How Hotel Companies Turn AI Into Competitive Advantage - Steven Moore & Joseph Benjamin, Actabl [Sponsor Bonus]

Josiah: Recently, Actabl announced an industry first when it comes to AI, and that is an offer to embed its AI engineering talent within hotel companies. This model, called forward-deployed engineering, was pioneered by Palantir and used by OpenAI and Anthropic to accelerate time to value of AI capabilities. In this episode, you're going to hear Actabl CEO Steven Moore and Chief Technology Officer Joseph Benjamin talk about the need they saw and what Actabl is offering to meet this need. If you care about generating results from AI and you want to do it in a way that preserves what makes your hotel company unique and special, this is an episode you won't want to miss. I've included some chapter markers if you want to jump to a specific part of the conversation. So without further ado, let's get into it.

[intro]

Josiah: I'd love to just kick off with you, Steven, because you've shared this before, but it's obvious that as CEO you have a really interesting vantage point into what's happening in hospitality. A lot of conversations, a lot of rooms you're in from different market participants. At the highest level, when AI comes up, what are you hearing? What's coming up? What are people wanting?

Steven: I've been having a lot of conversations. You ask to set up a meeting about AI, and nobody turns you down. It's the topic of the moment for sure. And all of those conversations have similar themes. I think there's a strong conviction that AI will transform hotels, hotel operations, the guest experience, the associate experience over the next several years. The conviction is more loosely held on exactly what that looks like. So people have ideas and talk about some fuzzy crystallizing vision, but it's not exactly clear where everyone lands, and I think that's okay. So strongly held conviction on the transformation potential, but loosely held on the execution.

Steven: I will say that the most advanced executives are all moving forward with the foundation, with the data. Regardless of what this looks like from an insights perspective, from an agentic workflow perspective, you have to have great data underlying this entire strategy or this entire opportunity. So there are certain things that even if you don't know what the end game looks like, you can take the first step. You could put those things in place today, and executives are moving fast on that. They know that they need to get that in place in order to be successful in whatever this looks like over the next few years.

Josiah: Interesting. I want to get your take on an element of that. You say AI's hot right now. We talk about AI, people want to take the meeting. We just had the NYU International Hospitality Investment Forum. There was the latest data sets where it seemed that the environment we are in is persisting in the sense of top line growth is nominally increasing, but costs are outpacing inflation. And profit margins continue to be under pressure even as top line has been re-forecasted up in the last couple of months. I want to get your take on, are some of these factors related? Is your sense hotel executives are turning to AI specifically to address these cost pressures and the pressure on profit?

Steven: Yeah, I think we've been given this incredible tool and this incredible gift to fundamentally reset the profitability expectations of hotels. Not just recover the compression over the last several years, but reset the floor even historically for hotel profitability. That said, given the environment and given the nature of hotels, hotels are not VC-backed businesses. There's no token maxing, there's no unlimited AI budget, there's no, "Oh, we can burn millions of dollars every year seeing what works with AI." So even though it's a gift, a tool that can unlock profitability, there's a pretty high bar in terms of being thoughtful on what problems are you going to work on, what opportunities are you going to go capture, and how are you going to execute those things.

Steven: So the bar around immediate ROI, partnering with someone that can help you realize that quickly and do it with minimal experimentation where you're burning a lot of time and resources, is critical. So yes, there's pressure. AI can help relieve some of that pressure, but you've got to be really specific and really thoughtful about how you unlock that.

Josiah: I'd love to hear you talk a little bit more about resetting the profitability floor. The idea of more profits flowing into hotels has ripple-on effects that benefit everyone in the ecosystem, so that's great. When you talk about the floor being reset, though, it seems like it's not just the outliers, which feels especially important for hotel companies that have large portfolios, where there typically is a distribution in terms of different performance levels. So when you talk about resetting the floor, is it sort of like there is this opportunity for hotel leaders to adopt this such that there's performance improvement across all of their hotels, and it's not just a star GM on one property that's doing something great?

Steven: Yeah, I think that's exactly right, and the star GM is a great example. It's just one of many. But imagine taking your star GM and replicating them in a digital form, in an AI-supported form, in a coach, in an agent, however you want to think about it, and then scaling that GM across every hotel that you have. So even your newest, greenest GM has this coach next to them all the time, that they're not anxious about asking dumb questions or asking too many questions. They can interact with it. That barrier is really low to help them drive profitability at the hotel. So that's just one example. It's a really interesting one to think about.

Steven: But I think that's exactly right. The profitability floor across an entire portfolio, not just a specific hotel where you're going in and doing something absolutely exceptional that you can't scale, but the ability to take a scalable platform and build AI into it, AI on top of it, that regardless of the chain scale, the location, the ways of working, you can tailor something to your unique needs that, in my mind, fundamentally increases the profitability for every hotel going forward.

Josiah: Amazing. That's a great segue. So speaking of resetting expectations upwards, technology platforms, and tailoring, I want to pivot to you, Joseph, and talk a little bit about this notion of forward-deploying engineers. What is this? There is so much going on in terms of AI. At Actabl, we're introducing so many different capabilities, but I want to focus this conversation on this notion of forward deployed engineers, starting out with defining it. A lot of people have never heard about this term. What is it? Where does it come from?

Joseph: Sure. I think the simplest way to say it is that our Actabl engineers work directly with the hotel company's team to solve business-specific data or AI problems, as opposed to building features and rolling them out in the product. So it's a very different model than that, where we start with what is the problem the customers are trying to solve, leveraging our data and potentially leveraging AI. And we're building that on top of the Actabl platform and the hotel data foundation we already manage for them.

Joseph: So it's more of a natural extension. Instead of starting from a blank page, we start with the customer's operating data, which in most cases is Actabl data, and work towards a specific business outcome with them. As we've been talking to a lot of different customers and listening, for some customers that may start with just a better data foundation, standing up a data warehouse so that they have a place to work off of Actabl data as well as other external data sources. For others, it may be more advanced analytics, forecasting, or an AI-enabled workflow. The common thread is that it's built around what is unique to that customer's business and expands on and leverages Actabl's data foundations.

Josiah: Interesting. So this is one part of an ecosystem. There's the data foundation that you spoke to. As you mentioned earlier, Steven, this is what I continue to hear CEOs talk about, that this is table stakes, this is where it gets started. There was just the Skift AI and Data Forum, Marriott, Hilton leaders all talked about this is where you start, this is how you get the advantage. So there's that. And then Joseph, with your teams, you've been designing and building different offerings to accelerate. So there is AI asset setup that makes digitizing a lot of work streams faster by just getting on board. So you have a standardized data foundation, you have these new AI offerings that are speeding up that process of getting started. You have AI offerings that are providing insights to the people on property, a new product, Altitude, that is providing insights above property, and this new offering allows for the application of what this company needs that's unique. I wonder if you could speak a little bit about the origins of this idea, because the most advanced companies in technology today have been using this.

Joseph: Sure. I think in tech circles, it's becoming very popular, and it's certainly not something that Actabl invented from scratch. Palantir was the pioneer in this model, where they really proved out this forward-deployed model where they would have engineers work very closely to the customer's operational problems rather than building software in isolation. So instead of "build it and they will come," it's "let's build what you need today." That was really the model and the mental shift that Palantir made popular. And more recently, you see a lot of the AI companies moving in the same direction, because what they found is it's great to have the models, but the AI implementation, how you leverage these models in these ecosystems, depends very heavily on context, on data, on what problem specifically you're trying to solve. So you really need to understand the data, the workflow, the business objectives, and how the output will actually be used, and bring all that together in a solution that works for the customer. And that increasingly is less and less prepackaged software and more of a combination of some of the building blocks that you just mentioned that we're building here at Actabl.

Josiah: So I'm going to get both of your takes on this, but Joseph, first I want to get you to expand on this a little bit more. It's interesting to me because it's not really a last-mile approach, but it is sort of like this. It's interesting to look at Anthropic and OpenAI with tremendous tech capabilities. When they think about how do you actually affect change, you need a function like this. I want to hear you both talk about this, but Joseph, you've led huge technology organizations. What do you think is going on here? Why is this offering needed?

Joseph: I think there are a number of reasons. The power and capabilities that the AI models, these LLM models, have unleashed is really unprecedented, has really opened up the surface area across all companies. So what traditionally required a software vendor to build for you now becomes more and more easy for you to build with the right data foundation, and to build a lot of these workflows and solutions independently. So there's a huge opportunity now that really didn't exist or wasn't as feasible two years ago, to build out these custom solutions at a lot less cost and a lot faster. I think that's really the unlock that has accelerated a lot of this movement towards this forward-deploy model.

Josiah: Interesting. Steven, I'd love to go to you and get your take on this, but maybe taking a step back and also getting your take on where this fits in the ecosystem, because there are so many new AI capabilities Actabl's bringing to market here. How do you think about that, the whole ecosystem of offerings and where this fits into that picture?

Steven: Yeah. So it's not a replacement for the embedded AI capabilities in the product. We're going to continue innovating in each of the products, from onboarding to using the products to the insights, the reporting. There's a lot of embedded AI features that exist today that we'll continue to innovate on, so it's not a substitute for that. Like Joseph said, this is for challenges and opportunities that are unique to specific companies, the operational challenges that you need to get closer to to solve.

Steven: And I think the reason this exists and the reason it is especially useful with AI is that AI is a tool, and you need to know how to use the tool. It's not just a tool, it's a new tool that is changing every week, month, day. It's changing rapidly. And on top of that, it's a new tool that is unlocking you to think about new problems. Problems that you may have not even considered being able to solve before. So that's where this FDE model comes in. It's not a substitute for the embedded AI innovation in the products, but there is a need to get closer to those unique challenges that AI is now powerful enough, fast enough, adaptive enough to fix, and that requires some partnership.

Josiah: I love it. I would also love, Steven, if you would speak a little bit to something that you've been sharing with our teams here at Actabl, and that is the dynamic that you see as CEO with CEO conversations you have in other hotel companies, where you have this baseline of 80% where there's going to be similarities between companies, and then the special sauce. I would love to hear you talk a little bit about that, because this might be a helpful frame for people to wrap their heads around this.

Steven: Yeah. It's an interesting, maybe philosophical observation. If you look at every hotel across the entire world, I'd argue that 80% of it is the same. It's a building, there's a room, there's someone to welcome you. You want that building to be safe, you want the room to be clean, you want the person welcoming you to be hospitable. That's pretty consistent and standard across every hotel across the entire world. But the thing that makes hospitality magical is that last 20% where you differentiate. You try to differentiate from your competition, or for certain guests, or in a location, and that could be brand, chain scale, ways of working, culture, any number of things that you choose to differentiate yourself on.

Steven: And so that 80 and that 20 add up to 100 in one hotel. But if you look at the tech space serving hotels, you have this split where you have vendors that may be serving that 80% standardized across every hotel in the world. So it's like, great, that's a really scalable solution. There's a lot of benefits to that standard operating procedure, making sure you have consistency. That is a good thing. And then on the other side, you have some tech vendors that are focused on, well, we're going to build something that's perfect for this one hotel, really addressing that final 20%. And that's a great thing for that one hotel, but you try to scale that across a portfolio or a number of hotels, and it just tips over.

Steven: So you kind of have these two categories, either serving the 80% or serving the 20%. And I think with AI, again, the power, the speed, the flexibility, the adaptation, you can bring that technology to match the reality of a hotel, where you need something to be scalable and a foundation that you can use across an entire portfolio. But at the same time, you have this last 20% that you couldn't unlock before that you now can with AI, and I think especially with this FDE model. So I think it's an incredible moment where technology can actually step into the reality of hotels that we've struggled to serve in the past.

Josiah: There's some good stuff that happens in that 20%, a lot of good stuff. But I'm curious, do you see some problems that live in the 20% that are important to think about as we think about this?

Steven: Yeah, well, part of the 20% is the problems that are in that 20%, but also the speed and the accuracy at which you address those problems. It's different to say, well, we're going to have a human go in, look at a dashboard, or even set up some type of threshold, identify what the problem is, try and calculate what the impact would be of solving that problem, and figure out how to go solve that problem. So from an insights perspective, it's like an always-on monitor, supervisor, all-star analyst that is looking at trends of what you've deemed important to your business, and identifying trends that you may not know are important to your business yet. So there's some real power in the insights, and not just the insights themselves, but the speed and accuracy of those insights.

Steven: So that's thing one. I think thing two, the really exciting part, is the agentic workflow that's on top of those insights, or that you could potentially build on top of those insights. So agents are not going to be helpful if you have bad insights that's coming from bad data. So this goes back to, you need the data to generate the insights to build the agentic workflows. But you think about all of the things your teams are doing, both above property and on property, in a digital world that could be automated. So you free those people up to do what only humans can do on property in a physical business. That is the exciting thing that I think you can unlock in this last 20%.

Steven: So that's how I think about the 20%. And going back to the uniqueness and the differentiation there, hospitality is fundamentally a people-driven business, and people are unique. They see the world differently. They operate differently. And that amplifies when you bring in teams of people. There are combinations of different personalities. So I think that's one of the most beautiful things about hospitality, but that just leads to different ways of operating, different ways of observing, and different ways of working. Those nuances are things that I think AI can tailor around to really amplify what makes your business great.

Josiah: Love it. Joseph, would you add anything there?

Joseph: Yeah. More of a tactical answer is, every company, every management company has to manage labor, profitability, service, assets, reporting, just overall execution, and that's where we have great products to help across all those things. But the other 20% is the ownership model, the management approach, portfolio strategy, forecasting processes. That's the 20% that isn't consistent across every hotel management company. Another example is above property and how above-property teams want to make decisions. So on the more tactical level, those are the differences we see, where we solve it 80% with the software, and then this 20% is where we can do new, unique things per customer.

Josiah: I imagine there's people listening to us, and they're saying, "Hey, we've built something really special with our culture, our way of operating. We have the special sauce here that sets us apart in the market. We want to make sure that we don't miss out on the innovation that's happening. We need a partner that's going to help us get there." There's a lot of consultants out there, different technology companies, people offering different things. I wonder if you could speak a little more to what makes this offer from us, what makes Actabl Engineering uniquely positioned to help hotel leaders execute on this?

Joseph: It's an "and." It's extending what our core platform already offers. Consultants can help, but if you were to bring in another consultant to extend on top of the Actabl data set, they can help, but they'll have to spend a lot of time understanding the data, the systems, the operating context. One of the advantages we have is we already work inside the operating environment. We have a lot of subject matter expertise within hospitality. We have the data foundations, a lot of the definitions. We understand the workflows and the edge cases. So we're uniquely positioned relative to a consultant to come in and work from day one to understand and figure out what that 20% is that our specific customer is trying to solve. So we're really applying this forward-deployed engineering model to the hotel industry, where we have a lot of advantages because we're already in the space, we understand it very well, and we have the data to expand and build on.

Josiah: Interesting. And then how do you see this offering as supporting a more agentic AI world? Are there components in the technical build that lay the groundwork for this?

Joseph: I think some of the things that we're building into our products, such as conversational AI, being able to ask questions of your data, and some of the building blocks on that, being able to generate insights and some of the knowledge and capabilities we're bringing around that. If you just ask an LLM an unconstrained question, you'll get the wrong answer 30% of the time, even with the latest models. So there's a lot of work and understanding and expertise required in terms of what context you provide the model, how you ask the question to get a predictable answer back. So those are some of the AI-specific capabilities that we're ramping up on very quickly in our own products that we can extend to some of these other different use cases.

Josiah: Awesome. Steven, the last item you had was around why can't standard software deliver what needs to be done here. I don't know if you or Joseph have any other thoughts on that we want to talk about.

Steven: One of the benefits of using AI and the ability to tailor and be flexible is that you really are enabling a toolkit in your organization. We know that hospitality is a relatively high-turnover business, and so you're building software for someone and their way of working, and "Oh, I want to see this data this way and ask these questions, and these insights are important to me, and I want to set things up this way." If you're building that for a person, again, back to trying to be perfect for someone, that's not going to scale. But if you're able to provide a natural language interaction where you can see exactly what you need to see, when you need to see it, how you want to see it, then it is not as painful if somebody moves into a different role, if they move to a different property, if they get promoted, if they leave the company. You have the next person that comes in. You don't have to redesign the whole system or the whole software. You're able to let that person set up exactly what they need on top of that foundation with AI. So the flexibility in a high-turnover, high-cost-of-training business is really powerful as well.

Joseph: And a standard software product has to solve patterns that repeat across many customers. That's critical. That is how we serve the market. But again, when the hotel wants to build something specific to how it operates, that's where all of this surface area is now available for us to work, not only as a strategic software partner, but also as a data partner as well.

Josiah: Amazing. So people listening to us, they're interested. You touched on this a few different times in different ways, but the next step is a discovery conversation for the reasons that you both have shared. This is tailored. This isn't just a set, off-the-rack kind of menu of things that you choose from. It's a conversation. I wonder if you could speak a little bit to that, Joseph. Someone who's listening to us, what might they expect in that conversation?

Joseph: Yeah, it's interesting because it's a different mindset than a typical product company. So we're really there to listen and understand where the customer's at, what they're trying to solve, what are the biggest pain points that our existing products as-is don't solve today. That could be a number of different things. So it's really listening, making sure we understand the problem and the outcomes that they're trying to solve, and then recommending where we can help and where maybe we can't. It's a very honest, transparent conversation, and it can run along, there's such a big product space here, a space to innovate on, that it can just take a number of different places. So we're really coming in to listen, learn, and then recommend where we can help and how we can accelerate some of those outcomes or solve some of the pain points.

Josiah: Love it. Steven, I'd love to go to you to close us out. For hotel leaders listening, what would be some closing thoughts or recommendations you have to take advantage of this opportunity?

Steven: Yeah. So at Actabl, we use this phrase a lot, start by starting. So just get going. And Joseph talked about the discovery process. The investment in that discovery process is really low. Your downside risk for having a conversation of how a forward-deployed engineer working with your teams on your specific operational problems can build some AI solution to help you win, it's some time, and maybe you walk away with some learnings. That's the downside case of this discovery call.

Steven: So I would just say lean into the discovery call. You're going to learn something. It's going to be something that will inform how you think about your AI strategy, and I think the upside potential is pretty significant. So there's some asymmetry there. Downside risk is really low, upside potential is really high. And then I would say, do it with a partner that focuses on tech, focuses on AI, and both of those things in hospitality. So honestly, if you're a customer, you're thinking about, "Hey, can I do this in-house? Do I bring in a third-party consultant?" I would propose that partnering with a vendor that's in the space, on top of products that you're already using, that knows your business context and is launching this FDE offering is a really good place to start. So I would just say, start by starting. We're excited to have the conversation.