AI adoption at Marriott Bonvoy - Drew Pinto, Chief Revenue & Technology Officer, checks-in on progress to date

For the past few years, global hotel chain Marriott Bonvoy has been working through what it pitches as “a massive re-platforming” of some of its most critical tech systems. That’s a big enough task for any company the size and scale of Marriott, but into the mix now must be thrown AI and its impact on the organization. 

As Drew Pinto, Chief Revenue & Technology Officer, observes:

When we started the design of that big transformation, AI wasn't what it is today.

But he adds:

The good news is that we built our architecture in a way that's very flexible and easy to scale and also to adapt. So bringing AI in is almost a nice additive component that helps amplify what we were doing....It’s opening up a whole source of data that we didn't have before. Things like property data or customer data, [we] used to be limited [in our ability] to do the things we wanted to do because of very localized old systems. [We’re] now unlocking all this opportunity to then apply AI to that to help those constituents that we mentioned, which we didn't have before. So if anything, this has been really good foundational work that opens up capabilities that we didn't think were possible before.

In fact, AI is becoming more and more of a critical component within broader strategy, to the extent that it is  now hardwired into the corporate DNA, says Pinto:

The trend that I'm seeing even more so is AI is now permeating everything we do. So projects that we were going to make investments in that maybe didn't have an AI component, that we never thought would be AI relevant, now suddenly, it's almost a requirement by us that within there, there are AI solutions and that we're designing for that. As part of it, it's not the full story, but anything we do, whether it's Finance or HR or Marketing or anything, we’re saying, ‘Where is the AI portion of this solution?’.

This is not just a technology conversation, he argues:

What [we] have really been trying to do is not just talk about technology for technology's sake, but [ask] how does it apply to driving the results of the business, whether the commercial results or the other KPIs we have? We're doing it for a purpose, so we're making sure the investments and the effort goes towards that purpose. We're now plugging AI into that instead of it being this discrete thing that stands on its own.

That’s something other organizations might learn from, he suggests:

I think that's really important because you can get enamored with a certain AI solution and think, ‘OK this is going to solve everything’ What we're learning is it has to be really integrated. Whether it's [about] data or security or the customers, it's not going to stay in one lane. They’re going to want to know that you've thought this through and thought their experience through. That's really where we've been focused.

Searching for answers

Partnering with AI specialists, such as OpenAI and Google, plays a part in Marriott’s thinking, with Pinto explaining:

We're really involved with the OpenAIs and Googles for a couple of reasons. Number one, this is all exploration. We're trying to learn, they’re trying to learn. It's all developing. That's why it's too soon to really say where is this all going to land. But what we want to make sure is that we were in there and learning along the journey.

Second thing, and probably even more importantly, is influence. We want to be able to be influencing this model to make sure that it adheres to our distribution standards, that it helps our owners and franchisees have a customer acquisition strategy and approach that's going to work for them [and is] going to be sustainable.

That's why we're in early and really trying to figure it out as we go. Literally, I get an update each week from the team, and there's a new merchandising idea that we might do with Google as part of this pilot, for example.

While it is early days, as Pinto points out, there are some early successes and benefits that he point to:

The one place that we've had Natural Language Search running for a little while has been our Homes & Villas by Marriott Bonvoy site. This was one of those test cases we did, and we actually launched it about two years ago. Our original plan was to run that for a short period of time and then move to our bigger services, marriott.com and our app, but there was so much to discover and things were changing so quickly that we took our time a little bit to learn.

It's been interesting to track customer behavior and kind of the results on that site, he goes on, citing two general observations:

Number one is my expectation that it's going to take a little longer for customers to adopt than we expect. We’ve had this out there and the #1 search that the customers still do in the Natural Language Search box is basically what we call rate, states and space...It’s the same inputs you put into the search bar today.

It’s the so-called ‘tyranny of the search box’ in action, he argues:

We've all been living with this for so long that the behavior change is going to take a little while. We've even done things like done suggestive prompting on Homes & Villas like, ‘Oh, here's the things you can ask it. You can discover homes like this’.

The other observation Pinto makes is that when you open up the box for customers to let them ask whatever they’d like, the response is a lot of questions that may have nothing to do with a specific desire to find a certain hotel or a certain home:

We get questions about, ‘What's my Bonvoy loyalty balance?’, or pet policies at different hotels, you name it. So for us, it's very important that we build a really integrated, really great experience that's very rich, because we want our customers to love this when they go there. We want to make sure that we can predict where they're going to go and help them get the answers that they're looking for.

Going forward, Pinto reckons that the two primary metrics Marriott looks to will be traffic and site stickiness, both in terms of direct channels and the partners that the chain is working with:

I think there's a big conversion opportunity here. [We] don’t know what that is yet, but just the fact that you're going to be able to really find the hotel you're looking for, the experience you're looking for. Within that, too, is then ancillary kind of basket size. We're one of the biggest restaurant, golf, spa operators in the world. We have all these great products. Our #1 challenge has always been in this very limited space, how do you expose that to customers in a relevant and very personalized way. That's all getting unlocked with what we're doing.

Overall, it’s clear that AI is a destination of choice for Marriott that is still taking shape. As Pinto notes:

One of the guiding principles at Marriott is this idea that success is never final. So there's always still more work to be done