Can AI-based pricing lift hotel revenue? Testing the impact of Mews RMS autopilot ft. Richard Smithies & Conor Winders

[00:00:00] Richard Smithies: Hotels with the autopilot feature switched on over a

period of 9-plus months had a 13% increase in revenue per square meter over an 18-

month period versus hotels that didn't have an RMS.

[00:00:25] Matt Welle: Hi, everyone. Welcome back to another Matt Talks Hospitality.

When you deploy tech, and you do it in the right way, your hotel should be making

more revenue. But that's a claim that is very hard to prove, especially if you're a PMS.

It's the core of everything else because, also, you know, we're just a tech layer, but

maybe you had a different business strategy, you know, a management change or a

new booking engine that you deployed. So, it is very hard to pinpoint what the effect is

of the technology that you implement, and what the impact is on the business. So,

today, I wanted to bring some data into the conversation, and we ran one of the most

rigorous studies we've ever done on our product. And we included over 6,000 hotels in

this study to isolate exactly the effect of one system, the Mews RMS. And what

happens if you deploy it and you let it run autonomously or assisted by humans over a

longer period of time? So, today, joining me, I have two individuals. I've got Richard

Smithies, who's our director of data, and he's really dug deep on some of the data that

we're seeing. And then you got Conor, whose team is the team that builds the solution.

So, these people work on different sides of the past and the future. But, actually,

what's really interesting is when these things come together to design the systems of

the future. But thank you for joining me.

[00:01:31] Richard Smithies: You’re welcome. Excited to be here.

[00:01:33] Matt Welle: Conor, what is the system? What is this RMS? Like, we acquired

Atomize, what, two years ago, and now we've relaunched that same product as Mews

RMS. Is it distinctly different?

[00:01:42] Conor Winders: So, the Mews RMS is effectively a reworking of the

Atomize product, but brought inside of the Mews ecosystem. So, it's now operating

directly inside the PMS, directly on the data that our hotels and our customers are

working with. So, the big change there is that there's no longer data being moved to an

external system in order to optimize and inspect it. It all happens much more in real

time and much more reliably. And what the system is really doing is all day, every day,

it's monitoring signals from the market. It's looking at your own data. It's analyzing past

data and making forecasts about the future to try to predict the most optimal price

that you should be selling your inventory at.

[00:02:19] Matt Welle: And which data sources do we specifically take?

[00:02:22] Conor Winders: So, we are looking at competitor data in your area, we're

also looking at your historic data as well, and then we're basically looking at the

booking pace over time, so the demand and the booking pace for your particular

property relative to competitors.

[00:02:38] Matt Welle: Got it. And you said there, you know, the RMS now sits inside

Mews. So, we've just iframed the old solution inside Mews, or what have we done

differently?

[00:02:47] Conor Winders: If only it was that easy. We've rebuilt the whole experience

to be really deeply embedded inside of Mews. So, rather than it feeling like a separate

section or a separate UI inside of Mews, it actually shows up in more interesting places

throughout the product now. So, it's deeply linked into our new Mews BI, for example,

as well. So, you can see the impact of the changes that the RMS is making right there

inside the business intelligence product inside of Mews as well. So, it looks, it feels, and

it behaves exactly like the rest of the Mews ecosystem now. Nice.

[00:03:16] Matt Welle: Richard, can you talk about the research that you've done? So,

obviously, when we acquired Atomize and they've been running on our marketplace for

several years, we had huge amounts of access to data once this came into the fold

with Mews. What's the research you did, and why should hoteliers or listeners trust you

as an objective source?

[00:03:32] Richard Smithies: I suppose the first thing is just, let's just frame what the

question was that we were trying to answer here, right? So, what we're trying to

understand is if a customer takes Mews RMS, do they see any benefit in terms of their

performance, number one? And secondly, are there any combinations or optimizations

the way that the product is used that actually really help to drive any change in

performance? So, we're looking for an overall kind of understanding of what's going on,

but we also want to help those customers that have taken Mews RMS to make sure

that it's set up optimally for them. And so, it's quite a simple question, but the analysis

is really quite complicated, I should say. You can get a simple read on this by looking at

all the hotels that have taken Mews RMS. You can look at what the revenue was

moving like in a pre-period before they did that, and you can look at what they were

doing in the post period. But it's just horribly simplistic and subject to so much noise.

So, for example, that hotel that took Mews RMS, they might have been on an upward

trend anyway in terms of their performance. So, all we're seeing is a continuation of

that. So, was that due to Mews RMS, or wasn't it? It could be that they were entering a

period of seasonality. And so, you know, we mistakenly assign all of the benefit to the

seasonal movement, which, again, would be a mistake. Or it could be that there's a

market head or tailwind that's affecting performance. So, just looking at each hotel that

took the product in isolation is not smart. So, what we need to do is we need to find

control groups. In the same way that we would if we were conducting an A/B test or an

experiment, you'd have a treatment group, the customers that took the thing or did the

thing, and then we'd have a control group. The problem is with something like RMS or

product adoption, where this thing has taken place over time, and you don't have that

perfect setup where you can just give some hotels Mews RMS and then just restrict it

to other groups. That wouldn't be fair, but also, like, you know, it wouldn't be

commercially smart either, right? So, the reality here is that what we have to do is use

different techniques to measure this. And so what we do is we use a technique called

causal inference. And what causal inference does is it effectively tries to find natural

control groups, groups of hotels that look very similar to the hotels that took the

product but didn't. They're in similar markets. They're a similar type of hotel. And also

that their revenue is tracking in the same way, in a parallel way, in this pre-period, so

that we can see not only that they're similar types of hotels, but that from a revenue

perspective, they're performing the same way. And then we look at the customers that

took the feature, Mews RMS, versus those that didn't, and we track the difference in

the revenue. And we did this. And initially, when we did it, we ran this for all hotels with

Mews RMS and versus all of their controls. And we found that there was some uplift,

but it wasn't massively statistically significant. And so we went deeper into the data to

try and understand, well, what is it that's actually really causing positive change for the

hotels where we see that happening, and what are the differences for the hotels where

we don't see any change? And there was one feature that stood out massively, which

was auto recommendations and auto pricing. So, as Conor probably explained, Mews

RMS has this auto pricing feature. When you deploy…

[00:06:42] Matt Welle: Autopilot, meaning it just runs entirely automatically. Right?

[00:06:46] Richard Smithies: Exactly. What we found was that for hotels that had this

switched on over a sustained period, in our instance, about 9 months, in terms of when

that starts to show up as being really significant in data. We got this very statistically

significant finding, which was that hotels with the autopilot feature switched on over a

period of 9-plus months had a 13% increase in revenue per square meter over an 18-

month period versus hotels that didn't have an RMS. And to give you a sense of the

statistical significance of this, the chances of that occurring randomly are 7 in 10,000.

We're quite confident in the result.

[00:07:22] Matt Welle: Why the 9 months? So, like, why are we not being able to show

that in after, like, 3 months, for example, if they switch on the autopilot, that they see

this uplift?

[00:07:39] Richard Smithies: Yeah. I mean, there's a number of reasons for that. I mean,

first of all, there's just a natural delay between the pricing change happening and

revenue showing on the hotel's books. Some hotels they wanna test the feature, they

don't necessarily switch on the feature straight away, so they they wanna kind of look

at it.

[00:07:43] Matt Welle: Like a trust kind of building.

[00:07:45] Richard Smithies: Yeah. Exactly. And then, also, it generally does take time

for the algorithm to learn. You know? It's actually in that learning phase; it's running

thousands of pricing experiments across all of our hotels in all sorts of different

markets. And so, 9 months isn't the magic point when this starts to add value. It's the

point where we see it show up statistically significant enough in our data that we're

confident of the results. The actual benefits come much earlier than that. But, really,

you know, to have a result that we wanna stand behind and that we're confident about

and that we feel okay to talk about externally, you know, we allow that period of time.

In a sense, it's me being super cautious, but I think it's right. I don't wanna kind of go

out there and present a result which I'm not a 100% confident about. And I'm 99.93%

confident in this one.

[00:08:31] Matt Welle: I love it. So, specifically, the feature autopilot, when you say to

the system, yes, I would like you to make the decision. I would like you to change the

rates without human interference, is the thing that was the trigger. But we also have a

mode where the system recommends a price, and a human just looks at it, saying,

“Yeah, direction, this looks good. I'll accept it.” So, what's the difference in output

between those two modes, and why do we still recommend the autopilot to be the

better mode?

[00:08:56] Richard Smithies: So, one of the things that's really interesting, when we

look at the group of customers that took Mews RMS, and we look at their match

controls, the average number of pricing changes that we were making per month in

those control hotels in the kind of pre-period was, I think, it was about 125 a month,

right? And when we look at the hotels that took Mews RMS, it was a 120, right? So, it's

roughly similar, and you'd expect that because they're controls, right? And then when

we look at those customers that have Mews RMS, and they're deploying pricing

recommendations, but they're doing it manually, so they're waiting to see what the

system says, but then they're effectively making those changes. It goes up to

something like 500. Right? Straight away, 500. So, that's just a massive difference.

When you then actually let the system just run, it goes up to 1,700. So, there's a 14x

increase in the amount of pricing changes that are actually being made, and that's an

average. So, if you actually look at this, for the top quartile where we see the kind of,

the biggest changes, the average there is 4,700, 30x times what those hotels were

kind of doing before they took out Mews. Now the thing is, obviously, I don't wanna

dismiss what a revenue manager does, right? They're experts in the field. They

understand a lot about the market. They understand things that AI doesn't understand.

They can understand the kind of events, or it's a very sort of market specific things. But

it's almost impossible for a human to be monitoring the market every 5 minutes like

Mews RMS does. That's how often it tracks competitive pricing changes. A human

can't do all of those pricing experiments constantly and learn. And so, what we see is

that this combination of Mews RMS with a revenue manager running auto

recommendations produces this, like, really quite significant uplift. Manually, we did see

some uplift, right? But it's just nowhere near as strong. And that makes sense, right,

because we're not using the system to kind of the full power that it has.

[00:10:53] Matt Welle: One of the feedbacks I always get from hoteliers who've used

revenue management systems before, they said, “Well, yeah, but it got this one price

recommendation wrong, where it priced it really, really high.” And that, you know, I

knew that there wasn't demand in the city. But, actually, the RMS would realize this

because there wouldn't be pickup, and then it would readjust that price automatically.

So, where the human is using it as a proof point, saying why the RMS doesn't work, it's

like, no, no, we're just doing a price test to see whether there is actually demand at this

price level. And if we see that the elasticity is not strong enough, then we adjust the

price again. And because we constantly reevaluate, like, that's the testing that we're

doing to figure out, well, how far can we push the algorithm? Is that a correct

understanding of it?

[00:11:31] Richard Smithies: And, also, I would say, like, a good revenue manager would

never just let the system go off and do whatever it wants. They would set constraints,

right? They would say, “Okay. Like, no more than this kind of maximum price, no more

than this minimum price.” But I'm not an expert on the system here, by the way, Conor,

is? I just feel like I'm the one doing the talking. So, what do you think, Conor?

[00:11:50] Conor Winders: You're doing an amazing job, Rich. But that's it exactly. I

mean, there are controls inside of the system that kind of help with this sort of thing.

So, you can have minimum boundaries, maximum boundaries, and so on. That will keep

the algorithm in check with what you, as a revenue manager, want it to do. So, it will

push the boundaries as far as you allow it to push the boundaries. And you're right,

Matt, in that. Sometimes it will set what appear to be very high prices. That's actually

based on data and what it predicts at that point in time. And if it realizes that that price

isn't achievable, it will adjust it back down. But you can actually constrain that picture

in the first place directly inside the system.

[00:12:25] Matt Welle: So, once I've allowed, I was at a hotel, and it didn't have an RMS

before. Then I switch it on. And immediately, I go on to autopilot. And then 9 months

later, there's an uplift. Can you talk to me about what the uplift would be? And is it the

traditional RevPAR uplift that we see?

[00:12:39] Richard Smithies: It's a 13% uplift after 18 months in revenue per square

meter.

[00:12:43] Matt Welle: Total revenue, not room revenue, right?

[00:12:45] Richard Smithies: Total revenue. Yeah. Exactly. That's a really good

differentiation. And so, yeah, and the next question you probably asked me is, why do

we use that rather than RevPAR on?

[00:12:53] Matt Welle: You're very good at this podcasting thing.

[00:12:55] Richard Smithies: Yeah. I mean, like, that's an interesting one. I mean,

RevPAR is, we live with RevPAR. Right? RevPAR is an industry standard metric. So, it

makes sense to talk in RevPAR terms where you can, because people understand

instantly what you're talking about. But I think, you know, hotels are more than just the

rooms that they have. Right? They have restaurants. They have bars. They have events.

They have spas. It sounds like I'm about to enter the worst rap ever, by the way, but

trust me, I'm not about to do that. So, there's all sorts of different sources that revenue

can come from. And so we believe it's important to track revenue changes on all of

these things. And I think the thing about where this is particularly appropriate to Mews

RMS is that because of the way that it works, Mews RMS can set higher prices

sometimes, where it understands that demand is high, perhaps higher than a manual or

a revenue manager might actually set the price, as Conor suggested. But, also, it might

set prices lower on those shoulder nights. And so what we actually saw in the analysis

was that RevPAR was up for these hotels that took Mews RMS. So, was the average

daily rate, and so was the occupancy rate. So, it's really unusual to get all of those three

things, and that's because of the way that it works. But it's the combination of not

throttling demand, not throttling occupancy when demand is high, but also getting

occupancy still running at a high level when demand is lower, which means that,

generally, you've got more people in your hotels. More people in your hotels means

more people using restaurants, and bars, and everything else. And so, that then helps

to contribute to both optimal room pricing, but also kind of overall revenue, and so,

that's why we use revenue per square meter.

[00:14:33] Matt Welle: And it's a metric that we've been talking about. We've never

actually visualized it inside our system, this TRevPAM or TRevPAF , I guess, if you think

in feet in America. It's a really critical metric because it shows your total revenue as a

hotel per available square meter. So, we think about the real estate, the building, and

how do we optimize the building in the best way possible. So, one thing is that we, you

know, work with the room rates. But another thing is to look at the building holistically.

And, Conor, I'd love to hear you talk about the Mews OS and how that supports kind of

what the RMS is driving through other, maybe, products where it's driving some of this

revenue uplift.

[00:15:08] Conor Winders: So, there's a lot to come in this area, and you're starting to

see over recent weeks and in coming weeks as well some of the power of the RMS

being deeply embedded in the OS with things like how we do group pricing and how

we do group displacement. And you'll see shortly the ability to, as you do a quote for a

group inside the RMS, actually create the availability block inside the PMS at the same

time. So really, really interesting, interplay there. But what’s happening in that situation

is the RMS is calculating a price, the optimal price for a group based on an assumption

of additional spend inside your property at that point in time. So, you get to weigh up

whether it is better to sell a room at a particular price to a particular individual or a

couple or whatever it might be or sell a block of rooms based on the additional uplift as

Richard was talking about, the revenue per square meter, revenue per square feet that

you would likely see as a result of taking that group versus taking the individual

bookings. And because this sits inside the OS and can work with things like availability

blocks, it's much, much faster, much more reliable than trying to do this any other way.

A lot of our customers try to do this sort of thing with spreadsheets or disconnected

systems that aren't accessing that real-time data and the kind of richness of the OS.

And what you'll see throughout the rest of this year is the ability to price things beyond

just rooms and beyond groups and things like that, but actually price everything in the

hotel is part of our vision. Being able to price your meeting rooms, being able to price

co-working and things like that, even parking, and so on, hot desking. Anything that is

ultimately something that you can price, we want the RMS to be able to help you in

doing that pricing. And again, adding up to that kind of holistic picture of revenue per

square meter or per square foot for your entire property. And that's really where we see

the kind of vision and the key benefit of being part of the full OS as opposed to sitting

on the side and only being able to look at room rates.

[00:16:59] Matt Welle: And, Richard, there's many other RMS solutions in our

marketplace. Do you see a difference between different RMS solutions and the Mews

RMS?

[00:17:08] Richard Smithies: So, what we saw was when we looked at Mews RMS with

the auto recommendations switched on for a sustained period of time, and we

compared that to competitor RMSs, because we can see in our data competitor RMSs,

we saw that there was an 11% uplift in the revenue per square meter over 18 months.

But the reason I'm sounding a bit hesitant is because that's not really a fair comparison

because we can't see how our customers are using those competitor RMS products.

We can't see their activity. We can't see if they're using a similar kind of autopilot

feature. If we could, then we'd be able to draw a fair comparison. So, the results are

kind of interesting, but it's not something that I can say definitively. I think what we can

talk about is, you know, what the difference is in our product versus our competitors,

and I suppose Conor can talk about that. But, you know, in the data, I just don't feel

confident enough to kind of lay that claim.

[00:18:04] Conor Winders: I think, you know, if, firstly, if I were recommending an RMS,

I'd be recommending Mews RMS. But if you were looking for an RMS anywhere else,

there are certain things that I would be looking at, and it's even the ability to do an

autopilot mode. Many RMS systems don't actually offer such a capability, and many

don't offer that real-time pricing that you get with the Mews RMS. Where, as Rich said,

you know, when autopilot is on, you're seeing somewhere kind of in the region of 1,700

price updates. Many RMS systems that sit on the side are doing one or two price

updates a day, and that is a very, very different world that you're operating in. So,

they're the sorts of things that I would be looking at when weighing up what is the right

RMS solution for a particular customer.

[00:18:45] Richard Smithies: Yeah. And a fun fact, actually, is that 17,000 price

recommendations are made a month, actually. It's just that they're not distinct pricing

changes. So, what they're actually doing, a lot of those recommendations are simply

checking that the price is correct. But, actually, you know, the number of changes of

prices, you know, 1,700 on average, but it just goes to show how much the system is

kind of reading and making assessments about pricing.

[00:19:15] Matt Welle: And, Conor, when it makes these price recommendations and

the hotelier is, like, not trusting it, they're like, I don't know if I agree with this. Can we

tell them what the logic was behind why we made a certain price recommendation?

[00:19:27] Conor Winders: Yeah. Absolutely. And this is actually an area we've invested

a lot in and continue to invest in going forward because it is one of the kind of key

enablers in building that trust to go from a system that's telling you something, and to

a system that you actually let run your pricing for you. Because at the end of the day,

you want to feel comfortable that if a customer or a guest gets on the phone, you can

actually explain why the price was set the way that it is. And so, the way that we do

that at the moment is we use a lot of the logic that goes into our model, and we

generate explanations for every recommendation that comes through to the app that

will actually tell you whether we saw competitor pricing, whether we saw demand

surge in the market, whether we saw seasonal changes and things like that. And that

ultimately led us to the price that the system is recommending. We've seen that make

a big change in terms of customers trusting the system more and moving on to that

autopilot mode, but we do think that is an area that warrants further investment, and I

think you'll see that improve a lot in the coming months as well.

[00:20:23] Matt Welle: Because today, I think only 55% of our customers have the

autopilot completely switched on. What do you think is holding back those 45% if the

data proves such a strong point?

[00:20:32] Conor Winders: Ultimately, a product like this does come down to trust.

And if you're somebody that this is your job and you understand this intimately, and

even things like, you know, understanding who your competitors are, why they're your

competitors, why a particular guest would choose you over competitors. So, like,

revenue managers have this rich, rich wealth of knowledge about their property and

their market that it can be very hard to accept that some of that knowledge can be

encoded into a system to do parts of that recommendation for you. And it does take

time to actually build that trust. And that kind of comes back to some of what Rich was

talking about in terms of when you turn on RMS, or use RMS at first and see some of

these recommendations. It can take a little while for the data to prove the value of

them, because you go on that journey of trust and eventually, allowing the system to

run the pricing more autonomously. But, ultimately, I think we can do a better job as

well, and we're starting to do this in explaining to our customers and explaining to our

users inside of the product the opportunity that's on the table based on the

recommendations that the system is making and that you're accepting or not

accepting. And we've done a lot of work on this earlier in the year as well, which starts

to show the difference between when you accepted a price or the system set a price

for you and when you didn't. And that data is starting to show directly in the product

too, to help kind of customers on that journey of trust.

[00:21:51] Matt Welle: A couple of years ago, I remember when hoteliers wanted a

revenue management system, they would need multiple months of data validation in

order to trust that the output of the algorithms is actually correct. What does

onboarding with Mews look like if they're already on the Mews PMS for a couple of

months? Like, can they just switch it on, or do we need to do all of this heavy data

validation that we used to do a couple of years ago?

[00:22:12] Conor Winders: This is one of the big benefits, again, of being part of the

Mews OS, is that the system can just be turned on, particularly if you're an existing

Mews customer, and that data is already there. It is literally a couple of buttons that

you need to click and selecting some of your competitors, and selecting the rates that

you want to optimize, and you'll be up and running in a matter of hours, really, because

the data is already there and the system already has access to that data. One of the

other big improvements that we've made this year, as well, is for a customer who might

be new to the entire Mews OS. So, there, you have options to actually upload some

historical data that can get processed very quickly. And, again, you can be up and

running in a few hours. But we've also built a new machine learning model that helps to

predict what your historical data would have been based on what we see in the market

as well. Now there are various levels of accuracy in how quickly the model is gonna

tune itself to give you the best possible prices. When you're inside the Mews OS, and

we have the data there and then, that's the fastest. When you upload some historical

data, the second fastest. And then when we make predictions, the third fastest. But,

ultimately, you're talking about a matter of hours to maybe a day in order to get up and

running with useful price recommendations.

[00:23:22] Matt Welle: And then my last question to Conor is, like, you know, what's on

the roadmap? Like, you talked briefly about some of the group functionality that's

coming. I heard you say something about pricing up other elements of a hotel, like

parking, for example. What does the next, you know, 6 months look like?

[00:23:35] Conor Winders: Yeah. I think, again, going back to what Rich talked about in

kind of total revenue and optimizing the entire footprint of a hotel is really our vision.

And that's something that we're investing heavily in over the rest of this year to be able

to optimize those different spaces and different space categories inside of a hotel. The

other thing that is maybe more on the backend, but will result in better results for our

customers, is more rich sources of data coming into the app as well. This is ultimately,

fundamentally, a data product where the better the data, the better the

recommendations are going to be. So, we are looking at better market insights, better

events data, and so on, the ability to bring in more competitors to bit and even in

situations where I might say my competitors are these four hotels, where the system

can categorically prove that you have competitor sets that you don't even know about,

and be able to bring in that sort of data as well. So, a lot of this is going on in the

background that will ultimately lead to better recommendations throughout the rest of

the year.

[00:24:32] Matt Welle: So good. One of the things I love about Mews is the scale that

we've now reached. We've got Mews deployed in, I think, over 10,000 PMS hotels, but

also 5,000 other customers that use other products. And you now have access to this

incredible data. And then, with AI, we can, like, mine that data. So, what other exciting

research are you working on that we will be talking about soon on another episode?

[00:24:54] Richard Smithies: I mean, there's so many things that are going on, right?

So, like, for example, last week, one of my team members was working on looking at

the fintech side of our business, looking at Mews terminals. And she found that when

we look at customers that use Mews terminals, they've got a 50% lower chargeback

rate. Like, that's just to a hotel because it's effectively just coming straight off their

bottom line. So, we're halfway through a study on the digital side of our business

where, effectively, we're looking at our guest portal, which tries to understand how well

customers are going through the digital journey of check-ins, using kiosks, etc. And

what we're seeing there is that there's a 5% uplift in RevPAR versus our control groups

overall for customers using that product. But for some customers, depending on how

they've optimized the use of that product, that increase can be way plus 10% if a

customer has, for example, set the kiosks in the right way with the right images, if it has

the right type of emails, if the hotel itself offers products that lend themselves to

upselling, like early check ins, breakfast, etc. These are, like, super interesting, and they

differ massively by the types of hotels that adopt these products as well. So, yeah, I'd

love to talk to you about that in a subsequent podcast.

[00:26:15] Matt Welle: This weekend, I was going through Claude, and I said, “Give me

all the hotels per peer category that are running the highest revenue total revenue per

available square meter.” And it gave me the names of the hotels because I'm like, I

wanna speak to these hotels. Like, they are doing something special. So, we talk about,

you know, massive datasets, but sometimes the individual stories of those hoteliers

that are at the top of the bar in one of those peer groups are actually really interesting.

So, I'm using your data to actually find hotels that I wanna interview on this podcast to

figure out, well, what could others learn from it? Because I think a lot of it, we're sitting

in our silo in our hotel, and we don't have to often think about what else could I add to

my upselling workflow, for example. But, like, getting access to this data is really, really

interesting.

[00:26:57] Richard Smithies: Yeah. And for me also, it's not just about if you don't have

this product, what's the value of taking it? What I really wanna do is to create value for

those customers that have already bought it. You know, I really wanna understand for

those customers that have bought Mews RMS, how can they get more value from the

product for the guest portal? What's the optimal combination? Because we see that

when customers do adopt things in the optimal way, this is making, like, double-digit

improvements in their performance over, you know, relatively short periods of time.

And so there are a number of reasons why they don't do that, right? And our challenge

is to automate that as much as possible to make it super easy for them to set up the

optimal configurations, but also use data like this to convince them of the value of

doing such things. And so it's super useful not just to kind of, you know, try and make

our products seem more attractive to customers, but actually to get real value for the

customers that have already made the investments. You know, it’s super important to

us.

[00:27:55] Matt Welle: I talk to so many hoteliers every week, and I ask, “Hey, how's

your business doing?” And often, it is doom and gloom. Like, the economy is hard.

Labor costs are up. And, you know, the only way that we think we can increase revenue

is by driving room revenue up, like, by increasing room price. But at some point, you

can't push that price anymore. But, like, leaning into all your ancillary services is really

powerful. Last week, I spoke to a hotelier whom I interviewed for Matt Talks as well in

Texas, and I said, “How's your hotel doing?” And I was expecting, like, a doom and

gloom picture as I always get. And he said, “It's incredible. We deployed Mews, and we

put the Atomize RMS switched on, and the end results are incredible.” And I was like,

“Oh, I wasn't ready for the positivity because I'm so used to the doom and gloom, but

it's wonderful when you actually get those pockets of really good stories when people

just lean into the automation, and it starts to make a real difference to their business,

and that makes me really proud.” Thank you both. I've thoroughly enjoyed this

conversation. I really appreciate you always joining me on these episodes. I think you're

gonna see both of these gentlemen back soon because they're building some really

incredible stuff.

[00:28:52] Conor Winders: Thank you.

[00:28:53] Richard Smithies: Thank you.

[00:28:54] Matt Welle: I hope that you're enjoying this episode. Because if you do,

make sure that you press that subscribe button. It really helps grow the audience and

keep you informed whenever a new episode drops. We have some of the best hoteliers

on this platform, and I love asking them the hard questions so that hopefully, you can

improve your hospitality business as well.