AI Only Works for Hotels in This Order: Data, Intelligence, Action - Stephen German, Actabl

Josiah: I feel like there's so many people in the market talking about AI. You and I have had some great conversations around, what does this look like? What should it look like? And you've put together a three-layer framework that not only has educated a lot of people internally at Actabl, but a lot of our customers. So I'd love to talk through it and to give our listeners a sense of what needs to be true and where might AI go. Let's start at layer one, as you described it, and that is the data foundation. I wonder if you could describe a little bit. When you think about AI and starting with AI, why does it start with data?

Stephen: Absolutely. Starting with data, you don't really need to look any further than Claude understands how important it is. When you go to Claude, what is the most valuable thing from it? It's the connectors. It's what can you bring in so that you can act on it? What can you bring in so that you can discover information about it or shape what you're doing? But it starts with that data. One of my frustrations is when I have tools and apps that I can't get the data out of, that I need to go and run exports from, or I'm needing to keep refreshing. It throws off the flow. It throws off the work that you're in. And so the data foundation, actually having all that data ready-made and able to be used, is by far the most important thing.

Stephen: What we have been doing on our data foundation, the core of ProfitSword and its BI, is really the more than 300 integrations that we have that we are already pulling the data in for, normalizing it, and making sure that you can actually compare apples to apples. You're not comparing disparate sources. You're working through one that you can trust because it's what your people are actually using every day. When you have that, you can rely on it, and you can then start to do the much, much cooler, more impactful work of surfacing the intelligence and taking action on it.

Josiah: It's funny, Stephen, because I'm at an event this week and I have had so many C-suite leaders at large hotel companies come up to me and literally tell me they live in ProfitSword, they live in Hotel Effectiveness, they live in Transcendent, because of the data. And I think for our listeners who kind of think about, if you're asking questions about the hotel business in your favorite LLM, it has generic information. Like you say, it's the data about your business that allows you to gather interesting insights. And the approach that we have, you talk about the integrations, the normalization approach, that stuff is so key.

Josiah: So it all starts, I think, for people thinking about where do I begin with AI, it starts with the data. So you need to get your data house in order. That requires you to digitize workstreams. We've been talking about that for some time around, whether it is preventative maintenance. We had recent capabilities roll out with AI asset setup, but it's all about digitizing. So you kind of build your AI house on a foundation of data. The next layer that you've been teaching us about is surfacing intelligence. And I know you are working on a number of things that will give our customers capabilities to know more. Talk a little bit about intelligence. What are some of the possibilities with intelligence that AI unlocks once you have your data house in order?

Stephen: Yeah, so the good news is the intelligence is actually quite a bit easier once you have the data foundation in place. Everyone has that knowledge in their head, that kind of decision tree of if this, then that. When I see this data, when I see this flag, I want to look at a trend. I want to compare, what did that look like to last year? I want to see if this other related component is also affected. I want to see if this is a blip or a trend. These are all the things that kind of run through your head when you see that.

Stephen: But what AI now gives us the ability to do is basically cutting to the chase, not looking at the reports to find that and not having to go through the decision trees. Getting the knowledge out of your head of what you check, what you look for, what is the underlying truth behind the data. That's what surfacing intelligence is all about. Setting up those agents and those prompts that think and work like you, that surface the flags that you actually need to care about. And citing their sources, of course, making sure that you can rely on it. But not telling you your CPOR is growing. Telling you your CPOR is growing because of labor, because of overtime, because of clock-ins that aren't being governed, because of poor schedules. Whatever that is. Getting to that end result is what surfacing intelligence is all about. Being able to spend more of your time actually taking, processing and working on it than trying to figure out what it is.

Josiah: And I think the promise of intelligence, it kind of seems like something that is instantly accessible to everyone. You may have had some experiences with a chatbot where you are getting great answers, but it goes back to your points on the data foundation. Unless that is in place, all the data's normalized, the structures are in place where the AI intelligence can then read off of that and work off of that, it's not going to be very intelligent. And just again, at this event I met so many people who have come up to me and told me about trying to just get business answers from maybe faulty data sets, but also just the parameters with which you're gathering that intelligence need to be correct. So you're building some amazing capabilities, but I think the intelligence is such a key component. Layer three of this is agentic AI. Curious on your thoughts as you think about agents and how does that play into the picture here?

Stephen: Absolutely. So when it comes to actually being able to execute, when you understand what needs to change and why it needs to change, and you have confidence, then you need to go to your system of action. You need to go to the place where you're actually executing, actually making that change. Whether that's your email or your budgeting and forecasting tool, your labor tool, whatever that is. When you find those insights and those things that you need to change, you still need to go into those locations and do it.

Stephen: What we see in the future and what we're developing towards is the ability to let agents go and do that work. Not having to go and find the location, or not having to go log in and find this page, this schedule, this location, but actually being able to tell an agent, this is what I want to do, this is how I want to change it, and just having it done for you so that the action is taking place without all those intermediary steps. And that you're actually getting the benefit of that intelligence, because insight without action is still ultimately meaningless. If you are getting a lot more insights, that is great, but if you can only act on so many of them, you still haven't actually gotten the full value of it. So being able to execute that at scale is the next frontier that we're seeing a lot of possibilities in.

Josiah: I think this is what gets me excited about AI in the hotel business. It can go so much further and deliver so much more value than AI in some other contexts, because hospitality is a real-world business. There's a lot of interactions. There's more than 180,000 hoteliers that are relying on Actabl technology every day to do the work they do. And they're working today off that data foundation. These workflows are making them faster, more efficient, helping them empower their teams, delight their guests. The intelligence capabilities that you are building are going to level up their ability to do that. And with agentic AI, we're using this internally at Actabl for our own work, and you're building a framework that is going to make agents possible.

Josiah: So to wrap this, I think for our listeners today, they need to be thinking about this. Focus on the data foundation right now. And then just get these systems set up. So workflows are also important. Actabl technologies need to be in place, and you need to get your teams on board with this, get the system set up. And with these components in place, then you'll be ready for an AI-driven future. So I'm excited about what's ahead for us and appreciate you walking us through that.

Stephen: Absolutely. And I think the final thing to know is just all of these are virtuous cycles. The old investment adage is time in the market beats timing the market. And that is very true of AI as well. If you have your data, if you are getting those intelligence signals, you can build on those, you can iterate on those, you can add more context to make them more intelligent. And the same is true with the agents. The whole thing becomes that virtuous cycle that just continues to build and improve off of each other. But the reality is the more you can do the sooner, the more time you're going to have to improve it, and the better and better it's going to get.