Some hear Matterport and think, obviously of real estate, 3D tours, Zillow, Redfin, and whatnot. Can you give us a point of view on what's going on in the housing market and highlight the impacts that your tools is helping to alleviate some of the pressures there? About half the business that we have comes from the residential real estate market globally. Most of that is, of course, in the U.S., where we were based and got our start. And the other half is in industrial and commercial and travel, and we can talk about that too. In the residential real estate space, as the market gets more challenging to sell homes, days on market extends. W e know a lot about interest rates and mortgages, I'm sure. Agents and sellers have to work a bit harder to trade those homes, and there are a number of studies that show that homes for sale with a Matterport digital twin associated with them sell faster and sell at a higher price. A ctually, we're growing in a more challenged residential real estate market. We're up double-digit% year-over-year in residential real estate, while the number of listings in the United States has declined by 30%. Right. So not only are we actually helping sellers find buyers, we're helping the deals get done. It's also helping to grow our business, and I think we're largely troughing in the sort of listing volume cycle around now. There are a number of forecasts by others saying that listings should go up about 10% to 13% next year. I think for us, that'll just be another tailwind on our business. Okay. Now, you also mentioned, beyond just the 3D tours, there are some really interesting use cases we've talked about in the past for digital twins beyond 3D tours. In your view, what are the most interesting extensions or potentially transformative use cases that Matterport is being used for today? The other half of the business, as I mentioned, is non-residential real estate. It's broadly facilities management, meaning factories, warehouses. It's retail operations. It's travel and hospitality, like this hotel that you mentioned, and Airbnbs and short-term rentals, it's insurance. One thing to note about the Matterport digital twin is that, when you and I see it, let's say, on a property website like Zillow or Redfin, we're seeing just the visualization layer, what the humans can see. It's the pictures, the high-definition resolution photographs and model. But behind every Digital Twin is a set of data. We have data on 35 billion sq ft of real property in this world, more than, I think, all the commercial real estate firms combined. That data include all of the floor information, the size, the material, the quality. The same with the plate heights and the walls and the ceiling, all the furniture, and including all the soffits and the windows and their orientation in the world, their sizes. We have all of this information, and so it's behind what we see as people looking at digital twins. But that's incredibly powerful if you're trying to renovate a property, or you're trying to adjust a manufacturing line to add a new tool into a factory, or if you're trying to underwrite insurance. And so the non-residential real estate exciting use cases are these, and that's where we're seeing even higher growth than what we're seeing in residential real estate. S trong double-digit growth in facilities management, and retail, and travel, and insurance. It's because of all these data that we have, this unparalleled, I think, data library of billions and billions of sq ft about properties, not just a customer's portfolio, but a lot of properties around and like them too. Right. I t's remarkable, the kind of extensibility of the product, right? You have use cases in retail where, I think Gap is one of your customers. They'll instead of sending somebody out for the holiday, "You know, hey, this is what we want all our Gap stores to look like," you literally just have them come in and show them what to set the store up like. Yeah, and have them come in. I think you mean send them a link to a digital twin. Yes, exactly. Yeah. That's exactly right. It's all virtual now, and broadly, this is what we're seeing in the real estate space. It's an enormous market that has yet to go through a real fundamental digital transformation, like shopping or music or TV has. We're starting to see firms embrace it. They're replacing the travel and the manual, taking of photos and writing of notes, the manual measurements, and the creation of manuals and visits to audit stores with digital twins and online collaboration and online note-taking. Right. There are many retailers that are using the product this way. The master store is built, it's Matterported. Right. That link is shared, as you noted, to all the store managers. "Please copy this. T he store managers can use their phone, do a digital twin of their store when it closes, probably takes them 10, 15 minutes, and now the home office can audit all of their work. It's great for branding, standardization, and merchandising. Okay. Maybe a quick overview of the rundown for what the product roadmap looks like today and some color on the near-term, long-term product roadmaps. Our focus today is largely around what additional data analytics and insights we can drive to help our customers with these major lifecycle moments. Right. We've done some announcements around generative AI and traditional AI, which has been a core part of our product since inception, for the last 12 years. W hat we're doing now is we're, I think, much clearer in understanding how the customers are using our product. Yep, let's say they're using it for fire and safety compliance. I have 2,000 stores, I need to make sure that all the sprinklers and the fire extinguishers comply with the local building codes. How that's done today, people go to those stores, and they do measurements, and they take pictures, and they write notes. We see in the digital twin, they're starting to use the digital twin to look for and label the sprinklers and do measurements. We see millions of measurements done every day in the digital twin library that we have, and so we productize those things. We work with those customers and say, "Well, we can, we can identify all of those objects for you. We can automatically measure their distances," which is one of the codes that we have to comply with. We can tell you whether that fire extinguisher is pressurized or not." We can do that all automatically with computer vision technology that we have. So that's where we're investing is taking the data and turning it into actionable information for our customers. You can get that granular. You can literally see the gauge on the fire extinguisher and see that it's charged? You actually. Yes, you can, and, and that's a requirement for some customers. They want to be able to read labels on machines as well as gauges. W e can do that. In addition, with a partnership we'll probably talk about a bit later, with AWS, we can actually include the live IoT data into the digital twin too. You can get live data If you've got a connected device, and you can also zoom in and read labels and gauges directly in the digital twin. I wanted to talk a little bit about the kind of freemium model. The free tier is doing a lot to drive adoption, nearly 900,000 subscribers, less than 10% of them paying right now. C an you walk us through maybe the use cases or what drives somebody to move out of that free tier into the paid tier? H ow much runway do you think you have ahead of you? Well, the free tier, first, well, why do we have the free tier? It's a top-of-funnel lead generation tool. for us, and it's designed to create a try-before-you-buy activity. With customers. All of you can sign up for a free plan and, you know, try a Matterport out in your living room with your phone, and a lot of people do that. You might still be business customers ultimately. We've generated Fortune 500 customers out of just this activity, and that's a very highly qualified lead. I really value having 900,000 o f those free subscribers. We look at, and we have algorithms that predict when and how they'll convert to paid. With those data from the data scientists, we then generate campaigns to nurture those subscribers, to encourage them to take those acts that we know, if taken in a certain time period, will help them convert to paid. It's part of the marketing funnel, and it's a campaign-driven or programmatic-driven, intentional, you know, sales process. To encourage them to come up into the paid subscriptions. And then, of course, along the way, if we see a real enterprise business opportunity that would be appropriate for our direct sales team or our channel teams- We'll siphon that person off into those flows, and they'll get a call. And, you know, we did a very large deal with an oil and gas company for all of their service stations in Europe, 40,000 of them. Just off observing that this person was creating digital twins at roadside gas stations. And so the inference is, that's probably not a consumer like you or me. This is not what you might do on a weekend, but I bet that's a business application. We investigated, it was. Now we've got another large enterprise customer. Awesome. Is the smartphone a sufficient capture device, capture technology, and enough so that you can upsell them on that, and that's enough? Or perhaps do you also need to upsell them on the capture technology? Can you talk a little bit about your hardware? I think you just recently introduced a camera rental program as well. What we're doing with hardware is we're trying to be very agnostic about how a customer creates a digital twin. Try to create. Another way to think of it is an indifference curve for the customer. The customer may wanna use a camera, use a smartphone to create a digital twin. They may wanna use a professional-grade camera that's got LiDAR and millimeter accuracy. They may wanna not do anything at all and just have a subcontractor do it or one of our service techs do it that we contract out for. Y ou know, we've got a campaign in sales that we call Always Win. Whichever way the customer wants to get onto the platform, we enable that. I think that's core to kind of how we operate. That's why we're focused so much on growing subscription revenue, in part, and open to having services revenue and product revenue be fairly balanced. You don't have to have a special camera to create a great digital twin. That's another part of your question. You can create a digital twin of this room with a smartphone. You can do home listings, Airbnb listings. You can do anything. What limits its capability is really the lens and the technology behind the lens not the Matterport cloud software. D epending on your use case, you might prefer a smartphone or a 360 camera. It's convenient. 360 can be very fast. It can be totally sufficient for your promotion, let's say you're promoting a hotel or a restaurant. But if you're cutting cabinets or perhaps you're billing or leasing based on square footage or, you know, drywall installed. Maybe you wanna have something that's inch or centimeter or millimeter accuracy. The platform works with all of those type of solutions. You know, the customers can basically just choose what works best for them. When you're talking about these millimeter precise technologies and partnerships, most recently you became an Autodesk Premium Partner. I n that vein, can you talk a little bit about, remind us the types of partnerships you're striking? A s you strike a partnership, is it. Clearly, it's utilizing, you know, some of your CAD, you know, ability to really drill down on the data you referred to earlier. One of the things that I think is really powerful about Matterport digital twins, to kind of set the stage here for especially in construction and architecture, is that most companies and owners of properties do not have as-built drawings of their property. That's for a variety of reasons. One, a lot of people just don't spend the time to create them. A lot of times, the site changes are so dramatic that the architectural drawings aren't accurate anymore. And so when a person or a company goes down the path of doing a change or renovation of the property, they don't have accurate information to start from. Well, with a Matterport digital twin, you do, and it's cost-effective, and it's quick, and we have file format conversions that allow for direct integration into AutoCAD, into Revit into BIM software, 3D design software. And so for the, I think, one of the first times, you can have a almost instantaneous, very affordable, as-built set of drawings, or digital twin from which to start to create your renovation. You don't have to pay somebody to do that, and that can take weeks, and months, and tens of thousands of dollars in the past. That's the first thing. W ith these partnerships, like with Autodesk, and we're in Construction Cloud or Procore, AWS TwinMaker, having that integration is very powerful because you've got an as-built, ground truth digital twin that everyone in the project can collaborate on. Some of these very big projects have people all over the world working on them. They can do it remotely. They can take notes. They can view the digital twin. They can spend all the time that they want in the digital twin, which they can't do on site. They can submit requests for information, requests for changes. All of these things that typically happen via text or paper or email, all can happen on the virtual location in the digital twin down to the inch or the millimeter location. "No, I don't want the wall there. I want it here," or, "I need... Do I need an outlet there? You know, all these kinds of questions are in the twin in the precise location, and then now it's documented forever. T hat's one of the things that makes the kind of the Autodesk integration pretty powerful. We've got a similar integration with Procore. I n the industrial space, we've got a similar integration with AWS, which is more focused on real-time operations, diagnostics and management around IoT. Okay. As you mentioned, 11 million spaces under management. That's growing 28% to 30% year-over-year. North of, what? 33 to 35 billion sq ft. To what extent is the, your, your Pro3 camera helping out there, or is it Are you seeing that inflect up because of the Pro3 camera, or is it, is it more the smartphone integration that's, that's doing it? We do make a camera, which you're referring to, the Pro3 camera that we have it contract manufactured. We do that because we've got a very, I think, you know, bleeding-edge technology and advanced technology in LiDAR and capture at a price point that is hard to compete with. W e still offer a camera ourselves. W hat's very interesting about this camera, which we call the Pro3, as you noted compared to smartphones or 360s or our old cameras, this LiDAR camera can do a 100-meter range at this 4K HDR resolution. T he size of properties is going almost vertical up, upward, and we're seeing 3x and 4x the amount of space captured with Pro3 versus other cameras. That's what's driving the rapid increase in the capture of square feet, and that's what's creating this massive proprietary library on which we can train the LLMs. That, it's so exciting because, you know, it's sort of, it's a, it's a wonderful by-product of creating a digital twin is that we're creating 4x the data that we used to create. That's a great segue into. I wanted to talk a little bit about, more specifically about AI. You can't get away from any conversation this year without bringing that up. It helps out with the smartphone capture. We got that. But beyond, can you talk a little bit about Cortex AI? Can you talk a little bit about Property Intelligence and Genesis? We've talked about them, but I want to put some meat behind those brands. Fundamentally, we've been an AI company for a decade. This is how we create the digital twin. W e have, we call it, we branded it Cortex AI Engine. W hen digital images and the data from cameras comes into our cloud software, that neural network starts to run, and it's trained itself on ground truth information of thousands and of digital twins and thousands of buildings and millions of square feet to create a new digital twin that's very accurate. Not only can it do that, it can also create a new digital twin when it doesn't have perfect information, meaning it doesn't have all the angles of the room. It doesn't have all of the angles of the chairs we're sitting in. Some of the chairs are occluded or blocked by tables. Our system can figure out the size and shape and color and quality of those chairs, because it's seen millions of chairs, and it also has logic to create spatial relationships within a space. Sizes of windows and doors and lamps and tables can be used to create the size of the chairs. T his is the Cortex AI engine. It's processing and creating, through an iterative process, these wonderfully accurate digital twins. T he final point on Cortex that I think is meaningful is that we can do this, we can create dimensionally accurate copies, even if we do not have depth information about the space we're working on.Y ou can use a smartphone that doesn't have LiDAR in it, and you can still get a dimensionally accurate digital twin. I don't know that many others, if any, can do that. That is because we have done so much work training our models on, ground truth information, the real world so that they can create depth and dimensions where there's none to be provided. So that's what our business is today. With generative AI, now we've got the opportunity to train LLM on our data set and create a whole new set of applications for customers. T o put meat on the bones, as you said, some of the applications we're talking about now is we've trained models to identify room types identify objects, identify dimensions, all automatically. Seems like not much, but it's actually pretty hard, and it's incredibly valuable. This is what people do most of the time on the platform is they're annotating their digital twins. We can do that automatically. In addition, we can, we have created a search bar, basically turned your digital twin into a Google database, and now you can search for anything. "Show me all of the light fixtures. Show me all of the appliances." You know, "Calculate the distance between the kitchen island and the fireplace. You know, you can just keep asking it questions, and it will answer just like ChatGPT does T hat's all kind of out there now in beta. You can read about that on our website. We've got customers in the beta, incredibly exciting prospects for general availability next year. You know, going further, you can imagine where this goes. We've got a nice video on YouTube about this that says: "Well, we can take your digital twin, and we can take all the furniture out of it. Now you've got an empty room or an empty house. We can write, "Okay, redesign my house in mid-century modern architecture. It will plop in all that furniture. It will take from RH Modern or Arhaus or Wayfair. I t will plop that all in, and it will look photorealistic. We can say, even better, "Just rebuild my house. Yeah. And I want to do it in a modern style." And it will start moving walls and replacing cabinets. And this is where I think you create some pretty powerful tools- Yep ... from the consumer if you want to do interior design or architecture, but also for the business that needs to lay out an office floor, a trading floor- Right ... a law firm, or a factory. On AI, so obviously you're highlighting great use cases and tailwinds there. Privacy is obviously important too. As you build this, perhaps the biggest data set of digital twins, can you talk a little bit about the balance between the data you use to train your models, the inherent privacy that you need, and is there an ability to take that data, maybe lock it down or secure it, but license it out and monetize it? Yeah. On the privacy side of things, we're pretty, we're pretty conscious of it, of course. These are consumers' and businesses', private spaces- Right in many cases. Not all of them are public spaces, and so, it's essentially an opt-in environment. The customers choose to participate in different aspects of that we've been talking about today. And certainly anything public, anything photoreal, that's all in the customer's hands. Like many applications, we have aggregated data, anonymized data rights to help train models. And so that's how we can give you things like your kitchen... "Here's how your kitchen compares to all the kitchens in your neighborhood. Here's what you should do to make yours more competitive if you're selling your house. We have the right to all the anonymized data. As we do specific applications, we just ask users if they want to opt into them. Frankly, if you do that, I think you get a pretty nice population of participants So we just treat each other with respect. That seems to work great. In terms of monetization, our strategy has been to add value to the existing platform and then do a pricing revision. And we did that for the first time in July. It was incredibly successful. We actually, I think you might have noted, we actually had a spike in our revenue growth rates. Both organic as well as with respect to the price change. So our approach is generally add value to the product, and customers would be willing to pay for it. So that's the main approach. But some of these more specialized applications, you know, I think you'll see us offer more nuanced subscription tiers and pricing plans to help monetize it. And then, of course, some of the big partners might want to have access to bigger portions of the data set. That's usually a bespoke deal. Okay. Okay. You continue to broaden your customer base, 100+ enterprise accounts. You just named one that you just locked down in terms of oil and gas. And these customers are spending $50K+ annually. Can you talk a little bit about your enterprise initiatives and how that's panning out? Yeah. We've retooled the enterprise approach in the last year, and created a different solution selling approach than you see on the website with the standard SMB plans. And so we're doing a consultative sale now, very similar to other enterprise software or cloud software companies. And so we're working with the customers to understand what their use case and use cases are. We're building the solution. Building means, you know, picking from our quoting tool, and that will create a price. And this is very exciting. We talked about it on the last conference call because for many, many years, we had a standard price for, for everyone. It was dollars per digital twin per month. And frankly, it was $1 to 2 per digital twin per month, regardless of whether it was a studio apartment or a wafer fab. It was very basic. Now, we've got a solution-oriented approach, I call it value-based pricing for enterprises, and we're seeing, as you noted, four-, five-, six-figure ARRs per digital twin in the enterprise. In fact, we just announced a long-term deal on the last call. It's an eight-figure deal, the first time ever that we've got a deal of that size. And so when we're building digital twins for big projects, let's say skyscrapers or airports or stadiums, or for companies that have many, many sites 10s, 1,000s, 10,000s of retail sites, it's a complex operation. They're spending a lot of money on managing those sites. They're very willing to work with us on ROI calculators and pay for that productivity gain and that savings that they're getting by using Matterport. And so we're seeing a separation. It's, you know, it's a year into it, but we're seeing a separation now between the sort of basic SMB plans you'd see on our website and the ARPA or the MRR, the ARR we're getting from the enterprise. It's pretty dramatic, and I think that has a lot of runway because every property in the world can benefit from a digital twin. Okay. Really quickly hit on, if you can, international, 'cause you've, we just highlighted one, but you've had some really good partnerships in, in LATAM, in, in APAC, in the Middle East, in, in Africa. What are you seeing in terms of your international growth? We're trying to use a channel strategy there, as you noted. Can't be everywhere, at our size yet today There are some great channels out there that, you know, other technology companies have created. So we're riding the coattails of IT services companies. CompuSoluciones is one in Mexico, for example. TD SYNNEX is a great partner of ours now. It's a global business, as many of you, I'm sure, know. And that's helping us reach different types of customers, like public sector, government, as well as different countries. And so that's working well for us. We have direct sales in Japan, Singapore, the U.K., covering Western Europe, and of course, North America. Okay. A couple more questions, and then we're out of time almost already. Your non-GAAP operating margin also improved year-over-year. OpEx is down 15% to 16% year-over-year. You're gonna hit operating cash flow break even in this coming year. Can you talk a little bit about where all that leverage is coming from? We've been very receptive to investor feedback about driving ourselves to profitability. We went public a couple of years ago on a, on a bit of a different thesis, and so we have focused on bringing our spending in b ut while maximizing growth as much as we can. And that's been working well. As you noted, we actually had an acceleration of subscription growth, in light of some of the programs I think that are working well. So it's a combination of basically three things that are gonna get us to our first goal, which is cash flow from operations, break even. One is, we've got to continue to drive revenue growth, and we talked a bit about SMB business and enterprise business. It's a little bit different, but driving operating efficiency, sales efficiency in those businesses is working, and that's both for new business as well as retention. We've actually had another uptick, 600 basis points in net dollar retention this quarter as well. One is revenue growth continues. We've got a very steady business in that regard, and it's accelerating. Two is reduction in operating expenses, as you noted. That's come down to about $36 million last quarter. I think it'll go down to about $32 million over the next year. Got a couple of additional programs in flight to get there. We executed a restructuring in July that created a step function decline in operating expenses. It's always challenging, but I think it creates the right framework and culture, and I believe we've got the right focus across the organization to get to this goal. That's number two. N umber three, a little bit of benefit from the balance sheet. Don't want to ever leave off working capital. That's a great place to squeeze out some cash flow. While we've got some, a turn we can do on inventory, for example. Any other standard, you know, tactics in the balance sheet, working capital lines will help. And so it's the combination of those three elements will get us there. I'd lastly just note that before we were public, we ran the business profitably. It was a profitable company Essentially at the same margins that we have today. So, it's a known playbook, and we're just kind of getting back to it. That's, that's a good segue into my last question, and that's just about, obviously, strong balance sheet, $430 million in cash, no debt. How are you thinking about capital allocation? Is there, is there more M&A work to be done, or do you feel fairly confident with your ability to kind of pursue the opportunities ahead of you with, with everything you're resourced with today? Y ou're right. $430 million of cash and no debt is a great place to be. Spent about $15 million in cash last quarter, excluding severance, about $10 million, and it's declining, obviously. So down about 56% year-over-year. So, we're not going to use most of that cash to get to cash flow breakeven. But what's in front of me today is what we talked about I've got great opportunities with monetizing the data library I have. I've got, I think, great maturity in the business, with people understanding what it means to be a public company and what to work on. The KPIs are, I think, pretty well known internally. I am moving ARPA up all over the enterprise. I really like where I'm. I think we'll just stick to that focus. I think it'll deliver the results that we expect. Okay. We're over time, but really quickly, in 30 seconds, we've touched on a lot of different things. Is there anything that the, that the street is misunderstanding or misconstruing about, about Matterport and, and the opportunities ahead of you? You know, well, I would just, I would just say that, you know, what I like to share with investors when, when we meet is, how well we do in residential real estate. I know we got our start there, and a lot of folks look at properties, websites, and see Matterport and see the visualization layer, those beautiful dollhouses. As we talked about today, that's just a piece of the business, and in that piece, we're still growing, even though there's a lot of news about the macro of residential real estate. I think there's a real opportunity here for this as that becomes more known and understood, for this stock to move from, you know, more of a trading with the residential real estate market and perhaps more like the software market. Okay. Well, we'll leave it there. Thank you, JD. Appreciate it. Thank you, Fitz. Appreciate it.