Subscribe
Share
Share
Embed
Each episode of GAIN Momentum focuses on timeless lessons to help grow and scale a business in hospitality, travel, and technology. Whether you’re a veteran industry leader looking for some inspiration to guide the next phase of growth or an aspiring executive looking to fast-track the learning process, this podcast is here with key lessons centered around four questions we ask each guest.
GAIN Momentum episode #111: CRM Secrets and the Power of the Voice Channel | with Scott Buelter
===
Adam Mogelonsky: Welcome to the GAIN Momentum podcast. We have another exciting episode here where we're gonna dive into CRM. I'm joined by the CEO of Ascend360, Scott Buelter. Scott, how are you?
Scott Buelter: I'm doing great, Adam. It's good to spend some time here with you and excited to dig in.
Adam Mogelonsky: Yeah. So let's dig right into it with the elevator pitch of Ascend360. What is it and what do you offer to hotels?
Scott Buelter: Yeah. So what we are, we call ourselves a customer data platform. And the core idea is that there's a lot of data sources that hotels, marketers, sales works with at a hotel, starting with the property management system. But they'll often have a spa or they'll have restaurants, and maybe those restaurants use tools like OpenTable to get reservations. They might have a golf course, but then they're also using marketing tools like website and email. Those are all throwing off lots of data also.
So the idea is, can we get all that data? Can we centralize it in such a smart way that we de-dupe down to the individual and also the household? And so we can sort of know a lot about guests and then treat those guests accordingly, right? So if you know someone is a very high-value guest, you want to know that, and you want to treat them accordingly. And if you know, as an example, that someone is allergic to peanuts, well, you should probably know that as well, especially if they've been to the restaurant multiple times.
Adam Mogelonsky: Yeah. I mean, that presents two opposite sides of the coin for a CRM or I guess a CDP underneath it as well. You have the positive side of identification of high lifetime value guests and how you can serve back better experiences to them, or also find lookalikes. And then on the other end, you're protecting your asset by the allergies, things that may cause guest concern, let's say.
Scott Buelter: Yeah. I think that's right. You know, we like to think about it from the perspective of the guest, and that is how does the guest want to be treated? And it certainly works if you think about it well from the hotel and how do we drive revenue and how do we drive ADR. But I think the two come together in the exact same spot, and that is if you treat the guest really well, typically a loyal guest is one who actually wants to be treated well. He's coming back for a reason, and if you do that, the revenue and the profitability will follow.
Adam Mogelonsky: Yeah. So I think most hoteliers, they can somewhat understand that, but you mentioned a term earlier, the whole idea of deduping and rid of garbage data in a lot of ways. So what's your process there so that way we can have accurate records on our guests?
Scott Buelter: Yeah. So a system like ours has to have some really sophisticated algorithms to dedupe guests.
Adam Mogelonsky: Yeah.
Scott Buelter: And what that means is, you know all these things about them: you know their name, you know maybe their physical address. You might have work emails, you might have maiden names, all this sort of stuff.
And what we effectively do is put that data through an enormous amount of cleansing, and also a bunch of algorithms that help achieve the goal. So you can imagine that Robert, Bob, and Bobby to a PMS looks like three different people, but to us, we've got nickname tables, we phoneticize the names, and we can actually say, "Okay, those are actually all the same person."
And of course, you do want to get down to the individual, and so you want to know every single guest, but you also want to know their household. It's easily possible that I walk up to the front desk and ask them a question, to them I look like a zero dollar guest, yet my wife has spent $12,000 or something like that. And at the end of the day, or in reverse, you want to treat them both the same.
Adam Mogelonsky: Yeah. So the whole idea of you have the person who books the reservation for the hotel, and then you have spouse who spends at the spa or activities. Now we're talking about some very complex data connections that need to be brought into the CDP.
Scott Buelter: Yep.
Adam Mogelonsky: POS, spa, you name it, all of which have different endpoints, and you need to configure all those endpoints to have them be brought in to be mashed and merged. How do you develop those data connections, and then how do you decide which systems that you want to pursue those data connections with?
Scott Buelter: Yeah. It is actually a really, really complex thing to do. Had I been smarter when I started the company, I probably wouldn't have tried it because it's not just simply that you need to go get the data and you get to pull it in, but you actually need the data in the right state.
So let's just say you pull data from a spa. You might have an arrival date and time, right? You need to know when that is, but that might change, right? The service itself might change. And so what you need to keep doing is getting back to the source, finding the exact same transaction, knowing is its current state still like this, or actually it moved out by three hours or something like that, or they moved to also get a pedicure or something like that. So now you know what exactly is happening. And then you have to start saying, well, where do you actually put that data? Do you put the arrival date to the spa in a similar sort of location as you put the arrival date to the hotel so that the user can find and use the data really, really easily?
So I'd say we've been at this for 13 years, and we've just kept getting better and better at it now. And you might say that if we're really good at three things, this is one of them. We've already done maybe 200 different systems out there now. We ultimately do choose to do them opportunistically. So if someone comes to us and says, "Hey, we use Stay in Touch as our PMS," we're like, "Okay, great. We'll go integrate with Stay in Touch." We largely do that at our cost the first time and then have a fee for clients later.
But it takes a lot to go integrate that, right? So you need not just all the PMS data, but you need all the changes. You need that stuff to come to you in real time so you can send confirmation emails or changes to the confirmation, all those sorts of things.
Adam Mogelonsky: Yeah. You know, you mentioned confirmation emails and then the I guess you could say the pre-arrival and the onsite upselling that you may do through the CRM journeys.
Scott Buelter: Yeah.
Adam Mogelonsky: Tell us about some of the advanced journeys that you can put guests on to increase value to the hotel once you have these systems properly connected?
Scott Buelter: Yeah. So I think the core idea of a pre-arrival is sort of multifaceted, right? One of the things that you want to do is just give the guest the information that they need to get to the hotel correctly, right? If they're gonna be taking an Uber and there's another hotel named the same, or maybe the lobby is actually around the side or something like that, right? You just want to tell them those things. But once you get to sort of a sophisticated level, what you really are doing is trying to cross-promote knowing what you know about them. So if in fact you know that they're arriving on the 12th, but they've not booked a spa yet, they don't have any restaurant reservations, and they don't have a tee time at the golf course, like you want to sort of know that.
And so a lot of our clients will sort of have three boxes, and if they have a spa, it'll have like a big green check mark over the spa, and then it'll have an X over the golf and an X over the restaurant with this sort of suggestion that click here, connects you into OpenTable, book time at the restaurant.
You definitely want to know all those things, right? The critical failure that a lot of hotels do is they'll actually send that notice for a spa booking with 10% off or something, and the person already has a spa booking, right? And now they're just calling in and saying, "Give me the 10% off." And then fundamentally, if you don't even have any bookings available, all right, there's no inventory, then that's a pretty big fail as well.
Adam Mogelonsky: Yeah. That is a big fail. So you talk about these check boxes that are beside a guest profile, whether they're a spa person, golf person, all that. So one of the biggest obstacles to devising these more complicated use cases is the fact that the CRM itself can be very complex in design 'cause you have so many different fields. So how do you really simplify the user interface for hoteliers so they can know how to build these value-adding journeys without having to be a data scientist?
Scott Buelter: Yeah. I think there's a couple things. I think the—maybe an important piece is just our staff that when a hotelier asks us, like, "What would you recommend we do in your pre-arrival journey?" We'll tell them all of the tricks and the tribulations, right? So if you make a three-week pre-arrival journey and one goes out 21 days, 14 days, and seven days, like, what if 20% of your booking happens five days or less, right? They get nothing because you sort of didn't plan for that. And so we'll talk through all of those things, which are what's the recipe that you should use? What are the best practices? What do we see that works? What do we see that converts? And so that they're sort of off to the strategic side of it.
And then there's the technical side of it. And actually there's a couple things. I think thing one is you do want to make your user interface as easy as possible to use. And I think we've done a pretty good job of that. And one of the things to do with that is you create what we call aggregated data, and that is a simple field that is a yes/no field that says, "Has spa booking."
And if they do, it's a yes. If they don't, it's a no. What you don't want to try to do is get into the transactions and figure out, okay, out of these thousand transactions, does this person have one, and on what date, and that sort of stuff, right? It's just a yes and a no. But it's also true that when we start with clients, we talk about it as a journey, and we have this maturity model, and we sort of say like: Look, you're starting from siloed data where you don't have even the beginnings of a sort of unified guest strategy.
And so the first thing that let's do is unify the data, and then the next thing we'll do is implement some of these recipes in sort of a simple way, right? And that might be pre-arrival messages that maybe don't have super-duper intelligence to it, but they're getting out there. And then you sort of watch them be successful, you watch them generate revenue, and then you decide, well, let's make this more and more successful.
So one of the best tactics we've seen is to create personas for the entire email in and of itself. So if you know that your guest is 26-year-old female, and traveling in for a wedding, right? You might have a persona that sort of is speaking to that rather than the 67-year-old silver-haired sort of person who's a foodie.
And that also has a journey. Like most of our clients will sort of start with a persona that is like families with kids versus couples. And that alone, you're gonna change the images in the emails or the sort of voice you use in the text messages. So the family with kids might show kids in all the images, and when the 60-somethings couple sees those, they definitely don't convert as well, right? They're kind of saying, "Well, am I even going to the right hotel?" And so what I think it is, is it is a three to five-year journey to build that really sophisticated program, but it works really well at the beginning also.
Adam Mogelonsky: Besides these personas that you're able to bake into the email campaigns, does your system do anything with A/B testing or multivariate testing?
Scott Buelter: Yeah. So it has sort of a journey builder, and in that journey builder, you start to say, "Okay, I'd like to start here." And if the data comes in, let's just say it's an email sign-up, you say, "Okay, let me send them an email. And if they open it, let's do something else." But one of the nodes you can put in there is to say, "Well, let's do an A/B test," right? "Let me send half of these people this email and half of the people another email," and sort of watch the results over time. And then you'll sort of know what's working and what's not working, and have some strategy behind that. Like when we think about testing and when we talk to people about testing, right? You should never do a test if you don't know what it is that you're trying to achieve, right? Are you trying to create more clicks? You should have a goal for what you're trying to achieve and why. And then you sort of can watch that test happen and make some decision based on that and improve things over time.
Adam Mogelonsky: Yeah. So over time, one of the things that I would look for a KPI is return guest rate, visits. So you're bringing all this data together, you're unifying it around a unified guest profile, golden guest profile, whatever the term is, and then that's allowing you to really understand your guests to then see what drives them to deliver great stays, and then from there, that allows you to send better campaigns to then drive repeat visit rate.
Scott Buelter: Yeah.
Adam Mogelonsky: Can you walk us through some of the other tools that you have and some of the other best practices you would advise hotels to do once they have some form of data maturity to then really drive this RGR?
Scott Buelter: Yeah. So what's interesting is the sort of initial outlook when a hotel starts working with you, they might say: "Look, we've got 250,000 people in this database. Let's send an email to 250,000 people." That's gotta work, right? I mean, it's math. The coverage is gonna be absolutely great.
And what's actually true is that actually isn't something that works. That gets you into Gmail jail. You're gonna be blocked. There's gonna be a lot of bounces.
What it turns out actually works is very relevant communications to people who are engaged. And if you sort of just say that's the beginning of the slices, what you'll find is that the open and click rates are extremely high and the messages can get to a place where they are so relevant that they're actually acted upon.
So when you think about like what are actually the most relevant messages, it's the messages that are based on the actions that that person's already made. So the simplest example is an abandoned cart, right? They put something into a cart and they walked away, and so you're following up with them based on the action that they made, right? Those emails are gonna have an 80% open rate and like a 3X conversion rate, like literally can be like an 18% conversion rate while some other email might have a 4% conversion rate. And it's because you're following up with something that they took an action on, right? They raised their hand. They said, "I'm interested," but something broke down, right? The inventory wasn't there, the price was not what they were looking for, or they just hadn't even confirmed it with their family that spring break really is that weekend, or something like that, right? You don't know at that time.
But what I'd say is that there starts to be like 25 strategies like an abandoned cart or like a browse abandon or like an annual booking reminder or, you know, if you know when their spring break is, all those sorts of things start being the sort of idea that it's highly relevant to the guest.
And so what you find when you start sending highly relevant guest communications is that those guests really do convert, and they're converting through direct channels as well. So I'll give you sort of an interesting note there. You know, some of our clients have gotten to a place where their direct booking rate is 75, 80%, and that's one of the things that they're watching.
But what also happens is when we send an email, we do what you call revenue attribution, right? We say, "Hey, we sent Adam an email. Did Adam book a hotel room?" And when you look at the sort of gross aggregate of revenue that's generated when we're saying that that email drove that person to make another booking, the numbers start becoming almost unbelievable, right? They're so high that our clients will sometimes say like, "Should I really believe this?" You know? "Did you really drive that much revenue?"
Adam Mogelonsky: Yeah.
Scott Buelter: I think our answer is nuanced, which is like, it doesn't really matter whether it drove it or not. Like, what we're saying is these people that are in your database are loyal to you, and you're reaching out and they're booking.
Whether it was that email or whether it's just the existence of you treating your guests really well like when they left last time so that they're gonna come back. But what we know is all the people in this database are so relevant to your business that they're the people who are driving not just repeat bookings like you sort of asked in your question, but they're the ones that are driving profitable bookings.
Adam Mogelonsky: Yeah. So your system, speaking of profitability, actually looks across to see, in terms of total spend per guest. Does it actually look across to identify segments? Saying like, "Okay, these are the ones that book direct and have the most in total revenue. These are the ones that are OTA guests, but may have potential to become direct guests that have high TRevPAR." Does your system give that level of predictive analytics?
Scott Buelter: I guess I'd say only to a degree, right? It does have variables that we create which says we know this is an OTA guest and we even know that they have an OTA email still, right? So in the 45 days after they booked but before they depart, that OTA email will actually work and you can communicate with them through it.
So we know those things, but we're not necessarily trying to combine that with other variables to create something that's too predictive. And the rationale there is that what we find is all of our clients really genuinely have different strategies on what they think is high value.
And so what we do is we give them the tools to create audiences that do exactly what you described, but they're the ones who are sort of defining who they want to talk to. And maybe they'll include that they're part of the loyalty program, or maybe they'll include that they've been coming back for five years or something like that, that just sort of fit exactly what they're looking for.
Adam Mogelonsky: Wow. So yeah, it's a system that enables the hotel to do a lot with those audiences, segments, everything.
Scott Buelter: Yeah. And I think something that is a little forgotten in the conversation is, you know, let's just say you've got the PMS and the spa system and the food and beverage with OpenTable all integrating together. So what might be 300,000 guests in the PMS over a few years, ends up after it's deduped in our system with the spa data and the other data, you might end up having only 220,000 guests.
But what that changes is something that's really fundamental, right? Because you dedupe them down, is that what was sort of a 15% repeat rate becomes like a 24% repeat rate, because you just have less guests with the same bookings, and you didn't really realize that they're actually the same, the people who do return over and over.
Adam Mogelonsky: Yeah, that's, I mean, yeah, the whole idea of dirty data screwing up all your metrics and for data maturity, getting that accuracy to then really segment and pull things across.
Scott Buelter: I like telling a sort of a funny story on this. We brought a large hotel live that has a very, very successful spa, and as I was looking at the data before we were gonna show them their data in our platform, I just realized that it was just broken, right? It was not saying the right numbers.
And so we got on the phone and we sort of apologized, "We're gonna show it to you, but we're gonna figure out what the issue is." And they're like, "Well, what's the issue?" And I went to the sort of top spending guests, and the top lady that year had spent $275,000 at the spa. And we were like, "Of course that's not real."
And they said, "No, that actually is real. She owns a private plane. She comes in with friends. She buys it for everybody. So that actually is real." And you start saying to yourself, okay, once you start understanding that, and of course they all knew that guest, right? But what about the other guests who spend $30,000 or something like that? They're all really, really big numbers and you want to make sure that you treat those people impeccably.
Adam Mogelonsky: Yeah. I mean, that's a huge, huge value case for that total revenue metric and then the specific spend and then being able to drill down not just total spa revenue, but going one step further to see whether they like massages or facials, as a very crude example. Drilling down further.
Scott Buelter: Yeah. I mean that's exactly right.
Adam Mogelonsky: Yeah. So shifting gears here, another place that's sort of a goldmine of data is the voice channel. And you've developed a good integration with a call center to thereby allow voice data to be with all of its richness into the CRM. Is that correct?
Scott Buelter: Yeah, that's right. One of the ones that we've built a lot of really good integration with is Travel Outlook. And so you start saying, well, how does voice differ and why is it still so valuable in hospitality?
I kind of think I say, well, the voice channel has something that's sort of fundamentally different. It is so high bandwidth that you're solving things that can't be achieved through the UI, right? And I guess I'd say, like, maybe the core idea is that every channel requires a guest to already know what they want, and voice is the one that handles guests who don't. So when there's complexity and ambiguity, right, where you're like, "Hey, we're actually coming with three adults and two kids, and we need an adjoining room, and we can't figure that out on your website."
Or, "Hey, I really want to book this room, but I see that it's a courtyard facing." So they've got questions and they've got doubt, and that doubt can really only get resolved through the voice, right? If the agent says, "Actually, it's a really large courtyard. It is absolutely beautiful. There's grasses and little waterfalls. You're gonna love it." Okay, you've just solved something that you just can't solve through the other channels.
Adam Mogelonsky: Yeah.
Scott Buelter: So I think voice is really fundamental and maybe that sort of place that is gonna become even more fundamental, right, where today you've got an agent taking a bunch of notes that then flows into the CDP.
You know, over time, you're gonna have that just call recorded, transcribed, and then sort of recharacterized into variables in the CDP. And you can only imagine that growing over time.
Adam Mogelonsky: Yeah, you know, like it seems like with the voice right now, the speech-to-text problem isn't the hard part, the transcription nor is the summarization, but it really is the categorization of all that data. You know, this person wants an adjoining room and they're coming with three adults, two kids. That's very complex to structure.
Scott Buelter: Yes.
Adam Mogelonsky: How would you go about solving this categorization of data problem?
Scott Buelter: You know, it's actually still really hard to solve. And what I mean by that is the tools are out there to solve it, meaning that you really can have an AI sort of categorize it and you can store the data, sort of in non-structured format. The challenge really then starts being, well, how do you use it later?
So what we'd actually say at this moment of technology is you really need to know the things that you want to know beforehand, and you need to focus on those. Maybe a great analogy is this: we've had clients that use our survey tool to go send surveys all the time, and they survey people, and they ask them things like, "Do you like yoga?" Or, "Do you like wine?" Or something like that. And that's a really helpful thing to know. But if you only survey 1,000 people and you find out that six people like yoga, what are you actually gonna do with that? That's very hard to operationalize.
And I think the voice channel sort of falls into the same exact thing. If in fact they say somewhere in the call that they would like a pack and play because they have a three-month-old, that's really helpful to know. But categorizing that and acting on it in marketing later fails for a couple of reasons, right? One is that the three-month-old's not a three-month-old in six months.
And then secondly, you're just gonna have very little data to sort of apply that information at some later date. So it is actually quite tricky. And our recommendation with the current state is that we actually sort of know what exactly we want to know and we get that from the transcription as well as we can. That doesn't necessarily have to be humans doing that anymore, but at least you are sort of getting it into those buckets.
Adam Mogelonsky: Yeah, I guess you could sort of, you have the initial transcripts, then at a future point you could go back retrain a data model, as you revise what you're actually looking for.
Scott Buelter: Yeah, I think that's right.
Adam Mogelonsky: Yeah. But you mentioned earlier cart abandonment for the online booking channels, which is easy to track, develop a workflow around that if you're able to capture the email. But what does the not booking look like on the voice channel insofar as understanding why you weren't able to close?
Scott Buelter: Yeah. Maybe just sort of walking through the cycle, which is like the voice agent is talking to the person, the booking fails or they don't book. So what you actually do want is the data to flow into the CDP, and then you want to follow up with that guest via email, try to continue the process of getting the booking.
You also want that data to be able to feed back into the call center so they could make outbound calls and follow back up with the guest depending on the reason that they decided not to book, right? I mean, if they don't book because you don't have inventory for the dates that they want, that's probably not a problem that you're gonna solve.
But certainly you can solve lots of the other problems. So I think you do want to decide in advance what it is that you're capturing, and I think you want to be really smart about that, right? I think that the idea of "too expensive" isn't actually a reason, right? That's like a wrapper. It's not a reason. And it doesn't mean that they can't afford it. It might mean that they don't think your unit has the value for that price. Or it might mean that it's too expensive for the requirement that you have to stay over a Saturday when they really wanted to come Tuesday to Friday or something like that.
So we tend to think that you really want to take a lot of care in figuring out what that "not booked" is so that you can follow back up with them, and so that you can sort of know whether it's still an opportunity or not. You know, if you sort of back up and say when they're talking to them, right, this is better than abandoned cart, right? They're trying to solve a more complex problem. That complex problem is gonna be higher value than the simple problem that someone else is trying to solve, right? The three adults and two kids or extending the wedding stay or whatever it takes to actually make that complex booking, which has to happen on the phone.
And then of course, you definitely want to be tracking are you actually closing those "not booked" leads? So at first you want to know why they didn't book. And a lot of times that's just like, "I gotta talk to my wife." And then you do want to find out that they did book, and you do want to even try to compare those bookings to your other type of bookings, because almost certainly it's worth the cost, because they're gonna be higher value.
Adam Mogelonsky: Yeah, and that's always been my theory is that the voice channel guests are very high value. So when you talk about costs in terms of, I guess, the customer acquisitions of all this, we're talking about some automated tools, AI or otherwise, to really bring that data in, to make it actionable. So what have you deployed in your CRM to make that happen?
Scott Buelter: Do you mean what have we deployed to convert prospects into guests?
Adam Mogelonsky: Yeah, yeah. Just to look at the data from the guests who haven't booked from voice channel to then pull that in to have a better analysis of why they didn't book, chain of reasoning you could say. And then also to follow up with them in an appropriate manner without having to have some person in reservations read the entire transcript.
Scott Buelter: Yeah. I think the first idea is you definitely want to combine that call data with the rest of the data that is in the CDP. And in fact, when the call center agent answers the call the first time, you want the screen to pop, and you want it to pop with what you know about the guest, right? Five-time visitor, they've been to the spa, whatever it is that you want to get, so that right at that moment they can say, "Oh, is this Mrs. Greenville? I see that last year you booked a two-bedroom condo. Are you interested in booking one of those again?"
Right? You sort of get through the immediate mix. And then of course, if they are a prospect, right, they've never seen them again, right? Our CDP does some sort of special things for prospects, right? It creates their record, but it also creates a date at which the record was created. It creates a source for where it's created, like from the call center, and it starts to count days from which that record is created.
And then we'll of course have all of the interactions listed that you have with that guest. So, hey, they called that night, you sent them an email, you sent them another email, we made another outbound call, that sort of thing. Because what we sort of know about those prospects is that they're hand-raisers, right? They want to come and spend time at your property. You don't know all of the magic at the moment, right? You know, like the data that you can get on the call is amazing, right? It's constraints and budget ceiling and competitive set, and on a call, a guest is going to volunteer all sorts of things like the occasion, right? "We're coming out to see the World Cup," or, "We're coming out for a family reunion," or something like that.
Now I don't know that we've actually solved the agent doesn't need to sort of familiarize themself with what we're reading because they're gonna have to do some of that. But what we do try to do is sort of give them the core metrics, right? Last stay date, number of stays. We also include an RFM score, right? And that RFM score is like one of the best predictors of repeat booking. But I think they're still gonna have to sort of dig in and know a little bit more.
Adam Mogelonsky: Why is RFM such a good predictor?
Scott Buelter: Yeah. So to back up and say what is RFM, right? It's a combination of recency, frequency, and monetary score. So recency being like what's the most recently they've been at the hotel. Frequency is how many times they've been to the hotel. And then monetary is what their total spend is.
And you break everybody up in the database into quintiles. So you're either a five, four, three, two, or one in recency, the same in frequency. And what that actually does is it creates 125 perfectly sized segments. So you have 125 segments of people that fit this criteria.
Let's just say that you've got some of them that are 5-5-fives. They've been there recently, they've been there very frequently, and they've spent a whole lot of money. And that sort of simple analytics is remarkably predictive. And it's weirdly markedly predictive across industry.
It's also very predictive in the nonprofit world. Like, if you're looking for a donation, the person you want to talk to is the person who just donated. For whatever reason, they're, again, the most likely to donate again, which is why sometimes you regret donating because you immediately get another mailer asking you to donate again.
I know I sort of danced around exactly like why does that formula work so well. I'm not entirely sure, but it does. Across industry, it's very well known and we like our clients to take it pretty seriously.
Adam Mogelonsky: Wow. What other metrics besides RFM do you think are of the most value to hotel marketers?
Scott Buelter: Yeah. We've got a couple of ones that we think are pretty fancy, which is of course have one called like, do they have a booking on file? I mean, a future booking on file, right? That's just one that you want to use because if in fact you're gonna do the big Labor Day sale, you don't want to send all the people who already have a Labor Day booking.
But the ones that get a little bit more interesting are things like midweek bookers or midweek local bookers, right? And we actually create a score for how likely they are to have stayed weekdays that are not Friday, Saturday, Sunday days. And the higher they are on that, the more likely that you can get them back.
And so if in fact you want to get people into the hotel in short time during the week, let's just say you're looking at next week and you see that you have a very high occupancy, that's an amazing metric to use, especially combined with local, you know, drive time, whatever, 90, 120 miles. Because those tend to be people who are demonstrating by their behavior that they have the ability to come to the hotel in the middle of the week. Doesn't really matter if they're 25 years old and work from home and they like that, or whether they're retired. They're just people who have the propensity to do that.
Adam Mogelonsky: Wow.
Scott Buelter: Yeah, and we've got the exact opposite score as well, so we actually do know whether they're a sort of weekend booker. We also keep sort of an average daily rate for that hotel, so you sort of know if their average daily rate is like $852 and what you're gonna be selling is $310, you're talking to the wrong person.
Adam Mogelonsky: Yeah. No, those are incredible things to track. And yeah, it does get very scientific, but overall thesis on that is, again, data maturity. Get the data in the right place so that way you can slice it up and, you know, if you don't have a good connection with the PMS, then good luck trying to determine if it's midweek or weekend to actually midweek who's local by cleansing that data in an accurate way.
Scott Buelter: Yeah, that's right.
Adam Mogelonsky: Yeah.
Scott Buelter: For whatever reason, it just sort of reminds me that like one of the biggest mistakes we see people sometimes make is they'll switch from PMS one to PMS two, and they'll move all their guests over, but none of the transactions, right? The transactions don't really fit into a new PMS, so they just sort of get thrown away.
And then they call us and say, "Oh, we'd love to work with you." And what it turns out is we can definitely work with you, but you're just missing a lot of really valuable data. And so the sort of plea that I would make to anybody who's out there that's thinking about switching PMSs is make sure you save the data somewhere, right? We can get the data from just about anywhere, but not if it's lost.
Adam Mogelonsky: Yeah. I mean, when you say get the data, talking hundreds of thousands of records that have multiple touch points to them in terms of what was spent.
Scott Buelter: Yeah, that's right.
Adam Mogelonsky: So that's not exactly an easy migration to bring that over.
Scott Buelter: Right. Yeah. We don't know that we've ever seen somebody, you know, migrate transactions from one PMS to another. Maybe some people do it. We, you know, that's not our world. But what we do know is that when we pull data from PMS one, and let's just say there's fields like rate code and group code. We're gonna put that into a field that's like rate code and group code. PMS two is gonna have a field that means something quite similar, and so we can ultimately lay these things on top of each other.
So we don't actually care how many systems somebody has. We do have a client with 21 different systems. It's a ski resort. They've got rentals, they've got lessons, they've got golf, they've got condos, they've got hotels, they've got ticketing, right? Lots of restaurants, all that sort of stuff, so it gets pretty complicated.
Adam Mogelonsky: But what you're saying is in that transition point, because a lot of people do feel locked in by a current vendor, and what you're saying is that your CDP architecture can give people the flexibility bring records into your CDP and then write them back into the new system to help execute that migration and keep that data accuracy intact during that process.
Scott Buelter: Not quite that. It's just that if you get it into our system and then you're on a new system, the data's not lost. It's in our system. And then if they decide not to use us, like they say, "Ascend, we don't want to work with you anymore. Give us our data back," we'll just give them all their data. We never actually own their data. We just, I guess we store it for them for this valuable use case.
Adam Mogelonsky: Yeah, it's an incredibly valuable use case, particularly with so much disruption happening in the hospitality world.
Scott Buelter: Yeah, absolutely. Things are changing.
Adam Mogelonsky: So Scott, to close out here, where do you see CRMs in the hospitality technology landscape landing in the next few years?
Scott Buelter: So I think an enormous amount is changing in our world as well, and I think it is because of AI. At the end of the day, like, AI can deal with an enormous amount of text while, you know, humans can't. So when we talk about the categorization problem, it starts going away with AI.
But I would offer a core vision for what customer data platforms are gonna sort of... How they're gonna change hospitality in let's just call it the next three, five years or something like that, right? And it's this point at which the data asset, the CDP, stops being a marketing tool and a marketing asset, and it starts looking more and more like the resort's operating system, right? Everybody in the hotel should be using that data to do what they can't do, which is know every single guest.
I'll give an example is we actually do now have a Chrome extension. So if you've got your PMS, like Stayntouch up, our Chrome extension's sitting right next to it on their screen. It's reading the data so they know they're on a certain guest, and we're giving them the entire guest history. We're telling them whether they have a spa booking. If you're in the restaurant, we're telling them that they're allergic to peanuts or something like that. And you can sort of imagine a world in which every single person in the hotel is relying on this data to make decisions about how do they treat the guest and those sorts of things, right? That maybe in three years it'll actually be an earpiece into our system that tells them that guy is actually friends with the owner, right? That sort of thing. So they know, "Okay, let's treat that guy really well. He's friends with the owner."
Adam Mogelonsky: Or we're gonna bring back Google Glass.
Scott Buelter: That's right. That may be. And in fact, I was at a PMS conference and they are focusing a lot on sort of the vision,
Adam Mogelonsky: Yeah.
Scott Buelter: to try to make identity a little bit easier in
Adam Mogelonsky: in the real world.
Scott Buelter: Yeah.
Adam Mogelonsky: Yeah. I mean, it's absolutely beautiful what can be done, and think it's important for hoteliers to understand what's possible and, from a use case perspective, then find good partners who can help them solve those challenges and the value of their business at the same time.
Scott Buelter: Yeah, that's right. What I'd also probably offer is that I think the challenge is not necessarily the tech stack anymore. Like, we've got the tech to solve some of these things. It starts looking like the org chart, right? A lot of seasonal workers, you know, people who maybe don't care as much, those sort of things.
And so there will be a journey where technology sits right along with people and we keep sort of getting better together. And I think it's a great future, right? I don't fear the AI journey at all. I think it's actually gonna make humans so much better at achieving what we all want to achieve in hospitality, and that's to treat the guest really well.
Adam Mogelonsky: I totally agree.
Scott Buelter: Yeah.
Adam Mogelonsky: It's been fantastic to have you on the show and really get into how a CRM can benefit a hotel.
Scott Buelter: Excellent. I loved it. Yeah. Thank you too. I sure appreciate it.