Good morning,
This week’s Stratechery Interview is with Booking Holdings CEO Glenn Fogel. Booking is one of my favorite companies in the world: they are a textbook Aggregator. And, in a fascinating twist, they are also among the best at managing Google, the biggest Aggregator of them all.
In this interview we discuss both dynamics, starting from Priceline’s acquisition of Booking, the evolution of their business model, and where the company is going in the future. Along the way we discuss network effects and just how powerful they are, the challenges of building customer loyalty and being a destination site, and how AI might finally fulfill the longstanding dream of recreating the travel agent experience.
As a reminder, all Stratechery content, including interviews, is available as a podcast; click the link at the top of this email to add Stratechery to your podcast player.
On to the Interview:
An Interview with Booking CEO Glenn Fogel About Travel and Aggregation
This interview is lightly edited for content and clarity.
Background
Glenn Faujelle, welcome to Stratechery.
Glenn Fogel: (laughing) Fogel is the name but I will accept.
I asked you and you said you wish you had what the French call you.
GF: My wife would change the name if it was Faujelle, instead she doesn’t. She goes by her maiden name.
Well, it’s appropriate because given Booking’s European roots, but we can get to all that in a moment. I’m very excited to talk to you, I love Booking, I feel like it’s a company I should have written about more. There’s just never been that right moment to really nail it, but you mentioned that you’re a reader, it fits so well with so many things that I’ve written about, all which I want to get to in this interview. But before we get to Booking and get to why I’m interested in it, I do want to start with you and your background, where you grew up, what you thought you would be, and if it would be someone who wants to make his name French because he’s such a big deal on the European continent.
GF: Yeah, so I’ll be brief. I grew up outside New York City in the suburbs, very nice suburbia growing up. I have parents who were very into education, they didn’t have college degrees, but they believed the importance of going to school, so I worked very hard. I was 17, a junior in the spring and I was looking forward to in the fall applying to colleges and I had unfortunately a very serious stroke, paralyzed me on my right side, wiped out all my language capabilities, and spent some time coming back up and ended up going to college.
How long did that take to recover from?
GF: Well, actually because it was the spring of my junior year, I was able to go to college in a year from the following fall so it was a year and some months, I did go. But I worked extremely hard there, I got a degree in finance at what was the University of Pennsylvania’s Wharton School as an undergraduate, graduated actually early to save some money for my folks.
Started working but not in finance, I was working in IT, I was at Morgan Stanley in what was then called MIS, it was their IT shop and that in 1984, but we still had punch card machines. We didn’t use them, but they were there. My first job was putting tapes on drives. Memory used to be in tapes — many of your listeners do not know anything I’m talking about — turned out, ended up doing coding, wasn’t great at it, said I should do something else.
It’s the ’80s, M&A was booming, I said, “Well, I’d like to go over to the other side at Morgan Stanley where all those merger and acquisition bankers were making good money”. Can’t just jump from being an IT guy even though I had the finance degree, so I went and got a degree at Harvard Law School. Came out, no intention of being a lawyer, started right at an investment bank called Kidder, Peabody & Co. Rode it into the ground, went eventually went to nothing, got sold to Paine Webber, at which point they let go most of the bankers. So I lost my job, and while was not fun, it did give me the experience of having been let go.
Right.
GF: So I have that credential.
Something you had to be the other end of many years later.
GF: Right. Met a woman and eventually she said, “If this relationship’s going to go on, you should get a job”, so I end up being a trader at Morgan Stanley and I did that for a couple years in the late ’90s and the Internet’s booming and I’m saying, “Wow, that seems a lot more exciting”, and I have a background in IT, I got a background in finance, and there was this job at a company called Priceline.com, which was really the darlings of the Internet bubble, so to speak.
Right, one of the most insane IPOs at that time.
GF: Exactly. It went public in the spring of 1999, right away over a $30 billion valuation, which back then was real money. I didn’t join until February of 2000, I had gotten the job at the end of ’99 and I said, “Listen, I’ll come over, but I just want to get my bonus for Morgan Stanley, they don’t pay you until February, so I got to stay until February”. They said, “No problem”, so I leave my trading job end of February, one week before the NASDAQ peaks, thereby proving that I should not be a trader, having just top ticked the Internet, gone long Internet.
So I did that, and the company’s worth about $15 billion at the time and we went down over towards the end of the year to just a couple hundred million dollars market cap. Stock eventually went down to a dollar a share, had to do a reverse split to get it up.
Was it a good thing you got that bonus to sort of carry you through?
GF: Yeah, it certainly, certainly was.
Turns out your stock grant wasn’t worth very much!
GF: But imagine what my wife was saying to me. “Let me get this straight now, you left a great paying job, a real good job at Morgan Stanley to this high-flying company?”, but it worked out okay. So the stock on reverse split, $6 a share, way back then, I’ve been there now 25 years and while we’re not quite up to $6,000 a share yet, it’s not far. So easy math almost 1,000 times in 25 years, so it’s worked out okay.
Yeah, you’re still married, so I think that’s the biggest endorsement that it all worked out. You’ve talked in other interviews about that stroke and how it grounded you through difficult times and to your point, Booking’s had a few of them. You go back through the Dot Com crash — well, I guess Priceline I should start out with, we’ll get to the Booking portion in a moment. The Dot Com crash, obviously there was 9/11 shortly thereafter, not great for travel, there’s the Great Recession, COVID-19, I mentioned the layoffs bit before, Google’s SERP changes. Okay, maybe that one doesn’t quite fit with the other one, so we can get to that one another time.
GF: No, it does.
But how did that shape you? And balance, clearly you wanted to chase the big thing, you want to take some risks. What benefit did that bring to you in your leadership style?
GF: Well, you listed a bunch and there were so many other very trying moments, many of which we’d have to fill up the entire discussion to go through all of them. But the one thing that’s consistent among all these things is, and it’s the idea that if you’ve been through a crisis, you’re better prepared to deal with the next one. Having actually been in the arena in fighting the fight helps you understand that when you’re in the next one, stay calm, understand what your options are, how to play it. Minimize where there’s potential losses, try and see what other avenues can you go down, maintain flexibility.
So again, it’s one of those things when things seem to go all to hell to say, especially the most recent, that pandemic, and there were a lot of people in panic. I’m like, “Hey, been there before, let’s not panic”. In fact, we’ve been through that whole COVID thing, the SARS-1 issue back in 2003 and we’d actually seen the effect of that before. So it was like, “Been there, no, it’s okay, we’ll be fine”.
Yeah, that’s a great perspective. I mean it’s funny because I was in Taiwan, went to Taiwan in 2003, and so I was aware of that at the time, it had sort of died down at that point, but it’s interesting to tie it back to that. You’d seen major disruption, people scared to travel.
Priceline
I did want to ask about Priceline, the initial model of which was, “name-your-own-price“, how insane was that in retrospect or do you still think it was a good idea?
GF: Well, so the idea of coming up with an opaque process, meaning opaque, you don’t really know what you’re buying, but you’re willing to do it to save a lot of money. Right away that’s not a bad thing. Now if we look in economics, the idea of consumer surplus or supplier surplus, and you see what you end up where you have people, consumers, who either are getting a benefit — they’d be willing to pay more for something, but the price is not at that level. The idea was trying to match up, capture what the person was willing to sell it for, is very good when you have a product you’re selling with very, very low marginal costs.
Yep, and somebody needs to stay at a hotel, or they’re not going to make anything.
GF: Or an airline, it much more was the airline. So that airline seat, the marginal cost of that seat is essentially zero.
Right.
GF: So if the airline can fill that up with something without impacting its regular distribution channels because you don’t want to have any type of price cannibalization, and the way that it came out wasn’t a bad idea at all. The problem really was is there aren’t a lot of verticals that that really worked for, and people, for whatever reason, they just don’t like the idea as much about putting in a bid with whether you get no chance to back out. If you put that bid in and somebody accepts it, you got to pay that amount and you’re on and there’s no refund.
That’s really interesting because I would think that a big problem with that is it’s offloading a lot of burden onto the customer. It feels very weighty. “Did I put the right price?”, “What do I do?”, “Should I be too high?”, “Should I be too low?”, or is that part and parcel of the fact you can’t back out, you’re locked in once you commit? Or are those two different threads?
GF: Well, I think there are two things, but I think they go to the same point.
Right, it’s too much cognitive weight on the customer.
GF: It’s tremendous. And the thing is, human beings are odd. It’s interesting, one of the things that bothers people almost as much as overpaying is not overpaying but paying more — even though they’re comfortable paying it — more than what somebody else got.
The sense of fairness, it permeates everything.
GF: But it does because when somebody says, “What is being rich?”, being rich is having more money than your brother-in-law. It’s all relative, and that’s one of the issues too.
So in the end, the idea of bidding, not great, but the idea of having opaque pricing with some type of trade-offs is still used a lot. We still have those types of things and people are willing to do it and let’s say we see it in many different instances where people are willing to take a trade-off in terms of what the quality of what you’re buying to get a better price.
So we’re going to get to AI in a moment, but as long as we’re here, would AI have made that work better by reducing the cognitive load? Or maybe another way to ask this is will AI-driven variable pricing work in the future in a way that makes customers feel good? And not feeling like, “They know who I am and so they’re screwing me”?
GF: It’s very interesting to think, “How will the world of the future be if everybody has an agent and the agent is able to instantly be always checking for prices?”. And at the same time, the seller also has an agent that’s always looking at what is the current price and how does that work? As markets get very, very transparent, it’s going to be very interesting.
Now the topping of all this though will be regulations because it’s also interesting where again, this idea of fairness, and which is an interesting concept, of regulators coming out and saying, “Well, is that fair or not?”, where you could have two parties, the seller and the buyer, both feel it’s fine and they’re happy. But the regulator says, “That’s not a fair market”, and starts putting in rules. So I can’t predict really what’s going to happen, but I agree with you, it’s going to be very interesting how it plays out.
Booking and the Agency Model
So when you bought Booking in 2005, what was your mindset when you bought the company? And the reason I asked, did you just think you were expanding internationally buying another travel site? Both you and Expedia were being very aggressive at that time, or did you realize you were buying an entirely different business model? I’d love for you to explain to me the difference in how Priceline acquired rooms back then and what Booking did and why that ended up being a big deal?
GF: Yeah, it’s a shame we don’t have enough time to really go into all the nitty-gritty.
Oh, no, I actually think this part is super, super interesting. So spend as much time as you want.
GF: Okay, so we have the name-your-own-price product at Priceline, and I come in and I’m corporate development and we want to expand internationally with name-your-own-price. So we’re doing it throughout the world, different ways we create different structures and one structure was we set up this joint venture type operation in Europe, it’s Priceline Europe.
Terrible timing, 9/11, and then you have recession and we pull back, etc. so it’s a very, very small operation now in Europe and we let go a lot of the people and I am now actually in charge of that. And I’m saying, “This is the craziest thing in the world”, because the idea of name-your-own-price to get a flight from London to Paris when you can go on Ryanair and EasyJet and pay five pounds, you’re going to save 50%, so it’s £2.50, really? Who’s going to do that? You end up with Frankfurt. You go London, Frankfurt, Paris, nobody’s going to do that. We got to do stuff the old fashion way, in addition, hotels. So we stopped the flight stuff, that’s ridiculous, but the hotel thing is okay. Name-your-own-price for hotels, but still people like to know the price and they like to know what they’re getting exactly.
So I’m looking around, who knows how to sell like that around here? And find a little company called Active Hotels in the UK. Meet up with the people, we end up buying that in 2004, $165 million US dollars with an earn out for the founders who I want to stay on because the real value in anything is the people, it’s not like a coal mine where you’ve got coal underground, this is the people, and so we create a structure that gives them the upside if we’re really successful.
So that happens in, when did that happen? 2004, in the summer of 2004, we end up in the fall of 2004, I am introduced to the people at Booking.com who are based in Amsterdam. They have almost the exact same model as Active Hotel, they’re disclosing everything just like you normally would by hotels but both Active and Booking have something slightly different than how the really big player, Expedia back then was the big player, and some of the other ones are. The difference is if you went to Expedia, you put your credit card in and you are essentially being charged right there. Merchant model, it’s called.
Expedia is acquiring the room and you’re buying it from Expedia.
GF: Well no, they’re not actually. Legally, they don’t actually own the room. In fact, what they have is the money in their bank account.
Oh, so they get float along the way.
GF: Which they are not planning to pay the hotel until after the person actually stays in the hotel and back then it would’ve been like a month afterwards. Really bad for cash flow for the hotels, really good cash flow for Expedia. Active had a different way of doing it called the agency model, which was, we’ll take the credit card, but that’s only to guarantee that you’ll actually show up. We’re not actually taking the money, the money was not run, the credit card was not run.
So essentially, when you showed up at the hotel under Active Hotels, you paid the hotel at the desk with your credit card, or cash, or whatever, then Active sent afterwards an invoice to the hotel to pay the commission on it, totally reverse in the way the cash flow works. Booking had the exact same model as Active Hotel.
So the issue was the more business you did, the faster you grew, especially with Booking because they had to pay Google up front in that month, but they weren’t going to get paid for months later, the hotel was going to pay him months later. The faster you grew, the worst your negative cash flow can get worse, worse, worse, worse, worse. So that’s one of the reasons we were able to convince them to join our little band.
So Booking came into, and we combined Active and Booking to one company, and we threw in Priceline Europe too. But Booking’s a better name, so we kept it as Booking.com, and that’s the story.
But here’s the question, though. You’re able to buy them because they have this problematic cash cycle.
GF: And we bought them because they were totally European base, pretty much, Active was mostly the UK, so they could share and we put the inventory together, so now they’re both better off. And look, you spend a lot of time talking about network effects, etc, well here where you really have a great benefit by being able to add a lot more of the inventory to show to a particular traveler. That’s one of the big benefits.
For sure, we’re definitely going to get to all that, but I’m very curious about this because from my perspective, the agency model is a critical component of what makes Booking what it is, into the behemoth that it has become. I talk about Aggregation Theory, and one of the things is the scalability aspect, all aspects need to be scalable and with this model, you have more from a hotel perspective, more scalability, more ability to come on, and you don’t need these custom integrations or whatever it might be. To me, that seems essential, so you acquiring the agency model seems super important to me, and it’s really striking to hear actually you talk about the real downsides of it, and the reason why it was a struggle for these companies.
GF: Well back then that was certainly one of the ways we were able to grow rapidly was because go to a hotel and say, “Listen, just give us your content, just give us your photos, just tell us the prices, we’ll put it on”.
We don’t need to make a business deal, yeah.
GF: Yeah, and then if somebody shows up, we’re going to send you an invoice, and please pay us.
Now part of the problems in that model for a lot of people who tried that in the past, was the hotels didn’t pay. So in the US for example, the actual collection rate for a travel agency was very bad, hotels would just ignore the commission letter, say, “Who cares about them?”, so that was a critical part too was collections, having a system to be able to collect the commissions. That was critical, and one among many different things that was done at the time that really made it different than a lot of other companies that historically had always done the commission-based agency model.
What I’m curious about is did you go in thinking, “Okay, Booking has this, we’re going to change it to our model”? And then was there a point where you realized, “No, actually for all the challenges that this model presents, it’s actually better”. Why is it better? Why did that matter?
GF: What’s good is having both models, Priceline was always the merchant model starting out. When we went into Asia, and we have the agency model at Booking.com, I was concerned, “Would it still work there?”, so we also looked at another company, very, very small company called Agoda. And Agoda based in Bangkok, they had the merchant model, and we ended up buying Agoda too, and now we keep them separate brands, separate things. Now, both have agency and merchant in different ways at different places, and the consumer gets to chose and there are many reasons people choose one versus the other. But having more choice is really an important benefit when you’re going to a customer.
And I’ll go on, it’s now in terms of the way people pay, it’s incredible. So I think we’re over 40 maybe now, different ways a consumer can pay us, and we take the money and send it to the hotel, which is really important. If you are a consumer in Shanghai, and you’re going to show up at a small little hotel in the South of France, and you want to show up at the desk and use your whatever, Alipay or your WeChat Pay, that French hotel’s going to say, “What are you talking about? Please give me euros”, and that’s not it, but we are in-between it. We will take the Alipay or WeChat Pay, and we will then put euros in the bank account of the French hotel. And by the way, it’ll be a lot cheaper for the French hotel than if they had gotten a credit card from the traveler, because the interchange is much higher than us putting money into their bank account.
So the answer to your question is payments is really important it’s the reason why we do a tremendous amount of our business is now done in the merchant model, not the agency model.
Interesting.
GF: There are a lot of benefits to it.
So the agency model got all these folks on board, and then over time once you had that relationship built in, you could circle back to the merchant model. That makes a lot of sense.
Network Effects
Going back to this buildup aspect, another critical piece of Booking from my perspective is that the hotel market in Europe, particularly 20 years ago, was far more fragmented. You just have a lot more independent hotels as opposed to the US being more chains and things along those lines. This seems to me to cut in multiple dimensions. First to the agency model point, that scalability bit and it being super easy to get linked up seems really important. And second, this really just landed you right in the sweet spot of an aggregator, where consumers see way more hotels than they would anywhere else. And from a hotel, you suddenly are seen by way more customers than anywhere else. This seems obvious now, I guess I’m stating things that seem clear about network effects, but did you realize this when you bought Booking — I’m just really curious, because the Booking acquisition was such a home run. One of the things you did as CEO was change the name, “Okay, we’ve held on to Priceline for too long, we are Booking”, again, it’s not a home run, it’s a grand slam. I’m just curious how much of this did you learn along the way versus, “Oh, this is how it’s going to happen”?
GF: No, these are all well understood concepts long before we started buying hotel agencies, like Booking.com. I had studied network theory long before that and I mean the basis of the Internet, or even going either further back, and there’s the famous when you’re in school you learn about, “Well, one telephone has a value of nothing”.
Metcalfe’s law, yeah.
GF: Of course it is. Right, exactly. You start getting into, “Well what is the value of a network?”, Ii it the square of the number of nodes, or how do you want to figure out the math of what is? But sure, of course that was all well understood and recognized why it would be such a valuable, valuable thing to have. But everybody knew that, and didn’t matter whether we were agency or merchant, everyone has built a network as quickly as possible, as fast as possible. What’s interesting, of course, is why did some of the networks that were out front, they got really good, but then they die out where others continue? We don’t have time to go into that, and theories of why did MySpace die but Facebook lived?
Well let’s do it in travel. Why did Booking, why did you pass Expedia? Why did you become the top dog?
GF: There are a lot of reasons that go into a question, and here we go into the idea of coming up with the narrative fallacy, which it’s a famous saying. Where we’re going to create a story about why the world is the way it is today, it’s not provable.
You’re tapping into what I keep trying to pry around the edges of, because I know the story you tell. There’s a story I look backwards and see, and this is all straightforward, and I’m just so curious about how you actually navigated this story way back then, when it’s, again, it’s really easy to look backwards and everything seems so obvious.
GF: Of course, that’s the issue of the narrative fallacy. And I can tell my story, and you have your story, many people have many stories about it, but there’s no way to actually prove out what is true and what is not true. I do know that I have a belief of certain things that are very critical.
I also know, and this is really important for anybody who’s in a business with anything in life really, is that so much of the ultimate results are going to be determined by things you have zero control over. I can’t control what our competitors do, I can’t control what governments do in terms of economics, I can’t control issues like natural disasters, and so many things are going to impact your business. In our business, my Lord, things hit us all the time, and the only thing we can do is control ourselves.
So try and improve the products, improve the services, so that both sides of our marketplace, our travelers and our partners, both see value in participating at our business, at their free will and that’s very important for us is that look, no hotel in the world has to work with us, they do it because we give them value to do it. And by the way, some hotels don’t. I even stay at hotels that are not on our network, because they don’t want to be on our network, but I want to stay in that hotel. There’s a place in Lake George in upstate New York—
That’s a lot of throat-clearing, though. So what is your narrative that you’re going to give my audience? What is the key thing why you surpassed everyone?
GF: I won’t do that, and here’s why, because I give it, and then it gives it more weight, and more credence, and more belief that was the truth. There is no truth. The only thing I can say is part of the reason is we worked damn hard, part is we had really smart driven people who wanted to succeed. Part of it is they came up with ways to provide more service, more value to both sides than other players did and being a friendlier partner, being better customer service, really focusing on it, customer centricity, those things were really, really important, but that’s kind of like table stakes for a lot of companies do that.
All right, fair enough. I’ll let you off the hook.
Managing Google
Tell me about your relationship with Google. I don’t know if it’s still the case, I know for a long time you were the biggest Google ad spender and I believe that really happened around sort of the tail ends of the 2000s, early 2010s, when that really ramped up. Was that driven by Google’s change of the SERP and moving away from ten blue links into their own travel product, and you realized, “Hey, game’s changed, we have to approach this differently”? What was the process there, and how has that evolved?
GF: Interesting, so when we were about to buy Booking.com, and we already owned Active Hotels, Active Hotels was very into SEO, all free search, very into that and they also had a lot of affiliates that were really bot farms, they were all farms, and we got a lot of business from them.
Booking, on the other hand, and this is another reason we wanted to combine them both, is Booking was really into SEM [Search Engine Marketing], they were paying, and they really became really one of the first players that I’m aware of that did really big A/B testing, really interesting science going into that stuff.
Right before we buy Booking, Google does a giant change in the algo, and we drop so much business on all the free clicks, it was like, “Wow, what’s happened?”. That’s when we really tilted more, “Okay, so we just have to keep paying more and more into Google”, and optimizing it so we actually have a good ROI, which we did, and continued to work on that.
That was before 2010, that was 2005, 2006, 2007. Then of course, you have the financial crisis, etc., but we still power through that. And that’s it, now I have no idea who the biggest spenders are with Google now, Google does not disclose, we do spend a heck of a lot of money with them, it’s a very good relationship, we’re good partners. We have a similar goal, and that is to get customers to get what they want as soon as they can.
I think that your approach to Google, for me, is a big part of the narrative story, and in two respects. Number one, there are other people in this space who got very large on SEO that basically their response to Google’s changes was to whine and whine incessantly for years and years and years and I feel there’s a very sort of pragmatic approach from Booking for years that, to your point, you can’t control what happens around you. You sort of play the hand that you’re dealt, and you dealt that by we’re going to pay.
What I’m curious about is, and this is sort of part two, is there an aspect where you’re dealing with Google, the ultimate aggregator, gave you more sympathy for the hotels, and dealing with you and knowing how to deal with them since you’re sort of sitting on top of them in that role? Or was there the opposite where just shit flows downhill sort of as it might be? I’m just curious, you’re in a unique position of clearly being an aggregator, one of the most textbook aggregators ever, and yet also you have to be aggregated by Google, and sort of navigate that, and figure it out.
GF: I do talk about this sometimes, it comes up. So we’ll have a hotel here, and let’s say we’re charting a 15% commission, and they’re not thrilled with having to pay 15% to us. Of course, we have to pay a big amount to Google to get the traffic to us, right? And so, you’re right, there is this relationship, and the big overall point is nobody likes to pay for marketing or distribution. Everybody would like all their business to come to them free, direct, loyal, again, and again, and again. People being people, that’s not exactly how the world works, so you do have to pay.
Then the next thing is you want to at least make sure that it’s a fair relationship and that’s really critical with our hotel partners, is saying, “Yes we are charging 15%, but we are getting you incremental demand”, and you understand this very well from your points earlier about understanding what’s the incremental cost to that hotel, so a hotel room that goes empty that night, that’s zero revenue, zero profit. However, you fill that up at $100, well what’s the real cost to that? Well, the incremental cost is essentially nothing, you already paid for the mortgage.
You’re just fighting over margin.
GF: Exactly! So we take 15% out of 100%. It’s actually a very good deal for them, which is why they do it so it’s all fine.
With us and Google is if we didn’t think we were getting a good ROI, we wouldn’t be spending it. Come on, it’s not just Google, well, whether it be Facebook, TikTok, Instagram, pick your channel, we are always looking to get the ROI, always looking, “Where can we optimize?”, and that’s one of the benefits we get to our hotelier friends, partners, is the idea is make sure that we are giving them a good ROI, and helping them out. Because if we don’t, they can go to Expedia more, there’s so many different ways they can distribute their rooms, because they want to maximize their bottom line.
I mentioned how your position might give you more sympathy for your suppliers in this regard. Now that you’re, I think, are you still the only European company that’s been designated as a gatekeeper by the EU?
GF: There are seven companies in the world that have been designated by the EU as gatekeepers. Unbelievably so, six of the seven are American, we’re one of them. We’re an American company, Booking Holdings is incorporated in the US, we are headquartered in the US. Booking.com is a subsidiary of Booking Holdings.
Right.
GF: The seventh company that’s a gatekeeper that’s not the US, though, maybe it will be very shortly is TikTok. So, there is no European company.
Does that give you more sympathy for Google now that you’re feeling some of their pain? (laughing)
GF: This is one of the things that I talked to the EU about, I talked to many people about this. I absolutely believe, I absolutely believe that the DMA, the way to set up its gatekeepers is not set up correctly. The idea that we, in one of the most competitive industries in the world, and I had a good headline quote in the Financial Times in London once, I said, “We have to fight for a booking every day”, because it’s true. You’re just one click away from somebody choosing somebody else, it is incredibly competitive.
So, the idea to put us in the same bucket as trillion dollar businesses like Google and Amazon? That’s absurd, it’s absolutely absurd. And yet what it does, it creates the costs, and the things we have to deal with that are much more difficult, thereby making it actually, we have a lot of European employees, so it actually makes it more difficult for European employees there, it has to be able to expand, etc. To me it just baffles the mind and I believe somebody said, “Well, you want to talk about reason to stories? My story in this is it’s politics.”
Well, all the hotels, to your point, they want all the customers you deliver them, and they don’t want to pay for them.
GF: Of course not.
They’ve complained about it.
GF: Exactly. That’s absolutely right, and I understand that. Listen, that’s part of a free world, if you can lobby better than we can lobby, I guess it’s not good for economics, it’s not good for society, but it’s good for an individual player.
Loyalty and Airbnb
I do have a bone of contention to pick with you, and you referenced it earlier. I, like many other professionals my age, was an extremely loyal Starwood member back in the day. It breaks my heart to see how Marriott has diminished those brands after acquiring Starwood in 2016 and the reason I’m blaming you, laying it at your feet, is it seems hotels realize that to avoid the way around you, or the way around aggregation in general is you need these super good loyalty programs and a good way to have that is to have the most hotels. So, I’m asking you, are you the reason why I now get crappy plastic cups, and uncomfortable pillows in Westin’s when that wasn’t the case before? Because you made all these guys consolidate, and try to get out from your fees?
GF: So, I’d first like to say that I am also very fond of Marriott at all of their brands.
(laughing) Of course you are!
GF: And Hilton, and I can go through the entire list…
That was on me, Glenn was grimacing the whole time for those that are listening.
GF: (laughing) Right, right. Look, here’s the thing I believe that-
I’m actually asking you in general about the hotel’s response to getting larger and things like that.
GF: It’s not just hotels. Look, everybody’s trying to come up with, “How can I get the customers to be more loyal?”, I mean, American Express just sent out the bill, if I want to have a platinum card, it’s going to cost $895. Meanwhile, Chase Sapphire, they’re being the cheap guys at $795, and I think there’s a Capital One that I know, $495, whatever. Everybody’s playing the loyalty, trying and bring more benefits, and they all show that if you pay this yearly fee, look at all the great benefits you’re going to get for it, it’ll actually pay more.
There’s nothing wrong with people coming up with new ways to try, and combine things to create more value and I have nothing against it. By the way, and I’ve been talking with all the big chains about this stuff. We need to have a relationship where I know they feel that if somebody booked with us who is a high status loyalty member, their company is saying, “Gee, is it fair for them to pay us the same amount of money they paid us somebody who’s never visited their hotels before?”, and I do agree with that, and we need to come up with a way that actually, while I’m not sure one will make any more money, or they’ll make any more money, but at least it’ll be a fair, and we used the word fair before, it’s a fair transaction.
If we could imagine a model where the commission is different if they’ve put in their loyalty number, or something along those lines.
GF: Totally, totally like that. Exactly. And by the way, and while we don’t do this, I mean look, some of these giant chains as big as they are, they don’t market to certain parts of the world. We market to people in Turkistan, or pick a place they don’t and we do things in over 40 languages, we have customer service for over 40 languages, they don’t. So, we are rolling benefits to them, that’s the thing they want is the incremental customer.
I put that list of crises earlier. Was Airbnb a crisis when it came along, or do you feel like in the end they just actually expanded the market, more places to stay, more types of trips, whatever it might be? Or was it actually — what was your process of facing this orthogonal competitor? You still talked on your last earnings call about alternative bookings, is it still a real center of focus for you?
GF: So, alternative accommodations in the last call, we talked about it over, I think it was 37% of our total room nights. Now when you add up what that number means, that’s almost a little under almost three quarters of the total number of room nights that Airbnb did, and we’ve outgrown them, 16 of the last 17 quarters we’ve grown faster. So, I’d say it’s a pretty big deal for us and it has been for some time.
So one of those things that seemed like a crisis, and has actually turned out to be a huge opportunity.
GF: It never was a crisis, it was just another area where we need to build, and we didn’t have the right processes, we didn’t have the right platform, we weren’t necessarily doing things as well as we could have. And by the way, I said this in my last call, I said, there’s still a lot of things we can improve in that product for our customers and our hosts in a way that they will prefer to use us more and that’s something which is good. That actually means there’s actually a lot of upside left for us, and I believe that, I really believe that.
Is your key advantage just the fact that if I’m searching, if I search on Airbnb, it’s only Airbnbs. If I search on Expedia, I can see both then compare, or is there more that goes into it?
GF: Well, I’d say that’s a big one right from the beginning and I’ve said this so many times that one of our advantages was that I did notice that [Airbnb CEO] Brian [Chesky] recently was talking about how they’re going to be in a lot more hotels.
I’ve been saying that for years. Airbnb, I mean obviously it was a great idea, could create an entirely new market. There’s a lot of strategic decisions they’ve made in the last X number of years that I don’t completely get.
GF: Well, so, I imagine that we will be seeing more of that competition coming from that area. But look, it’s all good. I like competition, it helps us all do a better job for both sides of the marketplace.
Connected Trip
So, something you’ve championed for a long time, well before AI, is this idea of delivering a Connected Trip, you bought OpenTable, and you’ve bought lots of other companies along the way. What is that? Explain, give me the high level thesis. How have you approached it? This feels a little like, “Let’s put ourselves in a pre-AI mind”, because I’m going to ask you how AI changes that. I’ll give you a preview of the next question. Is this a bit where it sounds good in theory, but it’s not actually the way customers operate?
GF: So, a couple ways to approach this. One way I always like to explain it is when I was younger, much, much younger, and my family would set up a trip somewhere. There is a thing called a travel agent that was a human being, many times was a woman, and she knew a lot about our family, and what we liked, and how much we could afford and she’d come up with some ideas, and my parents would have conversation with her, and such, and then she’d come up with a couple things, and we’d buy it, and we had everything, all the things we needed, the flights, how to get to the airport, the hotels, what to do there, all that stuff, and we give her one check, my mother would write a check to her and be done. God forbid anything went wrong, we just called her, and she would fix it all. Magically it would all get fixed, it’ll all be fine.
Now technology should be able to recreate that only much, much better, because the database doesn’t forget like I imagine the travel agent human probably did occasionally.
Well, that, and it’s much more accessible.
GF: It’s 24/7, 24/7, and in addition, which is really good, it has an ability to know so much more about what’s happening at the exact time, or in the future. So, for example, the human travel agent never would’ve been able to, when the plane is delayed by four hours to send out a text, and say, “Excuse me, Mr. Fogel, your plane right now will be late by four hours, but we have a different flight for you, would you like to change it? In addition now because you’re going to still arrive two hours late to your destination, we’ve already rearranged the car service for you, and in addition, so you’ll no longer make your dinner reservation, we’ve already tried to move that to a different time, but that restaurant doesn’t have availability at that later time, however, we have looked at three other restaurants that are in the near vicinity. Would you like to choose one of those instead, which has availability, and we’ll notify your partner so that person knows we’ll be there.” What I want is I want to have a third spouse in our, my wife, and relationship who does all of this for us. Only it’s not going to be human, it will be done digitally.
Right. This all sounds amazing. But you launched the Connected Trip idea a long time ago, and in retrospect, particularly when you look at AI, and how AI can potentially do this, were you just way too early, and this sounds good in theory, but actually there’s so many difficulties in pulling this off? Where actually we just have a holding company of a bunch of different companies that in theory work together, but it’s not a doable project?
GF: I do not disagree with you on your point that it’s taking a long time. Agree totally, but you have to think where we came from, all we had were hotels. We didn’t have a flight business at all, we really started that 2019, just started it. So, five years later, let’s round off, call it five years later, we now sell more airline tickets than any other third party distributor in the world except for China, a little different, so we’ll do it ex China. That’s amazing in just five years to do that.
In addition, the increase our Attractions now, we had no attraction business at all, now we doubled the last quarter, we doubled the number of Attractions. Our car rental business continued to grow. Rides, instead of renting a car, you just get a ride service from the airport, or to the airport, insurance, so many things.
We had to build all these different verticals, and now having done that now, bringing it together, using science, using data, analyzing it, making sure we are putting up the right offers at the right time with the right value to the right person so they get the best experience they can. And God forbid anything goes wrong, we as the person who actually has put it all together, with the relationship with everyone, we can fix it. Now that’s for the traveler.
On the partner side, the supplier side, we’re giving incredible number of different opportunities for them to provide, get incremental demand by providing different values, different things. We have a thing called our Genius program where primarily the supplier fund discounts, because we’re giving it to certain customers who pay more, who are willing to do it more often, so we give more value to our supplier partners.
Combining those two things, using again all the data analytics, it’s going to be wonderful. Now you’re right with generative AI, it may be easier to do all this stuff, don’t disagree with that. Your point, it’s hard is very correct, which is good, because if it was easy, there’d be a lot more people doing it.
What’s more important? Is it the demand side, or the supplier side? And I know the answer is both. They’re both really important, or does it matter in different stages of the evolution of your company? For example, I thought it was really interesting, you made the point that agency model was important to get that demand, but over time, merchant actually lets you deliver a better experience for both sides, and so you’ve seen some transition there. Is that the case different when you’re building a network effect, which one comes first if you want to add on a connected experience?
GF: I think the problem is when you’re overweight one side of that type of equation, you end up with real big problems and whether, I know you did a recent, it was very interesting interview with [Uber CEO] Dara [Khosrowshahi], shout out for the interview you do with Dara because I thought it was fascinating, but it’s very similar network effect. If you have too many drivers, you’ve got a problem. If you don’t have enough drivers, and a lot more people want rides, you’ve got a problem.
Same thing for us. If we have too much demand, and there is no supply, we’ve got a problem, the customer’s unhappy. Similar the other way, if a supplier puts out a line, they’re not getting any business, they get tired, and they stop working with you.
Right. It’s like, “What’s the point of paying this 15%? I thought that I was supposed to be getting lots of customers”.
GF: Well, no, don’t forget about they only pay the 15% if they get the demand. That’s a very important point to make out, which I continue to make to anyone who is interested in the issue of what value do we provide, I say, “Look at what we do here, we’ll take all the inventory, we’ll take the content, we’ll take the photos, we’ll give suggestions how to do things better, we will translate it into forty-something different languages, we’ll go out, and get demand, and we’ll make sure that your property is shown to the right customers. We do all of that, and I’m not going to charge you a thing, it’s free, absolutely free. The only thing I ask is if, if you get a customer who stays in your hotel, would you please pay us a fair commission?”
That’s it, it’s a great deal. It’s a great deal for everybody, and for everybody who comes out, and says, “Oh, well they have too much power, or whatever”, people come out and say — I say, “What are you talking about? You don’t like it? Don’t join!”.
Network Effects, Part Two
Something Booking is hailed for, we were talking about this before in your story. I appreciate that I don’t want to land a specific narrative, it is what it is, we work hard, all those bits and pieces, just between you and I, however, is the actual answer to Booking’s durability, is it really just a network effect thing? This ties into what’s more important, demand or supply? The answer is both, they’re both important and you have both. Are network effects still underrated in terms of how you survive through paradigms? How do you avoid disruption? And the answer is if you have a network effect, you have the time and space to figure it out, and that’s okay because to your point, you’re motivated and you have the business model where both sides benefit. But do we still underrate these concepts you’ve been studying for so long?
GF: I go the other way. I’d say that the network effect is overrated.
Interesting.
GF: Terribly overrated.
Well, tell me why.
GF: And a lot of antitrust people think that, and you can go back to the A&P cases way back, I don’t know how long, 50 years ago, 60 years ago, probably more actually maybe 70 years ago. The fact is that there are lots of very similar networks out there. You mentioned Bonvoy and Marriott, huge network effects there. Tremendous. All the airlines, tremendous. All our competitors, similar. Trip.com, they actually do more total travel value than we do probably.
I tell you this idea that there’s some sort of lock-in effect, which is really what people start talking when they start networks, they start going to this idea of some sort of lock-in, there is no lock-in and I have the proof of so many of our customers do not spend a lot of their travel budget with us entirely. They use lots of different ways throughout the year, they’ll use us for some other times. They’ll go direct to the hotel, sometimes they’ll go to a tour operator, there’s so many different things. The data proves really that at least in travel, I won’t try and go beyond my expertise, in travel, the network effects are very low. There is incredible competition and those that say otherwise are ignoring the facts and the data and the science.
I’m going to disagree with you slightly, but not completely, in that my contention is that the network effects enjoyed by your service, the aggregators, is because by definition you’re delivering value. To your point, hotels get free marketing until they actually make money and then you’re taking a commission on that. It’s not like they’re paying to get on there because you’re not the only way to reach customers, you have to earn it, quite literally. You have to actually deliver a booking.
GF: Yeah, exactly. In fact, here’s the thing. Do you know in terms of Europe, our biggest market, we only have a single teens percentage of the market of hotels? That means 85% or roughly somewhere in that neighborhood of hotels that are being booked are not going through us. So anybody come up and say, “You have a 15% share, but you’re a dominant player”, is that not one of the craziest things you’ve ever heard?
I mean, don’t get me started on European craziness, but I think the reason I’m pushing on this is network effects are a good thing. Being able to find any hotel in one place is convenient and it’s great. Being able to acquire more customers is convenient and great.
GF: I agree.
So that’s why I push back, I think there is network effects.
GF: No, I agree. I’m saying not as big. So we’re just sort of arguing semantics.
It’s not railroads, it’s not cables, that’s the problem.
GF: I agree, totally. I’m saying it’s not as powerful as some people think it is. It definitely is, there’s value definitely, it’s just to my thing and there are a lot of others who have the exact same type of network.
Booking’s Future
Where is Booking, when you look forward 5 to 10 years, is it still the arc we’re following? It’s this Connected Trip, we can tell, to your point, I think the problem with a lot of these Connected Trips ideas, particularly before, is it’s a good narrative. It sounds great, I’m going to have the travel agent deliver it and the number of edge cases and complexity in the real world is just so large, it’s never been doable. But is this a bit where actually, that’s probably right, but all that investment was worth it because suddenly it is doable. Did I just summarize what’s the next decade for you?
GF: Well, I think you did. I mean our goal is to get that Connected Trip really in a way that a lot of people really see the values. I’m going to go to Booking because I get a lot more value through the Connected Trip and we’re seeing a little, like we did announce in our earnings call, how the growth of people who are buying more than one vertical in a trip has gone up. I believe it was over 30% is what we said a couple months ago.
I just read it about an hour ago, so I remember that.
GF: Okay, good, so my memory isn’t gone. If you notice, it broke into the double-digit percentage of our total amount of transactions, so it’s showing it is growing, but we are still so early. I believe this is one of these things where you do hit a phase transition, you build up the verticals, you start going, but then at some point, instead of it going linearly, you get exponential growth as people really catch on. This really is much better.
Is this going to have to be a product-led growth strategy? Can you SEM yourself into this or is that you almost have to build a new muscle in that regard?
GF: I think the marketing of it will be important to help get people aware of it, but I don’t want to put on a firehose of marketing until it’s really good.
That makes sense. This is the other interesting point where you can have sympathy for hotels, you have your own loyalty program, you want people opening Booking first thing instead of going through Google to get there and you mentioned the Genius program and how your suppliers will help fund that because you’re consistently delivering more people. Is that a big focus going forward? Do you feel good about that? Is that going to be critical to getting the Connected Trip working?
GF: What I really want to be able to do is help our partners do better, I want them to make more money than they would otherwise, I want them to have happier customers than they would otherwise. One of the ways that we’ll be able to do that is by providing them incremental demand right away. That’s great and all that, and I understand where somebody say, “I wish I didn’t have to pay a commission for that customer”, but I want to help make sure they understand that they wouldn’t have gotten that person.
But there’s more. We know technology is now coming forward so rapidly, so many of these hotels, they don’t have IT departments, they don’t understand, they don’t know how to deal with this. We will do it for them for free. Again, all the things we do for them, and believe me, this is not dissimilar when you go back to the mid to late ’90s when hotels didn’t have websites, they didn’t know what to do or how to do this or whatever, somebody did it for them. Well that’s kind of now in this new world with all this new technology, we were able to provide the service to them again in a way that they will be better off, their bottom line’s improved.
Do you think there’s a shift where you keep talking about, “You’re really valuable, people shouldn’t be mad at us, we’re providing incremental revenue to assets that are already there”, but there’s one thing that I think is interesting about the ultimate aggregators is in some respects they create their own supply and maybe this is where Airbnb was really interesting and got there first, and you’ve done a very good job catching up to them, but creating their own supply of rooms that only exist because Airbnb. I talk about Facebook creating an entire universe of e-commerce retailers that basically exists because of Facebook, that gives them real durability. What market can you create as opposed to servicing hotels that exist and we’re going to give you incremental revenue there or has that already happened?
GF: I think you’re right. In both examples you gave, they did create new blue ocean markets that did not exist previously and more power to them for having done that. We, as you point out, we are basically making a markets more efficient, providing more value to people who are already into the marketplace. So for us to create a new market, we’d have to be going on to something that is not currently being done. Now, if we have such a thing, I would not be disclosing it right now.
Right, fair enough. But it’s fair to say, and if you were to even do the Connected Trip, but most of the stuff, you’re not talking about Airbnb, Airbnb is still like, “We’re going to create experiences X, Y, Z”, you’re like, “No, you can get your car and then get your hotel and get your restaurant, that’s what Booking does, it makes it more efficient”.
GF: And very important. But services, what we call Attractions, are very important because nobody goes on a trip to hang out in the hotel room, you go to do stuff. So whether it be the scuba trip, the snorkeling, the museum, the theater, whatever it is, we want to be able to provide that to them too.
One of the examples of the Connected Trip, I always give an example that I love, which is you go to Amsterdam, the two things most people want to do, they want to take a canal boat ride in Amsterdam and they want to go to the Rijksmuseum because it’s very famous museum, great things. Now the thing is though, we want to be able to provide, which we can now, we provide it to both tickets for both things. Imagine you have booked the Thursday for the boat trip and the Friday for the museum. On Tuesday, however, we notice, we obviously see what the weather is and we come back say, “You know something, the rain is coming for your boat ride day and it’s going to be sunny the other day, so why don’t we switch them around for you and we can do it all. Would you like that? Press yes.” And we will do it all for you, that kind of convenience and that looking forward to provide what could be a problem beforehand, that’s real value. Those are the type of things absolutely a part of the Connected Trip.
Is that another advantage of Booking in particular’s European roots? It always strikes me, all these European sites have their own technology stack from like 25 years ago, whereas U.S. is just like, “Yeah, that’s fine, we don’t want to do that, we’ll jump on sort of something else”. Is that a real opportunity to go in and provide a brand new technology stack for all these experiences telling them this is the advantage you’ll get?
GF: So we have a company that’s the leader in the space, FareHarbor, which we own, which works for us and provides the backend ticketing and reservations for a lot of small and medium-sized Attraction players. We are the biggest player in the world in that, so that is our stack that we do provide, free again, to things and think of it as like OpenTable, which we also have only, FareHarbor’s for attractions, OpenTable for restaurants, etc. But you’re right, there are a lot of different backend systems, but the wonderful world of APIs, you can tie to anybody.
Got it. Well, I could talk about Booking forever. Like I said, it’s such a textbook example and it’s so interesting that it’s on both sides of this equation, but did we get what you wanted to cover?
GF: Yeah, it has been really fun, really interesting. I mean, it’s fascinating. As you’ve done in many of your interviews, you just scratch the surface because there’s so many interesting things in the topics that you cover, but this certainly is one, and today with Gen AI coming down the pike for all of us, how will the world change? It’s going to be a very interesting thing
Google — are they going to be your most important channel forever?
GF: Who knows, right? Who knows about that? And I would even more so are the big societal issues and these are the issues not so much about how we do our business, but how do all of us do our business and what are the impacts going to be in terms of employment, in terms of jobs? These are interesting things that I wish our government friends would spend a lot more time on that as opposed trying to guess what’s happening in small vertical areas.
Do you think you might see a fork in the road where you want to deliver this experience in Booking and be that destination and tie it all together, but actually the ChatGPT’s of the world might actually be where all trip planning happens and you’re just going to have to, like you did with Google back in the day, accept that and be the best at that?
GF: It’s certainly a possibility, we’ll see how it plays out. We are currently and have announced many relationships, many deals with many of the players in the space, and there’ll be many more in the future.
Look, there are going to be people I absolutely know are going to go to a large language model provider and want to start their trip planning discovery there and I’m going to lament and say, “Why did they start with us first?”, it’ll be just like hotel, “Why didn’t they come direct?”, it’ll be the same thing, but I like to believe that we will create things that are so great the same way we have over 60% of our customers currently come to us direct. When you X out the B2B business, you go just the B2C business, over 60%, it’s mid-60s actually come to us direct. They’re not going to Google first, they’re not going to anybody. I would like that to actually be even a higher number down the road and they’re coming at the very beginning of their discovery area.
But if they don’t, you’ll figure it out and be there.
GF: If they don’t, we catch them later in the funnel and we’re still able to be able to get paid well, it’s all good.
Glenn Fogel, thanks for coming on.
GF: Thank you very much for having me.
This Daily Update Interview is also available as a podcast. To receive it in your podcast player, visit Stratechery.
The Daily Update is intended for a single recipient, but occasional forwarding is totally fine! If you would like to order multiple subscriptions for your team with a group discount (minimum 5), please contact me directly.
Thanks for being a supporter, and have a great day!
