like the least abstraction pieces is pretty despised in the real estate. You have to do that. And then we're not tackling cam with we've done a couple other things, but I mean, big, big processes that we haven't deployed yet, because they do need a lot of testing and controls, but we will like the way you procure is something that makes people's life very difficult. And we're about to change that completely very soon. So I m I've realized quickly that it's not just about the technical aspect of efficiency because then you don't want people to sabotage it and people can intentionally or unintentionally they can. um So just win them over by fixing their problems in a way that also solves mine. Yeah, it I mean that makes a ton of sense like if you're gonna it's hard even if you're you know, maybe antagonistic to this process if you're If your life is getting easier because you're going to solve it for them uh It's hard for you to be an anti. I'm curious You know, I often see sort of companies are sort of have these like couple buckets you've got these sort of like Early adopter evangelists are sure all in on this way in the of working there's kind of this big middle where it's sort of people are like, you know, I I'm not opposed to this. I'm not like for it. just, you know, I'm just here to get work done or kind of just go along to get along type. Then you'll have kind of these camps that might actually be like actively resisting this stuff. How do you like, how do you like work with the early adopters to sort of bring these along? How do you sort of, you know, prevent the like, you know, antagonistic folks from slowing you down and then hopefully like move more of that middle group of people into this like full automation mindset? Look, ah I'm not going to lie to you and tell you that we have it all completely figured out because we don't yet. And I don't know that anybody does uh from all that I've been researching and reading and talking to people. But the first group that gets excited about it and what not, we have in we call HHX, which is our innovation group inside Howard Hughes, a really good program to keep them engaged, to get them recognition. basically is to make them feel good for the value that they're creating. And that attracts some other people from the middle into that first group of evangelizers. For the people in the middle, it takes longer. But one of the things that works really well is again, let me show you how you can do something that takes you four hours and that you don't like to do. And maybe you can do it now in half an hour and you win some, but you also have to think about compensation schemes. And this is something that We're not the only ones who are dealing with this, right? In the typical process that a lot of most corporations have, where you're going to get your raise and bonus. How do you incorporate this value creation that people might have generated for the company into that process to make it more institutional recognition that using modern technologies is valuable? And we actually even compensate for that. That takes longer. You can think about that. That's an enterprise-wide cultural change. And we're going through that process. And then ultimately some of the people who are antagonistic, but I think, this is something that I'm sure everybody will deal with.

I think you need to give them opportunities to embrace at least to a certain degree. And if they don't, the reality is that people like that mostly end up self-selecting out. Because when you change the world from under them, they just don't like it. And then sometimes you also have to help them, help them be happy somewhere else. be honest, I don't say it as a euphemism. I nobody should work in a place that we hate. If there are opportunities to go find another job where you can be happy, then maybe sometimes I haven't had to do that, but I'm not, I'm not naive. mean, there is wrong possibilities and maybe sometimes that has to happen. Yeah, this is probably a good point to step back because it seems like Howard Hughes, both the individual and the business and the culture play a big role in how you and the company have been able to sort of navigate this moment. Like, tell us a little bit about Howard Hughes, the individual, the company, the culture he sort of set out that maybe the rest of us can learn from. Sure, a lot of people know about Howard Hughes from movies and other things, but most people tend to stay with the image in their minds of his later years, when unfortunately he was suffering from mental decline and the long hair and not coming out of the hotel room. But before all of that happened, he was the ultimate innovator. This was a person, and I like to emulate this way of thinking, even though I'm not nearly as successful as he was. But this was a person who would look at reality and say, I don't accept that. I am going to change it. And if I don't care that nobody else has figured out how to do it, I'm going to be the one that figures out how to do it. And it started with the drill bit and it only moved on from there. Everything he did in aviation, aerospace. mean, a of people don't know, but he had, he had influence on, on the moon landing. had an influence on many, many different things that have shaped the world we live in. and we don't even know about it, right? But if he had that mentality, and I like that mentality, reality doesn't have to be the way it is. It only remains the same way because we don't do anything about it. And I like to do something about it in the small part of reality that I operate in. And as far as the company, we're incredibly lucky. We have an amazing board of directors, mean, led by our executive chairman, Bill Ackman, Everybody knows incredibly successful. And so ah they challenge us to be better. do our best. And so that's an environment that works. When they're asking you like, be better, be best. Your company is different. We build cities. Nobody else does that. So do it better than anybody else. Challenge preconceived notions. Be the best that you can be. That's the challenge that we get from the board every time we have a conversation. And so it fits very, very well with this culture that we're building, trying to be better at it. Yeah, I have to imagine that mindset is very valuable in a moment like this, where the company is already culturally more disposed to innovation and thinking differently versus, you know, maybe a more traditional enterprise company where it's, eh, let's let somebody else go first on this. Let's wait to measure the risk and figure all this other stuff out first. That's right. And look, there are things where uh we don't think the risk we don't have to. We're not just reckless risk takers. contrary, we had to risk when it makes sense and we can be very financially disciplined and not be risk takers there. But in other areas, if anybody ever visits one of our communities, the Woodland, Summerland, Richland, Ward Village in Hawaii, or the one that we're building outside of Phoenix, the Ruales, you will see aspects that are different from other neighborhoods where you've been. So we've in our areas of expertise and areas where we can lead, we do lead and we want to lead. And we say it in our communities. When you, when you visit one of our cities, you might not even know that it's a Howard Hughes community, but you will feel differently. You will feel that's a special place. And we think that same mentality and everything we do. now during this transformation uh to a holding company, it's incredibly exciting. to think about how we're going to bring the same mentality to the new chapter of the company. How do you juggle? You've got everybody excited about this stuff. The company's ready to innovate. You're innovating like crazy. You're building AI use cases left and right. uh But this technology isn't a silver bullet either. There's a lot of hard work that's required to roll these things out. I often run into individuals who are like, you know, want to snap their fingers and say, solve this big hard, hairy complex thing. And the reality is like, ah, it's not quite there yet.

So how do you deal with kind of the messy middle of all this stuff? You might personally have gotten some quick wins, but I'm guessing many others in the company like... They have to go through their own learning curves as well too. How do you sort of support and guide folks through that? Absolutely. And I've gotten some not-frequence losses too. It's not a perfect process. But yes, I try to address this with communication and education in many different ways from informal communication, from formal texturing committees, from our technology committee of the board. Because sometimes, and I think this is where you're getting at, sometimes it's harder to explain what AI and blockchain are not. than to explain what they are. People think it's magic, right? And you have to say, well, now in order for that model that you saw at a conference and that you think, why can't we have that? There's probably five years of training behind that and a significant amount of money to make it look that seamless. Right? So I can't deliver that to you right now. mean, and sometimes we get technical, sometimes we don't, but in simple terms, we have to try to explain to people. that this is not magic and it's not going to think for you. And that just is that you want, maybe will happen in a few years. Maybe it never will, but it's not where we are right now. And sometimes people also don't realize, and this can get a little sensitive too. Sometimes people don't realize that in some cases they're asking for something where I can deliver or not, there is going to completely eliminate their own role. having a little bit of a kind of conversation of saying, I mean, you really said differently, but do you really understand what you're asking for? I mean, you have taken a step back and thought about it. But most of the time, it's not even that this is what you said that people think it's magic and it's not. And we have to spend a lot of time explaining why we're not saying no, just because we don't want to help you. But it's because it's technically feasible or it's expensive to do. I don't just want to take time. A lot of the things more than anything is that Yes, that can be done, but one, we have a backlog of projects, so we need to prioritize. can't do everything at the same time. And then two, it's not going to happen in three days. That takes training, takes deliver a thought, and you might have it at some point, but not tomorrow. You talked about folks removing their own role. I'm curious, if someone was so successful that they managed to automate themselves entirely out of a job and came to you and said that, what would your reaction be? I personally would love it and I would keep that person and I would beat the role and I would say, now you're working at HHX and you're going to help me automate everything else. Yeah, it's so funny. Like, every person I talk to, that's their reaction. And yeah, I think one of the things that is tricky right now is there is some fear that, you know, hey, you know, am going to lose my job? You got to automate me on a job. And think the practical reality is if you're so good to deliver that kind of value, no way I'm letting you go. Like, I'm going to find the next problem and the next problem after that to go tackle it. Exactly. And then I don't know, I can't speak for all organizations, but we tend to run lean. We try to be a lean organization. So that of people thinking I might be automated away, I'm excited to give them more time in the day to work on other things. They don't know, or maybe they don't know, they forget. We always have a backlog of projects and analysis and research and all sorts of things that we want to do and we don't get to. So I don't see it so much and again, other organizations might be different, but to me, it won't be, my God, I have to take a drastic decision. No, it will be more like, my God, finally the to-do list is really gonna start to get shorter because we have resources to do it. Yeah, awesome Carlos. Okay, last question for you. What advice do you give to other CFOs, other leaders who want to go replicate your success? What are the things that they need to do differently to be able to um do even better than you did? love that question. They're going to have to get comfortable. I would say get comfortable embracing risk. We're not trained as CFOs where you came up as a CPA like me. Maybe some people depending on the input banking and depending on what they were doing, they're very comfortable with risk. But by and large, we're not. We're actually trained to avoid, manage and transfer risk whenever possible. This is one where if you want to be successful, you have to embrace risk in this area. You have to be willing to fail, which again, for math people, it's very hard to say I'm going to be willing to fail. But if you don't do it, you're just not going to reap the benefits.

have to get comfortable with this risk. You failure. Otherwise, I don't think you can make it successful. Can you give an example of what kind of risk that actually looks like for folks? Well, you can look at the different people. But for me, the first risk was when I went to the board and asked for money for something that was completely unproven. mean, think about it. Can we have a CPA? What does this guy know about this? He lives in Excel spreadsheets, asking for money to research and implement blockchain and AI technology. Why did they trust me with that request? I didn't have the profile for it, right? I wasn't the chief information officer. wasn't the chief technology officer. I don't have a background in IT. And here I'm asking for money to basically deploy and research new AI technologies. If I had failed, I mean, I don't know that I would have lost my job, but I certainly would have lost a great deal of credibility. And I didn't, but it was a risk for me. And in other areas where you actually have Google ask for a budget, to deploy an enterprise wide AI application, for example, and you're putting yourself on the line again, you're asking for money that it's not going to be used for something traditional quote unquote inside the company. It carries again, credibility and execution risk that I mean, if you're not comfortable with it, then you have to get comfortable with it Carlos, thanks for joining us on Agents of Scale. For those of you listening and want to hear more, hope you subscribe, follow us wherever you get your podcasts, and enjoy the rest of your day.