Arnie Weissmann: Last time we spoke, Carnival Corp. was launching the Ocean Medallion platform, which includes a wearable that enables guests to, for example, shop, gamble, order a drink from anywhere on a ship and automatically open stateroom doors as one approaches them. It was scheduled to go live on a ship last November, but a full-scale launch was postponed in favor of a limited rollout.
Are we not ready yet to undertake such ambitious, complex service systems on a ship- or propertywide basis? Is this perhaps a little bit more difficult than people assumed a year ago?
Shelly Palmer: It doesn't surprise me that a program like that would be delayed. It's harder than it looks. What I think is important to understand is that this is a guaranteed future because it reduces friction for purchasing, and people like to purchase stuff. This is what these systems do, and they do it really well. It's a guarantee. It's not even slightly not going to happen.
And I wouldn't read much into the delay. I don't know the specifics, but it's something they're going to be able to do. It's going to be awesome, and people are going to like it. The Disney organization has had nothing but success with their MagicBand.
Weissmann: John Padgett, who led the Ocean Medallion initiative, was previously with Disney.
These systems, however, do raise questions about achieving the right balance between technology and service that's provided by human beings.
Palmer: There are people who just hate the [digital] experience. They don't want it. They want the friction. They want a bit of a wall between them and spending money. They believe that [these systems are] an invasion of privacy in a way that's unacceptable. It's pretty generational, but it doesn't mean it's not real.
It's still early days, and everyone's got an opinion about what service means.
This matching of people with preferences is probably the hardest part, because everybody's threshold is a little different. Sometimes you're in a restaurant, and you're depressed because the waiter only came over once and then left you alone. Sometimes, the waiter comes over and interrupts every three minutes, but all you want is to eat and have a conversation. And it could be the same person who wants two different kinds of service on the same day.
It would be relatively easy to [give a diner] some kind of tablet, or let you use your phone to order. A millennial might like that. Somebody my age might think that's really obnoxious -- like, they can't walk over here and talk to me? I want someone to tell me the specials, I want a sommelier to tell me about the wine. They could push that through an app, but is that the same as having someone who you know went to sommelier school come over and talk to you? For some, it might be; for others, it won't.
Can I imagine a time when everything is automated, when we're living in an antiseptic world where I can walk in someplace and don't have any interaction with people? Yeah. Can I also imagine that right down the block there'll be a small boutique hotel with 24 rooms that are relatively expensive, that has a complete staff that is artisanal in nature and it feels like a small luxury hotel in Western Europe, circa 1920? You bet.
Travel is as personal as the people who travel. I think the hospitality industry is doing what it can to use available technology to find a happy medium. The cohorts are going to tell you how to apply technology to them. It's a really hard job. It's going to take time.
What everybody needs to understand is that technology is an enhancement, and the biggest mistake everybody makes is not understanding that just because you can do something, that doesn't mean you should.
My best hospitality experiences have had a point of view, a theme, have been experientially driven. The people who've made that experience are experts in what they're creating, whether it's something as kitsch and prepackaged as the Grand Floridian at Disney World or as authentic as the Cipriani on the Giudecca in Venice. Each of these hotels knows what it is.
But [a cookie-cutter business hotel brand that Palmer asked not be identified]? It doesn't mean anything. You can't name those three hotels in a row and call them all hotels. One's a place to sleep for a business traveler, right? One is a vacation for a family that costs more than going to Europe, and the other actually is going to Europe.
Weissmann: In advance of CES, I was inundated with press releases about robots, much more than last year. Did robotics progress significantly in 2017?
Palmer: Robotics are progressing on an exponential curve, the same way every other technology is. They're getting smarter, they're getting more affordable, they are becoming more capable. But I don't think it's nearly as consumer-accessible as some other technologies have become.
Weissmann: The main travel application for robots I'm seeing is in hotels. You can have amenities or room service delivered to your room.
Palmer: The robotics that impact the travel industry will clearly be ones that interface with consumers. But delivering amenities is a parlor trick right now, not a paradigm shift.
Robots are fun, but there are still navigational problems. They have to roll on some kind of wheels. You're not going to have bipedal robots or quadruped robots walking around the halls. They're expensive and they don't really do much yet. There's a robotic receptionist in Japan, and there's a hotel that has robot staff. It's all just a way to get you interested in the hotel itself.
But the delivering of food is still a ritual in America. It's not a process. It's not just someone bringing you a tray, like in a hospital. There's a whole presentation that comes along with room service. That's not what robots are about.
We have tools available now that can solve real problems. If you're a hotelier, the last thing you want to see is a line of 50 businesspeople waiting to check in. But I can walk in and the hotel knows I'm there -- my smartphone checks me in, tells me my room number and will be my key.
And what's really going to be interesting involves natural language processing and understanding. Voice-enabled rooms are very important.
I'll go into my room and say, "Alexa, lights on," and the room knows that I want the lights set a certain way, or [if I ask for the television to turn on, it knows] I want a news channel. That's the kind of stuff you're going to see, and it's going to be very well accepted.
Artificial intelligence (AI) can help make your room and amenities match your preferences in a better, more vibrant way. Everyone's carrying a quad-core computer with five radios in it in their pocket, so the bigger advances are coming from ways to explore and exploit -- and those are technical terms -- [smartphones] to help people have a better experience across the board. Am I going to order the pizza I always order, or am I going to find something new? This is where the explore or exploit algorithms show up.
When the hospitality industry gets it right, they'll serve each cohort the way it wants to be served. They'll do that by using some kind of well-matched training set, either machine learning or some kind of neural network AI, to analyze data and really get a deep understanding of what it is you will like or how you behave, and they'll match your behaviors to appropriate technology. And that is no mean feat. That's going to take a while to get right.
Weissmann: To some extent, it is, as you say, generational, but it's not all about technology for millennials. They use travel advisers in even greater numbers than baby boomers do. Does that surprise you?
Palmer: Nope. Not even a little. At the end of the day, whether it's a millennial or a baby boomer, people pay for experiences and expertise. You want someone to provide an experience that, in many cases, is more valuable than things.
You'll do research, but you're looking for expert advice. We as a society have seen the quick, painful death of truth over the last 24 months, and people are looking for things they can trust. Generally, people trust people who have some track record.
This trend is across categories. As access replaces ownership, you're looking for an expert-curated experience.
Weissmann: The CEO of Expedia, Mark Okerstrom, recently said that when it comes to man versus machine selling travel, humans are winning now, but eventually "we think machines can win." Given what we just spoke about, could he be right?
Palmer: He's spot on. Here's why:
If I gave you a 10-by-10 matrix about your business and asked you to tell me about it, you'd have no trouble explaining it to me. "Oh, this is a double-income, no kids household. These are seniors. These are empty-nesters. These are millennials."
But if that spreadsheet was 25,000 columns by 250 million rows, there's nothing you can do except train a machine to look at that data set. And once you do that, a bunch of things start to happen. Patterns are found that aren't obvious because you can't scan that much data as a human.
So, while over the course of a lifetime, a trained professional might be able to amass enough acumen, knowledge and experience to understand how to present various aspects of their hotel or restaurant, and they may understand how to solve problems, the data set that an OTA will have is going to dwarf anything humans could ever do. And they'll turn it into action in ways that people who aren't doing that will not be able to compete with.
Weissmann: But what about human intuition? In "Blink" (Little, Brown and Co., 2005), Malcolm Gladwell wrote about how an art expert can look at a forgery that's technically perfect but sense that something is wrong. Is intuition a human advantage that is beyond the capabilities of computers?
Palmer: I think looking at the extremes is really not a good approach to thinking this through. The vast majority of situations in our world are incredibly middle-of-the-road and simple to deal with.
There's a scene in the movie "I, Robot." The main character, played by Will Smith, is a detective who hates robots, and there's a main character robot, Sonny. Smith's character, says, "Can a robot paint a picture like Rembrandt? Write a play like Shakespeare? Compose a symphony like Beethoven?" Sonny looks straight at him and says, "Can you?"
Most of the work in travel is process-oriented. Is the room cleaned? Did the meal get there on time? Was it served hot? These aren't subjective questions.
So when you talk about what AI can or can't recognize, why are you trying to have a computer do something it doesn't need to do for you? Why would anybody rely on creating a cognitive brain that understands humanity if you can build a cognitive brain that can search out patterns you can't see and enhance your ability to serve your customer?
If you look at AI the wrong way, if you look at machine learning the wrong way, you think it's going to do your job and replace you. Not only are you mischaracterizing what it's capable of, you're completely missing the point of the tool set. You don't let your email client write an email for you. You write the email and it delivers it. You don't let [the GPS app] Waze tell you where to go; you tell Waze where you want to go, and it gives you the best route. That's how you partner with a machine, and that's what all of the OTAs are going to do, what every hotel chain is going to do.
If you're using digital tools that partner with AI or with machines, whether it's a mechanical device like a robot that has some kind of interactive system or if it's something as complicated as a true neural network that's analyzing billions of transactions, you don't let these things just tell you what to do. You partner with them. People try to anthropomorphize AI and data-driven systems, as opposed to just using them for the things they do best.
The people who are going to win will partner with a machine that can look at 250 million rows of data, find patterns you trained it to find and allow you to serve hundreds of thousands of customers as if they were personal friends of yours.
Weissmann: You're suggesting that the larger the company, or the more access it has to data, the bigger competitive advantage it has. But I, for one, don't want to provide data about myself every time I interact with a company. The reward has to be huge or I won't do it. It's intrusive and irritating.
So are we going to end up with all our preferences in some Big Brother database that anyone, including smaller companies, can access, or will large companies use their data power to sideline and make everyone else irrelevant?
Palmer: I would opt for number two. I think that the more data you have, the better it's going to be. If you travel enough with the OTAs for them to care about you, they're going to make things happen for you that small companies can't.
Weissmann: Still, companies need my cooperation, at some level, to serve me. Is my reluctance to provide that cooperation unusual?
Palmer: No, I don't think that's unusual, but what I do believe is that we, as a society, have not yet figured out who owns our data. We haven't really sorted out privacy. It's kind of generational. If you're of a certain age, you're really not interested in giving up what you perceive to be your privacy, whether you ever really had it or not, whether you understand it or not.
For the most part, business travelers understand that they're in a loyalty program, so let's get the most out of it. They will give up data for the quality of their enjoyment.
If you're in the Hilton Honors program, and you're at a certain level, they automatically upgrade you to a concierge floor, they'll give you complimentary water and internet. That's the easy stuff.
The crazier stuff happens when you're not super high-loyalty. They know you have other hotel choices, and you're not quite at the level they need you to be at. What can they do to incentivize you to make your experience so much better that you'll consider a Hilton over others at a given time?
What makes a data set interesting is that it can convert those people by giving them amenities. I'm not just singling out Hilton; it's any hotel chain. The bigger you are, the better off you are.
Small companies that have the computational capability -- all this stuff is going to become commoditized in fairly short order -- but not big data sets to analyze, well, all the computational capacity in the world doesn't really help you. You must have the data sets.
It's so early in the history of all of this. We haven't had our meltdown yet. The bad thing hasn't happened. I don't know what that's going to be, but you can be sure something incredibly weird is going to go down, and everyone's going to go, "Oh my god, machines are evil."
We as a species and a society are learning, realistically learning, what it means to live in a world where machines can do a bunch of stuff. What are we going to let them do? What aren't we going to let them do? Those are big questions.
Weissmann: Security seems to be an issue in myriad ways. Cars have been hacked, and the driver has lost control of the vehicle to hackers. Could airplanes be at risk for something similar happening?
Palmer: I don't think so, for a whole bunch of reasons. There are pretty decent safeguards in place. I'm not saying it can't, because anything can happen. But so far, this isn't a "thing." It would be really hard, way beyond the capability of most people.
Weissmann: We just talked about using data to woo travelers to be superloyal. But will airlines, for instance, use data to make flying more pleasant for even infrequent travelers, to try to convert them?
Palmer: I'm a very frequent flyer, and I can say with a high degree of confidence and consternation that airline travel is like riding a bus. Flight attendants never say they're there "for your comfort" or "for your pleasure." All they're talking about is your safety. A safe flight is the only thing they're trying to give you. Everything else ... is ... nothing.
While some transpacific flights are magnificent and a few transatlantic carriers can still show you a good time, mostly you're packed into 17-inch seats that are 29 inches apart. That's suitable for cattle, not people. I don't think that the airlines have any economic incentive to change that. If you don't have the extra 50 or 100 bucks for more legroom, you don't get it.
Weissmann: Airlines -- and hotels, and meetings planners -- were concerned that telepresence would cut into their businesses and people would meet electronically rather than face to face. That didn't happen, but will technology improve to the point where it might?
Palmer: Telepresence sucks. Every instance of it is terrible.
We don't yet know what impact augmented reality (AR) will have. If they design AR to be like telepresence, it's going to be useless. But if they make it like you're sitting at a conference table and you're looking right into someone's eyes, it'll be a whole other story.
It doesn't have to replace meetings. You only need to take a 20% bite out of certain things to hurt them hard. You ankle-bite 20% away, they're going to feel it.
I've put on a pair of AR headsets and had some incredible interactions. Is that the preferred way? I don't know, but if I've got to get on an airplane, travel 16 hours to Korea for a two-hour meeting because I must meet someone in person, because that's their culture, and then fly back the next day for 16 hours -- and I have an opportunity not to do that? It's not the $20,000 of airfare I'm pissed about, it's the 32 hours of my life I can't get back.
Weissmann: Last year at CES, I came across one company that claimed to translate foreign languages through an earbud. Fast-forward a year, and I'm aware of three companies -- Google probably being the game-changer -- offering a form of this. But for now, the reviews of Google's translation with the Pixel 2 earbuds have been pretty negative.
Palmer: It's early days, and I refuse to get into Google-bashing on this. Google Translate is the largest AI project on planet Earth. The things under the hood that Translate has done are well documented, and the neural network they've built is astounding. This is a most remarkable project, and when all those stars align with natural language understanding, automatic speech recognition and natural language processing, there won't be anything else on the planet that will impact humanity more dramatically.
That is not hyperbole. Language is a huge barrier, and they're making it a trivial barrier. And they don't have to get it all right. They knock 50% of the friction out, the world changes.
I'll look at a menu in Chinese, I'm going to see it in English. I'll put in a pair of Pixel 2 earbuds and say, "OK, Google, help me speak Spanish," and a person speaks in Spanish and I hear it in English. Right now it's not great, but give me a couple years and I've got a universal translator in my ear. You tell me that's not going to change travel?
At CES, you're seeing the beginnings of a massive change in the way humans interact with one another. Translation is a giant part of a travel experience; imagine if you could understand everything that was said to you. Imagine if you understood everything you saw that was printed. That's what's happening.
To subscribe to Shelly Palmer's blog, go to ShellyPalmer.com, click "Blog" and scroll to the bottom of the page.