citizenM CEO on the Future of Hotel Tech, AI, and Guest Experience: Lennert de Jong

Josiah: Since its founding, citizenM has always been a leader in hotel technology. I encourage you to check out some links that I've put in the show notes for some prior conversations I've had with leaders in the company on this topic. But today I'm thrilled  Matthias Hüttebräuker, our innovation correspondent, is back with citizenM CEO Lennert de Jong for a conversation about hotel technology, AI, and what's next. Let's get into it. 

[intro]

Matthias: Is this in general a lesson to be learned for bigger and smaller groups that the leverage that a big group provides should be used rather where we don't see it, and integrating systems and empowering employees and stuff to ensure, as you've done at citizenM, that on a guest contact level, people could be rather intuitive because they're free to so many stuff. And the standards are everywhere but in that last mile where people can actually be people and humans can be humans. And I think that's the function of your whole reverse pyramid and also of the way you see tech and use tech? And is there a broader lesson to be learned from citizenM? Or maybe you want to explain how you did it and how you run it?

Lennert: Well, I hope so. And I'm quite passionate about that. Having built this first check-in experience in 2008 when other hotels were saying, citizenM is a disgrace for hospitality by providing self check-in. Those were also the people that were saying that their guests are never going to make a booking online because they prefer to speak with the reservation staff. So I think the next step, if you think about the hospitality industry as a whole, we're still doing one thing wrong, and I'm talking about the hospitality industry in general, which is that we treat this front desk like this, I wouldn't call it a garbage bin, but the front desk is a place where human beings look at a computer and they can handle all kinds of exceptions and they can handle instructions and routing instructions and folio instructions and VIP instructions and payment instructions. So a lot of people wonder why you go to a hotel and they spend so much time typing behind the front desk and figuring things out because it's a human being that's doing stuff that central organizations did not want to automate. If there's one thing I think you can learn from citizenM is that if you don't hire human beings on their ability to interact with a PMS, but you hire nice human beings, you are forced to automate these things. And so a very simple example, the charging of virtual credit cards provided by reservations. We have bots running that in the background. So that when you get to our self-check-in machine, or you check in through the app, everything is arranged for. And the human beings can just stand next to you and say, hey, Matthias, welcome back. Did you fly from anywhere in Europe today or did you have a longer flight? And why don't you come over for a drink later on? We have fresh cocktails and I'll be at the bar. And that's a very simple example, but yeah, I'm very passionate about that specific topic. You and I talked about it before. I always hated the London Underground. I could not figure out which zone I needed to buy. And then you got this on-peak, off-peak, single, double, whatever. And I always got it wrong, and then you get a ticket out, and you need to stick it into the gate in a certain way, and I always got that wrong. And then at the exit, the gate doesn't open because you got the zone and the on-peak, off-peak wrong. When the London Underground said, we're going to change that experience from a clunky ticket machine to just put your phone down, first the Oyster card and then the phone, or a card, and the gate opens, and it opens again at the exit, we're just going to automatically charge, that was a fundamental differentiating experience that allowed the London Underground to get more traffic. Instead of biking or taking a taxi, I would say, I actually like the London Underground. Now, go back to the hotel industry. At every hotel, it's the same thing. I mean, my last name doesn't really go well in most of the countries. Your last name is even more difficult. You get to the front desk and there's some phonetic recognition of you being a guest, then there's a couple of things that the front desk needs to take care of. So who are you? Can you identify yourself as you? How are you going to pay? And how do I know you're not going to steal from the minibar or when you charge something to the room? And then they need to issue you access to the room. Now all of these things, identity, payment, guarantee, access to something, could be handled through technology. So for me that is probably the part that will hopefully be solved in the hotel industry, that we get an operating system at the central place where people can take care of that and then there's just going to be a human being that is going to be friendly. It doesn't need to be trained. It's difficult enough to find people that want to go to training on PMSs and all kinds of hotel systems. Yeah, that could be a way out of this front desk being the garbage bin of all central operating systems not really working and human beings doing stuff what computers should be doing and AI should be doing.

Matthias: Yeah, it's actually true. We're using humans a lot as interfaces, which is not what they should be doing actually. And so basically what you're saying after introducing the 1.26, I think, the check-in terminal, which I think you introduced as the one minute check-in, now you want to go from one minute to one touch. Is that correct? And all those other companies who are still in 2025 announcing proudly that now they have check-in terminals, they should stop copying that and go right to the next stage. So check-in terminal is just another intermediary step towards the one touch kind of check-in that London Underground has and stuff?

Lennert: No, I'm not saying we're there yet. We already did make a change in citizenM and we're rolling it out over the next couple of months to all of our hotels. If you remember the old check-in terminals, they had a piece of hardware with a payment machine and an RFID reader and a printer and a passport scanner. And now we only have an iPad and a payment terminal. And the payment terminal already takes care of the creation of the room key. If you want a room receipt with your room number on it, it comes out of the payment terminal. And the iPad you can use to put things in. So I can imagine at some point also the iPad can be gone. It's not really needed. I really believe in this, that we're part of Marriott with 10,000 hotels. That's probably the scale that could be interesting to get this piece of technology on the front desk that takes care of guest recognition. Who are you? How are you going to pay? What do you mean for us? And in the background, do, of course, all the analytics, that you also know what Matthias spent in the bar. And I've always believed, and that was my payment excursion, that probably the payment industry or the payment rails, as we call it, is probably the one rails where you can take care of all of that. Identity, recognition, and of course, payment already being part of it.

Matthias: Yeah, I agree. You just said another interesting thing. You spoke about the centralization of tech and if we maybe move to AI, because how could we not these days? Everyone is quick, I mean, for hospitality is there is quick maybe, to invest in some AI enhanced tools. And they're thinking that now that they bought the tool and they're halfway there. But the other half is integrated systems. It's unified data and it's actually the bigger half and most people don't have it. But it seems at citizenM, you've always had this mindset of let's integrate tech and let's unify data and stuff like that. And even though you don't really talk that much about AI these days, publicly at least, I do think that you probably, to a large degree, are already running AI. Is that correct? What do you see the future of, or what does it take to make this step towards AI, which always includes a step towards total digitalization? What are the success factors? What are the scenarios you're seeing for the near future?

Lennert: Yeah, at citizenM we've always had a very integrated approach to things, so I think if you take a look at it from an industry perspective, there's probably three ways that AI is going to play a role in the hospitality industry. The first is this big existential, how are people going to find my hotel question. Because everyone, the consumers are now on ChatGPT and on other AIs and the way that they find stuff is changing. I see it myself. My default browser one day changed, I probably allowed for it to ChatGPT. Anything doesn't go to Google anymore, it goes to ChatGPT. Now, booking.com and Expedia made a business out of listing thousands of properties in a city and having a user interface where people could interrogate. I want a five-star hotel, I want a swimming pool, and I want a room with a view over a river. And then, yeah, you get some result and that whole dynamic. So the first question is, how is AI going to change the way that people find my hotel? And how am I going to pay for that? Is it just going to be that everything that we've seen so far, it's going to transact through the OTAs anyway, or might it be an opportunity? And how is that going to work? The second part I think is what we're working on, when you have integrated systems, you can start implementing AI yourself. So we have Gen AI running all of our FAQs, everyone that asks us a question, whether it's for WhatsApp, through the messages, through email, AI answers it. 96% of all booking.com queries are handled by AI, generative AI. And then the second part that we now start to run is Agent AI. And Agent AI is then making the next step, is Matthias stayed at citizenM. Actually, you didn't pay attention to your email, so the invoice you can't find. So you ask us, can I get an invoice? Now, that is a physical human being logging into the hotel PMS and sending you the invoice. Well, that we are solving with AI as well. And we are investing a lot in these repetitive processes, just coming up with as many use cases where there's a lot of volume behind it to solve it through agent AI. Now, if you are an independent hotel where you're operating on an Apollo or a Muse, there's also a lot of out-of-the-box AIs. And you have runner.ai and a couple of these, and there's a marketplace in Apple.ai where you can choose an AI. And I think that's the third use. So if you don't want to figure it out yourself, then find some micro cases and make sure that you solve that. It could be in the check-in process. It could be extending the invoices, pre-stay communication, finding a restaurant. So I would say those are the three things that we probably see in the industry.

Matthias: Lennert, you just said, and I want to emphasize that, you said there will be different ways of finding hotels. Now, these days, a lot of people talk about the shift from SEO to GEO, Generative Engine Optimization, but I feel, and you seem to also say that it's going to be more than that. I recently listened to Sundar Pichai on the Decoder podcast, he said that just like with a shift to mobile, there will be a revolution in interfaces. There will be totally new ways in which we interact with information or with people or with communication and systems and stuff. So do you see or which dramatic changes in interfaces do you see? For example, will you, because you mentioned booking and stuff, will you think that we're not only will be interacting with the distributions platform differently? But do you see some dramatic change or reshuffle in the whole distributional landscape, for example?

Lennert: Yeah, I mean, I'm not a wizard here, Matthias, but what I. But you're a visionary, which is pretty much a thing. Yeah, well, let's see about it. But I think the one thing that is really going to change is the user interface, how you find stuff. And I was at an AI conference a couple of weeks ago and somebody made an analogy and if I would have remembered the person, I would have given credits to him, but sorry for that. But he compared it to the good old travel agents in the old days. You would not be able to find, between hundreds of hotels in Amsterdam, which hotel you were going to stay at. So you would walk into a travel agency and say, dear, Mr. Travel Agent, I would like to stay in Amsterdam. And he would not give you hundreds of options and all kinds of filters and look at hundreds of pictures and thousands of pictures and reviews. He would actually say, based on my experience, I think these are the three hotels that would be really good for you. Then we got the internet that allowed a lot of more properties to come online, the democracy of selling your rooms and the OTAs took advantage of that. So they said, we are the place where you can find all hotels. The one thing that crept into this business obviously is, how democratic is that process? Who shows where? Are they really showing the best options or is it by sponsored results? And there's a lot of commotion about this direct booking war, of course, in the hotel industry and lawsuits and booking.com getting penalties in Switzerland for unfair commissions. But they've been moving around hotels up in the ranks and changing their algorithms. What I think is that AI is probably going to redemocratize your search results. And they're probably going to be more relevant to really what you want. Does it take out an OTA or a bad bank or a big chain? I don't think so. What it probably will do is it will make the best hotels come out a little bit better without having to spend a fortune on it and hopefully give more relevant search results, but we're not there yet. I mean, ask ChatGPT, I want to stay in Amsterdam, which hotels can you recommend? It's not as great as uploading the results of your doctor on the ChatGPT and say, can you explain this in a way that I understand, that it does a much better job at that than democratizing search results. But I believe that's where it's going. And then if you're doing a good job, then yeah, probably what we've seen is that hotels with a high review score, already got more rewarded than hotels with a low review score. And that probably is going to extensify because ChatGPT and other AIs will cut through the chase and quickly evaluate everything that is out there, booking.com, Expedia, Google reviews, TripAdvisor reviews, and look at all the semantics. So based on what I know about you, your family traveling with two kids, 10 and 13, this is probably the right hotel for you. I think that's where it's going to go. And then the other monetization of it is going to be probably a little bit tougher on your own direct channel, if you're a 40 room hotel in the city, as if you're part of Marriott with good direct integration. So the OTAs could probably also play that role. If I would be booking.com, I would be developing that search function.

Matthias: Interesting. But, but as you said in the beginning, not of, about not being a wizard, probably. As with mobile, I kind of think we will probably be the biggest changes that will come are the ones we're not seeing yet. So probably just like mobile didn't just change how we use the phone or access the internet. It basically, and I think Apple used the tagline on the very first iPhone. I think they said this changes everything and it kind of did. So at some point, AI will change a lot of things that we don't even think about right now.

Lennert: No, I'm probably going to interact with it. We don't know yet. I mean, I bought these META glasses a couple of weeks ago and I must say, I had these Google Glass glasses as well a few years back and they were clunky, but the META glasses are pretty awesome. I was cycling through on scootering through Washington and I asked META, can you make a picture of this building? Took a picture of the building. I said, what building is it? They recognized it as a historic museum. I said, what's the art on the building? Who made it? And it was talking in my ear about that. And later on, you can look back at all that stuff. But yeah, I find that those kinds of uses of AI are actually pretty useful.

Matthias: Yeah. Yeah. So, so actually coming back to what we discussed a couple of minutes ago about citizenM superpowers of agility and stuff. So, so we're basically entering a time as I tell you, basically in every industry, but also in our industry, where we are extremely sure there's going to be a lot of change. We don't know exactly how that change is going to look like. So agility is important. And the ability, what has made you guys so strong to change big? To understand this is eventually not only about automation. This is eventually probably not only about process re-engineering, but eventually it'll be about business re-engineering, new business models coming up, new ways of doing business.