ON-DEMAND: Your Data Driven Journey to a Smarter Retail Experience

Published on
May 3, 2022
Raffi Vartian
Vice President, Business Development and Strategic Partnerships
No items found.

This is a recording of a webinar, previously published on Loss Prevention Magazine. Brought to you by Cisco Meraki and meldCX.

In this webinar, technology experts from Cisco Meraki and meldCX discuss how video analytics derived from modern security systems can provide insight into customer buying patterns and behaviors, and be used to drive store sales and improve customer experiences.  

Most importantly, learn how you can leverage surveillance data and advanced analytics to:

  • Improve detection of visual events, including register fraud detection, shipping and receiving monitoring, damage to goods and property, and slip and fall incidents
  • Streamline the customer experience by better understanding buying habits including traffic patterns, dwell times, checkout times, and occupancy metrics
  • Create custom analytics solutions to address new, unique business problems, such as stock optimization or creating tailored buying experiences

Featuring:

  • Matthew Moreno — Product Specialist - IoT, Global Enterprise at Cisco Meraki
  • Raffi Vartian — VP, Business Development & Strategic Partnerships at meldCX
  • Jacque Brittain — Editorial Director at Loss Prevention Magazine

Transcript:

[00:00:00] Jacque Brittain: Hello everyone, and welcomehow video analytics derived from modern security systems can provide insight into customer buying patterns and behaviors, and be used to drive store sales and increase customer experiences. to today's webinar, your Data Driven Journey to a Smarter Retail Experience, leveraging Surveillance Data and Advanced Analytics sponsored by Cisco Meraki. My name is Jacque Brittain. I'm the editorial director at LP Magazine, and I'll be your moderator today. Physical security and loss prevention teams in retail have relied on video surveillance for decades to help with safety and security, but now the list of responsibilities is growing as these teams are being asked to support other areas important to the day-to-day operations of the business.

[00:00:39] Fortunately, modern video surveillance can provide the tools necessary to address these new asks. The same system that supports security and loss prevention efforts can be leveraged for additional operational benefits. Video analytics derived for modern security systems can provide insight into customer bind patterns and behaviors used to drive store sales and increase the customer experience.

[00:01:06] In this webinar, our panelists will discuss how you can leverage surveillance data and advance analytics to improve detection of visual events, streamline the customer experience, and create customer analytics solutions to address new, unique business problems. But before we begin, let's review a quick some quick housekeeping items.

[00:01:29] The webcast is designed to be interactive between you and the presenters. The console you are looking at can be completely customized at the bottom of your screen, across the bottom. There will be some widgets that you can use to ask questions. During the webcast, there's a q and a there. There is also a question or help box that can be used.

[00:01:54] To help you if you have any common technical issues that we need to review during the presentation. With over let's, let's begin by introducing today's speakers. With over 20 years of experience in the enterprise technology services and physical security industry, Matt Moreno is responsible for the development and management of the Cisco Meraki Internet obtained solutions for global enterprise.

[00:02:19] Over the years, he has developed and managed relationships with many large national projects, including mega city transformations, with joint partnerships in commercial and government agencies. Marino has ex has experience working with retail food. Food service, healthcare, manufacturing, and telecommunications clients.

[00:02:40] His security industry experience comes from all three sectors, which would include partner, manufacturer, and end user consulting. Raffi Vartian is a solutions driven strategist, passionate in helping organizations realize their growth potential by implementing Future First Technologies with more than 15 years of experience in the technology industry is help multinational companies expand markets, explore and pioneer new solutions, and deploy game-changing technology into complex environments.

[00:03:17] Through the years, Vartian has helped develop both the technical acumen to understand the omni omnichannel landscape and lateral thinking approach to problem solving, helping customers solve business problems through AI and Internet of Things technologies. Gentlemen, thank you both for, for being with us today.

[00:03:38] I'm gonna turn it over to. Thank you.

[00:03:41] Matthew Moreno: Thanks Jacque. And hello everybody. But I gotta to say 20 years, is that what it's listed on my portfolio now. , man, that hurt and growing and growing and growing and growing. 20 plus years. That's right. Oh, I am ancient now. Anyway, well thank you for everybody for joining and it's actually a quite a privilege and a pleasure to be on this not only with LP Magazine, but also with a close friend and partner Raffi on this.

[00:04:04] So I like to lay out some of my own house sleeping rules when I do these presentations. Is that, hey, there's a q and a button on there. Ask them, right? Jacque's gonna interrupt me, Rai's gonna interrupt me. You know, we like to make these as conversational as possible. Plus it makes it a little bit more fun for us as well, because, you know, hey, this is your time.

[00:04:21] We want to use it to your benefit. And that is very rewarding for us. But I'm gonna start now. I'm gonna get into this. And by the way, Jacque had mentioned a whole slew of different items that are gonna be part of this presentation. We're gonna try to get to them all right? The reality is, is there's a lot, but there's a lot to consume and a lot to, you know, to to understand and a lot of kind of, you know to kind of just discuss.

[00:04:43] So let's get started right now and we'll go through it. But I wanna start with the idea of smart spaces, right? Smart spaces is really what this. So really what is smart spaces? Well, in simple format, it's taking the physical environment and marrying it to the data environment, right? That's how at least we at Cisco Meraki identify the idea of smart spaces.

[00:05:06] So we have to say, well, are they being reimagined? Well, yeah, they are, right? What is that commercial space? At least it was, you know, 10 years ago it was very simple. It was a retail environment in a box doing what a retail environment does. It was an office setting and you know, doing what an office setting is.

[00:05:21] And then of course there was public spaces like parks and malls and recreational areas, et cetera. But now that line's a little bit hazy, right? Well the reality is it's hazy because that's what people are expecting out of them. They prefer that experience to be both commercial and retail and, you know, office and it kind of all blended in one.

[00:05:39] Right. So the stats say that. So we have to be prepared for that and as we're going forward. Right. Okay. Then of course, customers and end users, right. Their, their expectations of these types of things have changed, right? If the, if like, you see that stat there, right? All of our customers expect intuitive, safe, and secure engagements.

[00:05:56] Yes. But the experience that they're coming into when it comes to a retail is incredibly important, actually more important than the actual products that they buy. So how do you blend all of this together? Right? And that's what we are calling smart spaces, right? Tho that physical environment to the data that we're actually collecting there.

[00:06:14] And then lastly, right, the you know, the, the, the adaptability piece of this, right? Well, you've had these both brick and mortar places, in this case, retail operations and centers, and maybe even warehousing. How do you adapt quickly in the sense of deploying new technology to be able to accept and gather all of this data that's happening in there?

[00:06:33] Well, we have a lot of solutions for that, and that's really what we're actually gonna really drill down to today. Not only from the hardware side, but how do you collect it, manage it, and then display it in a solutional format. But the reality is you may have already seen a lot of these things, right?

[00:06:50] Video surveillance and using analytics on video surveillance isn't ne necessarily anything new. But with Meraki, everything's new when it comes to the cloud and how we actually manage it on the cloud, right? We're a cloud first driven platform. That's what the, the company was founded on. Of course. But you know, cameras have always been inhibiting to go into the cloud because of the idea of bandwidth and bandwidth constraints and, you know, big files and et cetera.

[00:07:15] And also there's some security aspects of that. Well, we have a really cool solution to that and that's how I started within the Meraki portfolio. So to build those safe environments, let's examine one and kind of, you know, work backwards from there. So in a retail operation, yes we're gonna have cameras and that is definitely something I'm going to focus primarily on today, but also in those environments are gonna be what we at Meraki is part of the entire full stack portfolio is gonna be access points or wireless aps maybe even some sensors, like maybe you have some of those environmental sensors today in your store, but maybe you need or Meraki iot enabled sensors.

[00:07:57] But then those are hardware devices collect.

[00:08:07] How do you deliver

[00:08:19] Raffi Vartian: sensors

[00:08:32] Matthew Moreno: access points

[00:08:39] flow, right.

[00:08:44] Did you get me on that?

[00:08:49] Raffi Vartian: I think you broke up a little bit there, Matt. I was having a hard time hearing you. Gotcha.

[00:08:53] Is

[00:08:53] Matthew Moreno: that any better? Yeah, I can hear you fine. Okay. No, I just pivoted a little bit to kind of show off the API partner as this vignette is where you would come in and like, you know, like what is it, what we're showing here in this one little easy one.

[00:09:07] Raffi Vartian: Yeah. We just, we just missed a little bit of the last thing. It's just your connection broke up a little bit. That's all it was. Yep.

[00:09:12] Matthew Moreno: Go for it. No. Okay. Anyway, so as I'm saying like, you know, in this retail vignette, kind of showing where these kind of hardware pieces would be, right. And, and the usability in any kind of like, spotting and setting for that, that's where it would Okay.

[00:09:26] Of course, Vienna is on here, of course, which we're obviously showcasing today, collecting that the data. What is the use case here in this one Raffi? Well,

[00:09:33] Raffi Vartian: it, it's a collection of things, right? You know, one of the things that we're gonna be diving into is how do you. It's one thing to collect all the data and how do you make it actionable effectively, right?

[00:09:42] So there's a connection between cameras and the data that you're kind of getting out of it. We'll go into it more when I've got a couple of slides to go out of it. But it's really it's not just about looking backwards into what has happened in your environment. It's realtime analysis of what's going on to see how people are engaging with products and solutions and really getting information in a live perspective that it would take a long time to sit there with kind of pen and paper to be able to kind of take all those things down.

[00:10:10] So that's kind of really what, what we are gonna try to focus on. Right? How do you make that enormous amount of data that you can get from video into live and actionable data?

[00:10:18] Matthew Moreno: Exactly. And it starts at that camera, right? That's the highlight of what we're actually talking about today. So at Meraki, and this is one Mariah, this is where I focus on in this specific Arena products is the Meraki MB Smart cameras, okay?

[00:10:33] They are what we call edge based cloud managed cameras. And there's some legitimate, like things I don't want to, you know, just brush over here, like scalable, secure, agile, and smart, right? Of course, we're gonna talk about each one of those, but I always like to say each one of these cameras that we put in any environment, whether it's 60 in a large, you know box store, or if it's three or four in a small bodega, the idea is that the camera is the solution, right?

[00:11:00] Overall, it's going to give you everything that an entire, you know, older end-to-end type of, you know, what we call box architecture or old video surveillance systems give you. But everything's on the camera. That's the key here. And how we do it is through this type of architecture. Right? Everything on the left of that cam, on this, on this architecture is in the camera, so a great imager and lens.

[00:11:22] Of course, all of the recording that's going to be done, what's ever in front of that lens. Yes. Everything there. And then of course, the actual chips that we've chosen. Why did we choose very specific chips? In this case, it was a Qualcomm chip for our APIs and the metadata that we're collecting so that Raffi can put his awesome solution on the actual camera.

[00:11:44] What it does, it actually gives you some real big benefits, right? If everything's at the camera, that eliminates all that other hardware that you normally have for an infrastructure, right? No more pizzas, box servers, right? No more imports, no more work stations that you need that proprietary software on there.

[00:11:59] As a matter of fact, we don't even really have software. We just use dashboard via the you know, the, the, the browsers, right? Any browser that's compatible in the world of contemporary web and then of course access to anywhere, right? If I'm saying that I'm only using browsers, that means.

[00:12:13] Theoretically remotely access any of this video. It doesn't have to be right on on prem in that machine, you know, and downloading it and working through export files and all that, nope, doesn't exist. Everything's done right there in the browser. And then smart processing, again, just gonna emphasize this as much as I can.

[00:12:30] Putting everything on the camera makes the entire story come together so that you understand what that environment looks like. But I would re remiss and I actually would probably get in a lot of trouble if I didn't actually talk about hardware security, because not only in the world of physical space for physical security and cameras, but the idea that you know, how do we, you know, in a sense secure that intelligent endpoint if everything's in that camera, right.

[00:12:57] Recording video, all of it. Right? Well, you do it on the hardware security, you do it on the trusted anchor module, on the actual device. We manufacture our own cameras. No other person in the world is allowed to touch. Or put a label on Meraki cameras cuz they're ours. We manufacture from lens to core.

[00:13:13] And doing so allows us to put that trusted anchor module directly on the camera. That's basically a a way for this, the dashboard, which is how we actually view video to authenticate the actual camera itself. All video being encrypted and, and both at rest and in motion. Basically saying that when the camera's just recording on the camera, it's a form of encryption.

[00:13:31] And then when it sends it to the cloud, it's a actually a different form. Secure user access, right. And firmware on the right. Yeah, I know that's also kind of a big pain point for geographically dispersed organizations like retailers because, you know, if you have a firmware update from a pass that needs to be deployed on an endpoint from, you know, your older video surveillance systems, it's a big heavy lift, not with Meraki and not way that we architect architected this system because we can just push that firmer directly to the camera right from dashboard.

[00:14:01] Okay. That's the key of this, right? So not only is everything at the edge intelligent, giving you great video, but it's nice and secured and that cool dry place that you don't have to worry about anymore, right into the cloud. And that's how we manage it. But we actually have a lot of different cameras to choose from.

[00:14:17] It's not just one, it's not just that little dome. We have, you know, bullet cameras and 360 cameras and little wireless cameras that can be put ad hoc. So, you know, on a different time and a different space. I'll certainly go over all of the different products and portfolios, but this is about the data that it's collecting.

[00:14:34] Okay. And the data that's collecting is key because I have to be able to give it to Ravi in a format that actually you can consume. So, but first and foremost, the purpose of buying cameras and the world of loss prevention has always been about these types of functions. You still need a camera to do traditional loss prevention things, making sure that no one's stealing something off the shelf, making sure that no one slipped and fall or you have an audit.

[00:14:57] You know, in general, keeping the public safe is what cameras are designed for. Yes, our cameras have those functions built into them. We have 'em in formats from, you know, from different types of megapixels to. Higher 4K video. Of course, all the quality, you know, is, is spread across different technologies.

[00:15:16] Yes. But additionally, even some out of the box solutions, that gives you a little bit more than what older video surveillance have given you. Like the APIs of course. Right. We do have some core analytics that you can exploit. People counting, heat mapping, things like that. Yes, of course, you know, dashboard health, that's another big thing for geographically dispersed organizations, knowing when a camera is down, because that's the worst case.

[00:15:39] If something happens and you didn't know the camera was down. Now all across the board, all across all of Meraki in the dashboard, you're gonna get all those great, you know, base and core functions itself. But, How do you get more value out of that camera? That is really what we're at here today. Right? So you understand the portfolio, you understand the architecture, but really what can you do to get that cameras in that pervasive format that we were talking about?

[00:16:06] Right. Well, first and foremost, let's start with how cameras are purchased. Usually, and in the case of history has always been the security director, loss prevention. The asset protection person has saying, Hey, does that a hundred percent of that camera cost have to come from me? But what you're gonna learn today, that's not necessarily the case anymore, right?

[00:16:27] Yes. As long as it's like. The purpose of buying the camera has always been that, you know, the idea of safety and security, and that's certainly the driver for it, but maybe it doesn't actually all have to come from that same budget, right? What if we took half of the budget and then we identified maybe a different percentage to somewhere else?

[00:16:45] If you can get the data and give it to somebody in the merchandising component, right? Looking at endcaps in that store for that matter, is that something that's valuable to them? And maybe they'll quote unquote, chip in for a portion of the of, of the cost to deploy cameras or maybe even store operations looking at risk or real estate and utilization in the back room, things like that.

[00:17:05] The camera's still functioning of what a camera does, right? It's still looking at that scene saying, Right. You know, you got an aisle, right? Making sure no one's slip and falls. But really is it really looking at those end caps, right? Is it really looking at all of those zeros and ones? Is it looking at the area that utilization on all of 'em?

[00:17:25] Yeah, it's technically doing it. Smart cameras are doing that on your behalf. We call it pervasive video because we want it to be pervasive, right? If I can make the camera insert itself in an organization Right. The way we have been describing it since start to finish here, right. Doing multiple things.

[00:17:44] That's how you get the most value out of your endpoints, right? That's what the reality is. So as that example, as I'm going through all of this, right? Sure. The base functions are there, the core business is there, but that API is the ticket. That API is your unlocking to the, to the aspects of data that give you everything that you could possibly see of what the solutions are that represent on the right side of this screen.

[00:18:07] Our ecosystem partners, our ecosystem partners are tried and true and, and really important to Meraki and the, and in the Meraki app space, very similar to how you, you know, consume apps on your iPhone. You can actually add our, the applications to the camera from a Meraki app app, app store, or marketplace, excuse me.

[00:18:26] And for that example, this is where I get the transfer over to RAI to explain what we do. That's

[00:18:32] Raffi Vartian: question. Sure. That's No, I just, I think it's really important, Matt, to define what pervasive video is because , it's, we don't want to, we wanna make sure that we're not talking about invasive video.

[00:18:40] Right? So pervasive is that it's going on all the time, right? Invasive. We've seen some manufacturers that are kind of out there that is, it's kind of like surveillance as a service and surveillance of individuals as a service. So I think it's important to kind of draw the line between cameras always on, camera's always recording, and camera's always getting data as opposed to camera's, always tracking individuals, storing individuals, and taking kind of personally identifying information.

[00:19:05] There's a big line of break that we're seeing because when you're, you've got that, you know, so your traditional security customer, right? You know, they're using it for very security focused issues, but when you start branching into operations, they're branching into marketing. There's a, there's a layer of.

[00:19:22] Personally identifying information that you don't wanna violate. So I just wanted to make sure that, that we kind

[00:19:26] Matthew Moreno: of No, it's, it's value. I mean, that's exactly right. I mean, a camera's gonna do what a camera does. That's what we wanna like, like, again, from the purpose of buying cameras in general, from the eons ago, from my beginnings in this business, 20 something years ago, whatever it was to now has always been safety and security.

[00:19:44] And it better be a reliable piece of hardware that does what it's supposed to do and do it well. The next phase of this world has then had been analytics attaching an analytic to a camera by adding more, you know, services, et cetera. But we're saying now, like the next even further phase a Web 3.0, if you will.

[00:20:00] Something fun like that is, well, what can we do on the actual end? Can we do it on the end point? Could do it on the camera, and that's the architecture that Meraki supports to giving you all great camera footage, doing what cameras do, but then also collecting the data so you can exploit it in a different measures by using the APIs that we have.

[00:20:19] For solution sets. Oh, I don't know. Like meld. There you go. .

[00:20:24] Raffi Vartian: Thanks very much for that. Matt Jacque, was there anything that you wanted to interject with? Did, started, I

[00:20:28] Jacque Brittain: have a couple questions here. When you start to look at your systems, Matt, and, and the way that they're laid out in the store, obviously when you are using the camera for security loss prevention, asset protection purposes, a lot of times you're focused on areas that are more remote, for example.

[00:20:45] Correct. You know, areas where you potentially have problems. Okay. A lot of the data that you're talking about gathering is going to be more the flow of the store and the way people move through the entire store. Which means that cameras will be looking. Potentially in other directions, in other ways it might require that we have to add cameras and look at you know, the way that we lay out our store and that type of things.

[00:21:12] When, when we're talking about these types of solutions, how do you go about helping your customers determine the best and most effective use of that equipment? When they're installed in the store.

[00:21:29] Matthew Moreno: Sure. There's a lot of different answers to that. That kind of lead into the, the it, it's a, it's a partnership, right?

[00:21:35] It's a partnership between somebody like myself who's been in this business for a long time, knowing where the best way to manage, or, sorry insert cameras in certain angles and certain lighting conditions, et cetera. As well from a proof of concept perspective. And then there's the partnership with our a p I partners identifying the use cases that he, that we're trying to solve and then position the cameras that way, or at least maximize their field of view to be able to.

[00:22:00] Get the data that we can in an accurate format. And then also it takes our partners, right, our partners and our resellers in the channel as well. I mean, you know, you can't buy Cisco direct from anywhere. We need our channel partners and integrators to be able to supply, and they are the experts in the field identifying where these cameras go into the best way.

[00:22:21] I mean, look, I could identify, you know, what camera's best in one type of store, but you know, there's, you know, a few hundred people on this call that would probably have a different opinion of what mine is about where they go, right? And so the collectiveness of myself, the end users, the. The security integrators and then our partners or APIs is how we actually transform a lot of this.

[00:22:44] But the one statement you made was about the you know, adding new cameras. Well, for someone who sells cameras for a living, that sounds like a terrific idea. However, the reality is, is that we are, we're trying to do is actually replace your existing cameras with newer, smarter cameras, right? And then, and then instead of just giving you a like for like, just to be able to use a physical security camera, looking at physical security things, well what else can we see in that view, right?

[00:23:11] What else can we extract with that data so that the cost of replacing it is kind of like superseded by the value you're getting from the data that you're getting. That's where I think it gets into the stratosphere of the ideas and what we can do with cameras in the future. That is incredibly attractive to me.

[00:23:30] Jacque Brittain: A couple. There are already a couple of questions here. Let's go ahead and address those Sure. As we get 'em. How, let's see. Is video data stored primarily on the camera or in the cloud? If stored on the cloud, how do you address bandwidth implications Number two. What is the percentage commission that Cisco charges developers for selling their analytics through their marketplace?

[00:23:59] Matthew Moreno: Okay, I'll let oh, well, okay, I'll answer the, the first one really easy. So, primarily the way the architecture is supported best is to, is to record directly onto the camera itself. That means that no large files of video are being sent to the cloud in a consistent or a continuous stream. And then what we're doing at that point is we're only sending the metadata to the cloud, which is less than 25 kilobits of upstream data directly to the to the Meraki dashboard that way.

[00:24:27] So that, that is primary. The way to do it is also just best practice, however there is not an infinite amount of storage on every one of the cameras that we sell. We put solid state storage on these things the best we can, right? The largest size that we can based on whatever the parameters are that the customer needs.

[00:24:43] But sometimes there's just not enough space in those cameras and the customer needs a large amount of storage for whatever reason, whether there's policy or policy in the organization, or if it's a piece of, you know, a regulation that they have or whatever, that you need to have extra storage.

[00:24:57] Well, we also support cloud archive. Cloud Archive is our way of sending that video directly to the Meraki dashboard and housing it for you on any increments of time that you would need between like 30, 30 days extra to 365 for a whole year if that was the case. So that's how we do it, but again, it does impact your network when you do it.

[00:25:16] So we have to kind of, you know, architect it a little bit differently for you, but you still get the great You know awesome use cases of video being used in, in, in a lot easier format using a dashboard, no software, you know, et cetera. And plus all of the metadata that we're really out, out for on this call, which is what, you know, what we can do with Mel.

[00:25:34] The second question for you that was asked, I believe it was in response to how to or does Meraki charge for our ecosystem partners to be into the app like our marketplace? I. I think so I think there's a, there's a credentialing component, right? They have to meet our standards, of course, and they have to meet the security as well as the solutions that they're actually claiming that they can do to be able to, to resell their their value on their marketplace.

[00:26:00] But I don't think we charge for anybody who be on there, do

[00:26:03] Raffi Vartian: we? Roughly? No. No. It's not like an Apple store thing where it's like 30% cut of, you know, all sales because we're really not transacting directly with the end customer either. To your point earlier, we're working with channel, and I've got a slide on this one about, you know, kind of who we're working with and kinda working through.

[00:26:18] So you still, these are still solutions. It's not an iPhone. You know, we use those kinds of examples as a description of how the architecture is done, not necessarily how the business model is done, if that makes it easier to.

[00:26:33] Jacque Brittain: Okay, great. I, I have one more question here. I might be stepping on part of your presentation here, Raffi, but you brought it up, please.

[00:26:40] Sure. The whole idea of the personal identifiable information. Yeah. How, how do we go about, I'm, I'm sure people are gonna have this question because it's, it's obviously something that a lot of companies are talking about. Absolutely. How do you segregate that personal, identifiable information and make sure that it's not used inappropriately?

[00:27:02] Raffi Vartian: It's a, it's a phenomenal question. So first thing to, to kind know about us, and I'll, I'll go into this from a high level and then we'll have specific slides to talk about it is that we're kind of GDPR compliant kind of outta the box, right? So Europe is driving a lot of security. On personal privacy regulation, right?

[00:27:18] Or all the rest of the world is kind of catching up a little bit, so like it or don't like it, right? Europe is driving a conversation. So we're GDPR compliant because we don't take personally identifying information, meaning that there are, you know, there's the, the facial recognition models that can look at, you know, the 62 points of the face and really kind of use video and abuse it, right?

[00:27:38] That's the idea of surveillance as a service. The easiest thing to identify from frame to frame, you could see that Matt and Raffi and Jacque, right here, we look very different, right? But if you, if you only are really interested about the data, about the metadata, about how people shop, about what people are interacting with, about what people might be doing to hide something or steal something, or something around those lines, you don't really care who the individual is, right?

[00:28:03] You're looking for behavioral information. You're looking for information about. The person, right. Age, gender, sentiment, those kinds of pieces of information because the point that we made earlier, right? You know, if you're marketing or operations or anything like that, you don't wanna touch that personally, identifying information.

[00:28:21] So we do it at the edge, right? And we kind of encrypt it. We're working with Meraki to take those MQTT feeds which is kind of the live stream of data, and be able to basically just take that data and be able to go down and screenshot things as well. But whenever we do screenshots or we do kind of any processing, we do automatic face blurring so that we're not actually looking at the individual.

[00:28:40] There are exceptions to that use case, which I'll talk about as well here in a couple slides. Which is personally iden, I'm sorry, person of interest detection. Meaning that these are folks that are already known to either law enforcement or, or loss prevention where we can. A few snapshots of an individual effectively train the system to only look for that person, encrypt that information at the edge.

[00:29:02] And then when individuals pass in front of the camera, we say, are you a match? Are you a match? Are you a match? Anyone that is a match, the 0.01% of the time that that person comes back, if you will, you can trigger an alert, right? And the 99.99% of the time that that person is not a match, you scrap the data immediately, you destroy it at the edge before it ever gets to the cloud.

[00:29:23] So there's, it's complex. But there's, there's, there's a, a lot of kind of tricks of the trade, if you will, in order to keep all that information secure, encrypted and ensure that it's not not abused. Does that make sense? Absolutely. Thank you. Okay, wonderful. Absolutely. So let me, lemme go through some of these slides now to kind of show you a little bit about who we are and kind of what we do and the reason we're kind of on the call here.

[00:29:46] So let me go ahead and see if my computer will actually work this time. Ah, there's the button. Sorry about that. Everyone can see that. You're good. Okay, great. All right, so you already know my name. You already know while we're here, we don't have to spend too much time on this. But a little bit about, sorry.

[00:30:02] Our company I described before we're about a hundred people strong headquartered out of Australia. I'm calling you from Chicago, from my house in Chicago where I've been for the last two, better part of two and a half years like everybody else it seems like. And we are two parts. Mel CX is melding con customer experience.

[00:30:17] That's what it stands for. So we've got a part of our company that focuses on device application and peripheral management for things like kiosks and digital signage. And then the other part of our company that we're gonna be talking about now, viana, which is all about computer vision. And those building blocks, those modules are things that we'll describe about today.

[00:30:34] We don't spend a lot of time doing dashboard talk. We talk a lot about use cases and what a business can do with the technology because we get everybody cut, starts glazing over when we start showing dashboards. So do a little bit of that, but mostly it's about use cases. So The, our, our head of technology says that we teach computers to see like humans.

[00:30:52] That's why we've got this kind of infographic. It spooked me a little bit when I first joined the company and we started to talk about those things. But really what this means is nuance, right? So like you talked about before, Jacque, do we really care about who that individual is? No, we don't. We care to teach.

[00:31:08] We we want to teach our system or build, do combine the building blocks to tell a story about what's going on. If you're interested in demographics and around how people are interacting with products, you know, that, that, that takes a, a couple of models putting it together and you've got a really teach system to think with a little bit of nuance analysis in real time and then make decisions, making decisions is really about triggering.

[00:31:32] Those event triggers can be something's happening. I'm gonna send an alert to loss prevention, or I want to be able to trigger content on a sign, or those different types of things. That's kind of what those three categories in those buckets are, right? So like we talked about before, you can see we even do it in our promotional materials that we're blurring out the face and we're really taking out individualized information, right?

[00:31:52] About the person, but not about who that person is. We don't frankly care most of the time. Right. The reason we can do this is because we use what's called synthetic data to train our models. So I think everyone's either played video games or has a kid that has video games right within their house.

[00:32:08] The same technology that's used to create video games and Hollywood movies, photoreal. Humans and, and objects and things like that. That's how we train our data and that's how we can do it extremely quickly and do it very, very cost effectively, right? So we're trying to get all those things kind of trained up and really go from the outcome of what the customer's looking for backwards to kind of training that model to create a representation of what that environment is, to be able to train those models pretty quickly.

[00:32:34] That's a lot of words, so if you feel free to interrupt me anytime that you guys wanna, right? It's a, a lot of data coming out of my mouth, right? So let's talk about, go to. We've talked about it before. We've got phenomenal partnerships with Matt and team over at the Meraki side. We also work in some very complex environments that needs additional computing, if you will.

[00:32:52] So we do that either through, you know, Cisco kind of uc, S'S platform with the Intel chip set. So there's a lot of computer vision solutions that are focused on the gpu. We're focused on the CPU, if you will. And then we utilize g ccp and also the Azure stack to be able to do a lot of the training and hosting.

[00:33:08] And we contract through kind of distribution, if you will, right? So we've got distribution relationships with Ingram set all over the world to be able for people to pull down some of the business. So it goes back to the question earlier, does Meraki charge a commission on the board on the ecosystem website?

[00:33:24] No. Right, because we still have to go through systems integrators, I s V, right, to go after kind of those enterprise and mid-market companies. So why are we so intertwined? I think is a, is a, is a question I get a lot. And I think philosophically we're really at the same point of where Meraki is cloud first, right?

[00:33:44] There's no application to download on our side either, right? So it's really about simplicity and out-of-box. It's all about the cloud and cloud manage. Anything from the, the UI itself to all the data that you can kind of collect and be able to analyze. We've even got something called manifest that we can kind of store individual actions.

[00:34:00] We can get into that more. If anybody was interested in learning to take a test drive of the product. At some point after this infinite scalability, right? As many cameras as you want, as many models as that you, that you want, that you can send down there. It's kind of a one click install from us that we can say, what are you interested in?

[00:34:15] Are you interested in planogram compliance model? Are you interested in demographic information models? You can kind of send those down. And then again, we're just kind of philosophically in the same place where we're not taking personally identifying information. We're not trying to do surveillance as a service, right?

[00:34:29] So let's talk about this. Jacque. We, we, we discussed it before. I don't wanna belabor the point, but this is kind of what we're looking for. So this ID is a fake id, right? We do this for marketing purposes, but if I were to walk into any kind of environment where we're running our analytics we can basically look at me Raffi as an individual and say, okay, let's blur the face.

[00:34:52] But we're gonna take a little bit of data from the face. We're gonna look at. Basic age information, gender information, and then sentiment as well. So sentiment is not always used, but that's kind of more of kind of like a marketing environment. But then we're looking at everything else. We're looking at the, the glasses, right?

[00:35:08] These are my vanity glasses that I use for webinars. We're, we're looking at what kind of shirt they're wearing. They're, are they carrying a purse? How they interacting with certain things in an environment. Even gate tracking how people walk in order to be able to identify individuals and how those individuals move in a certain environment, but not who that person is.

[00:35:28] Again, it's not necessarily important to us. And we don't have a basis of problematic model training, if you will. And we created all synthetically from scratch. So we don't have to worry about, we don't train our models to even look for things like race, right? That's, that's not kind of who we are.

[00:35:42] So we're taking that privacy, first idea, right? And the non-face activities around the individual to be able to. Anonymously track behaviors within an environment. Make sense so far? Okay. Now this is a stat that we had pulled and I think everyone on the call is familiar with the fact that there is loss in retail.

[00:36:05] It kind of belabor the point to really bring out stats, right? Because we all are here for a specific reason. So, you know, there's a lot of loss that's out there. What we have found through our initial engagements is this is an undercount, right? There is a lot of black holes within. Any kind of environment, anything from manufacturing to warehousing to back of the house, to actual, you know, store shelves.

[00:36:30] There's a lot of things that are either not admitted or that the data simply doesn't exist. So we go in and talk to customers and they say, yeah, we've got about a 2% loss. And then you start showing the actual data and it could be double or even triple that number. So the scale of the problem, I think is undercounted with any kind of surveys, because this is a self-reported survey.

[00:36:49] That's kind of the reason why I wanted to talk about it and bring it up. I don't know if anybody on the call has found that this, this is, is true, that there is a certain amount that you know for sure, but then a certain amount that you don't necessarily know for sure. We've got a customer all, it's not all straight posts, but we've got a different customer that doesn't I'll put it, I'll put it this way.

[00:37:07] We're examining the pick and pack part of their direct to consumer part of their brand, and they're, they only get complaints. and they know, know that they're lost when there's a complaint on an order, right? Cuz they're not tracking it in real time. They've subcontracted out to a third party. So they only know when people call and complain, well I dunno about you, but if you got extra, you've got an extra box of whatever product that you're ordering.

[00:37:30] You typically aren't calling it complaining. You're saying, well, I must have done something. Right? So there's a certain part of loss where you're kind of oversupplying customers that's not even being reported. So there's a lot in there that we can utilize with just unambiguous data right? About what we're doing.

[00:37:46] So lemme go through a couple of these. I'm, I'm starting to get a little long in the tooth when it comes to the things we're talking about. Let's bring it all together. Australia Post is one of our first customers and one of our most loyal and most interesting, if you will. So we're talking about insights from a retail environment.

[00:38:01] Also through Covid we started to create models that are looking at cleaning and keeping surfaces clean and how people are kind of interacting with those surfaces. I'll show you in a second about how we're applying that actually to kind of planogram compliance. But we've also created an entire solution that's just surrounded about self drop off for parcels and about how people are able to transact more quickly within a self-checkout.

[00:38:22] And it was great, great project. Of course, everything got stalled a little bit with Covid, but it's kind of ramping back up again. But what we learned from there is that people are self-checking and putting parcels and saying that I've got a, you know, a a thousand dollars iPhone that I'm putting in a package and sending to someone, buying the insurance, putting the package together, spending the $10 for the package, and then.

[00:38:43] Putting it in the box, but not putting it in the box, hiding it right, and kind of pulling it away, and then filing an insurance claim. That's a very specific use case, right? But we have the ability to kind of train the models to look for those types of things. That's what proof of lodgement is for us, right?

[00:38:59] So we say, okay, after the person is gone from this self-checkout, have they actually dropped it off and does the scan match? What what, what are they dropping off actually matches the scan? Because you've got a lot of complexity, right? People are very, very creative on how to steal from larger organizations.

[00:39:14] So we have to kind of develop that use case to be able to build a model to look for the certain behaviors that we're looking for. That's the, can we call proof of lodgement? This is kind of our self-checkout kiosk. I'll blow through this a little bit. This is just kind of a, a slide for us to show off a little bit about our ability to create environments like this.

[00:39:31] But let's kind of go into the, the basic components. So we talked about it before, how Mel CX is about the, the, the building blocks, the AI building blocks to kind of combine together to get that result. So basic level of people counting was something that we really hadn't considered about a year ago.

[00:39:46] We were looking for more kind of complex use cases, but what we found is that even people counting was pretty antiquated technology, but it was also the basis of how we can start capturing information again anonymously about individuals kind of as they walk in a store. It's kind of the point of capture about traffic flow and, and where people are and kind of how they're moving.

[00:40:06] And then also it's the beginning point of the behavior within an environment about what people are doing. To your point earlier, Jacque, right? People are disappearing into the quiet areas of the store and hiding things. Well, how long are they there? What are the behaviors? What aisles did they get to before they started to disappear into the quiet areas of the store?

[00:40:24] Things like that. If you don't have this baseline of data, it's hard to get that information. Zone engagement is just our way of describing how to re-identify individuals from camera to camera, again, anonymously but to figure out where they've. And how they've gone throughout a physical environment that could be used for marketing purposes, for you know, if you're a big box and you wanna be able to charge, you know different kind of electronics manufacturers prices and say that we've got, we are gonna guarantee that we've got, you know, young males you know 18 to 24 that are going in and playing with your you know, computers or, you know, PlayStation or whatever.

[00:41:01] We can extract that data out from it. But it's also the same cameras, the Matt point, Matt's point earlier are being used for security purposes as well. So the camera with the intelligence, right. Applications that sit on top of it, the model looking about where people are going, but using that for multiple stakeholders within a business for different purposes.

[00:41:20] But it's not just about individuals. It's also. You know, vehicles as well. So one thing use case that has been presented to us is that you might have a chain of of stores, right within a, a corridor like an I 95 on the East Coast. Well, you, you might say kind of where I grew up where you've got people that are hitting multiple stores, right?

[00:41:41] Because you might have a policy to non-intervention when somebody comes in and actually, you know, steals and kind of, you know, just, just runs out the door. Well, you might, the individuals might be hard to track, but you might know what a car is. So if you have a camera on the outside with the bullet cams, the 50 mb, 50 twos, right?

[00:41:56] And if you're looking for the specific car, that might give you a better lead time to prevent theft with your environment or let law enforcement know that somebody that has known to have stolen right is hitting all these different individual places. So the car might be the point where you can actually interrupt that theft.

[00:42:13] Go ahead.

[00:42:13] Matthew Moreno: Yeah. Yeah. The term casing, right? If they're casing ing multiple spots and looking for trends looking for spikes and things like that. Absolutely. Yeah. Those use cases when it comes to vehicle surveillance, right? I mean, the obvious has always been just making sure that the parking lots are relatively safe, but now you can actually use it and track data from a loss prevention perspective.

[00:42:34] That makes sense. But hey, but also those are customers, not then five nines, percent of them that are, that are coming under your lot are the good ones, and that's what you really want to know about, right? Yep.

[00:42:47] Raffi Vartian: But to your point, there's two sides of the coin. There's the loss prevention side of the coin, and there's the marketing and analytics side of the coin.

[00:42:53] Looks like we've got somebody in the chat, Jacque, I don't know if you wanted to take a look at that.

[00:42:56] Jacque Brittain: We've got do cameras come preloaded with specific analytics running in them? How many analytics can run on each camera?

[00:43:04] Matthew Moreno: Sure. I can take that while we're, please. Yeah. So yes, the cameras do come with some basic analytics out of the box.

[00:43:11] We do people counting and vehicle counting as well as heat mapping. And then of course, with those two types of analytics, we can also use those to filter on your searching for the traditional sense of doing your investigations with your cameras. With that stated, those are the only ones that come with the camera after that.

[00:43:29] When you're adding, you know, like a license plate recognition or some of the other ones that are melded, I described those that come from the the models that, that he's describing from the, the meld, right. The meld piece. But I also, just another time for me to emphasize this. What's really cool about the architecture of edge based cloud manage is that you can add an analytic.

[00:43:51] Specific to the camera. So instead of doing what I, I call that peanut butter spread of what a server has to do across all of your cameras, right? We can do it on an individual camera, so the camera that's outside, put an l p r, you know, model from meld on that one, the camera that's in the warehouse. Put a a space utilization model on that camera for the camera that's at the front door, do the people counting and gender and customer sentiment analytic model on that one.

[00:44:20] So now the choice is yours, right? So, no, we don't have any analytics other than the basic ones in the camera. That's an advantage for you to be able to add what's more important to your environment on that particular camera. That that's key.

[00:44:37] Raffi Vartian: Yeah, absolutely. And and to your point, we've got customers that are we have now the ability, Meraki has enabled the ability for us to optimize and, and distribute those models down to the firmware level on the camera.

[00:44:47] Exactly to Matt's point, right? So you take the base models with the basic kind of collection of data, right? And then you can kind of retrain and then send those models down to the cameras themselves, which is, I believe, kind of really kind of burst in the industry. And we're excited to be one of the initial launch partners for this.

[00:45:01] Yeah. Get about 13 minutes left on the call. I'm gonna go through a couple of these quickly, Jacque, if that's okay. Make sure that I get through the slides and then we can kind of, you know, come back to it. I don't wanna be the micro machines guy and go too fast, but wanted to kind of describe some of these things.

[00:45:13] So this is a slide talking about the ability to filter out staffing and, and, and customers, right? So we have the ability to train for things like uniforms or name tags or all the rest of that to figure out how you're gonna separate out from an analytics perspective. Again, the model that's running on the camera, the, the people that work for you and the, and your customers, right?

[00:45:34] It's a big separation point that you'd have to dig through manually be able to look at those things to be able to generate data on. Okay, so this is really bleeds into where do you want those people to go? If they're not an employee, are they going in the spaces that they're not supposed? Right, it's restricted zones within any kind of particular environment.

[00:45:53] This is kind of based on kind of metadata and not the personal of interest information I talked about before. This is very complex, so I, I can go into it. I could do this for an hour, but we don't have an hour, so I'm just gonna put it this way. What is the biggest thing that we can see from a human interaction perspective about when they're gonna go look and steal things from a shelf?

[00:46:16] It's not necessarily, you know, who the person is or what they're wearing or any of those kinds of things. It's this, right? It's looking side to side and trying to figure out if there are people on the sides of the aisle, that kind of human behavior. And then, you know, pulling something into, disappearing from the shelf, if you will.

[00:46:34] You know, hand tracking and things like. It's the combination of those models that gives us the ability to really dig in and look at theft detection, kind of blunt theft detection, if you will, from the shelf. It's not always the most important thing in a retailer's environment. It may be just the component, but it's something that we can train for for a specific retailer's environment.

[00:46:54] Just kinda like this, right? You know, stealing from the actual register itself. Like blunt theft is a portion of what theft really kind of looked like. But what about the kind of not, you know, customer not present interaction. Where're they, you've got a big box retailer or something like that where there's a ton of returns and if they're all stacked up in one corner and the person starts to kind of double return if you will, right?

[00:47:16] And then start peeling off gift cards, you've got a little bit of sweetheart action with somebody in front of you, right? Where a friend comes in and just has a fake return and all of a sudden they're walking out with a hundred dollars gift cards. That's a huge problem, right? But it's something that we can have the ability to train for.

[00:47:29] And it's something that we really have to have a conversation with the end customer about what that's all about. Person of interest. We already covered the ability to effectively create a black box environment of known individuals and creating millions of different examples of that person to be able to train the model successfully in order to identify those folks.

[00:47:46] And, and this is one that I wanted to spend some time on planogram compliance, right? So how many things are on the shelf at any given time is the baseline of measurement about what has gone missing or what needs to be in stock or outta stock. Again, going back to Matt's point earlier of multiple stakeholders within an organization, being able to utilize the same camera, the same data, but slice it up in different ways, right?

[00:48:11] In stock or outta stock, is a huge problem with grocery stores, right? You've got your instacarts of the world that are coming in and DoorDash that could, might, might wipe out all your mustard, right? I don't know, whatever the, the thing is that people are the most interested in these days, right? . So getting that that, that's an atypical spending spree if you will, from what retailers have been doing for the last a hundred years and what they know from individual customers' behavior.

[00:48:35] They wanna drive people in, they want to get them to walk around the store, get that delight right, make sure that they're not stealing and then, you know, stock it a give kind of given level. Now all that stuff is very, very much changed where you might, I can just sit here and go on my app and click a button.

[00:48:49] Somebody goes to my local you know, jewel Osco that we have here in the Chicago area, and just take as much as you can, right? So having that realtime understanding of the planogram is something that we're seeing more and more and more. It serves store operations, it serves marketing, but it also serves loss prevention as.

[00:49:06] Right. And then we can go back to the house again, we don't have time to go through a lot of these use cases, but security monitoring of people within warehouses and tracking packages and things like that. You don't typically think about it as loss prevention, but there's a lot of loss that goes on in the warehouse, right?

[00:49:21] And there's a lot of things that get broken and people get hurt and things a lot along those lines. Slip and fall detection we talked about before. Again, this is in the warehouse environment, but can also be applied within retail. You use those cameras to have professionals look to see if somebody's kind of faked it, right?

[00:49:36] But fundamentally you wanna know immediately if somebody's gone from a standing position to a prone position on the floor, right? And we have the ability to kind of train for those things as well. We're also looking at the Meraki ability to do audio analysis as well. Again, anonymized, not personally identifying audio information, so spikes, right?

[00:49:55] Or keywords that are kind of coming out as well. , we, this is, this is another solution that we have. Hold on one second.

[00:50:00] Matthew Moreno: Yeah, go ahead. So there was a question from please from one of our pan one of our attendees again, terrific questions. It's really about this one in the case of pricing, and obviously, you know, pricing is the easiest part, right?

[00:50:11] As we all say. But the, the question is, are all of these analytics priced differently? Is the SaaS model or is it a one-time charge? So ums kinda two questions in one there about the model, so I'll let you explain that one.

[00:50:24] Raffi Vartian: It's a great one. So there's two components of what we do. So any of the custom model training is kind of statement of work driven, if you will, if you're looking forward to train for uniforms or things like that.

[00:50:33] We go between average, between about 60 to 90 days about how long it takes to kind of scope and train those things. Those are one time fees. Those models can be pervasive in your environment. You can use 'em wherever you want effectively, right? We try to keep those costs relatively low in order to be able to scale the SaaS business.

[00:50:51] That's what we're all about. So we're a a software as a service kind of per camera, if you will. And there's different tiers of the kinds of models that you can deploy down as you get more complex. It's, it costs a little bit more, right? We don't publish pricing up on our website. We're very much tra channel and partner driven.

[00:51:08] And we can explore that with anyone who'd like to discuss it. .

[00:51:12] Matthew Moreno: But I will, I always say this, when people ask about pricing, and it's the fair question, of course, right? It's obviously the most honest that we can give. Yeah. But if, if the, if if everything works in the reality that we're trying to, you know, to make inside of a store, right?

[00:51:26] In a store, in a warehouse and public space, in the smart space, the cost of the model should never be over. It, it, it should make the TCO right. Less, right? So, yep. That the model that we're adding to that camera, whether it's a slip and fall, right? It's something new to your environment, maybe, I don't know, cost of, you know, a hundred dollars, right?

[00:51:47] For that model or whatever it is per camera. But if the return is quicker over two years with the lifespan of the camera, Theoretically the cost matter is none, right? I mean, we're not making it free by any stress of the imagination, but the business case is made for you for these models. That's the whole purpose.

[00:52:05] In the past, loss prevention has always been a loss leader, right? It's always been a a cost to doing business. What we're saying now is, yes, we agree that you can still need cameras for camera's sake, right? For those older, there's millions of reasons why you've had them in the past. But if we can extract more value by putting these awesome models on the camera, yes, it adds a little bit more cost to the camera.

[00:52:26] Maybe you're splitting that cost within different lines of business in the organization, but the return is now coming tenfold than you were ever getting in the past, thus making it really affordable to the organization in general. So,

[00:52:39] Raffi Vartian: it's a great point, Matt. It's a great point. And, and what we're seeing and I'll, I'll wrap up with this, right, is what we're seeing is that the biggest cost center that everybody has is labor.

[00:52:48] right? Fundamentally, it's about who is staffing, right? The different, who is looking at the camera in real time, you know, is that person you know, so caffeinated that they could do it perfect a hundred percent of the time for 10 hours a day, five days a week. The answer is no, right? I mean, the, the, the technology is there for a reason.

[00:53:05] We can train it to a certain percentage for a, you know, for a reason. And it is an assist to the individuals that are working with that in an environment without dramatically expanding your, your labor equation, right? So we're trying to optimize the things that are already in store and it shouldn't be seen as a, as a threat at all, right?

[00:53:23] It should be seen as a value add. So it takes a little bit more work to develop that roi, right? Because you have to sit down with different business units, not just security and loss prevention. You've gotta sit down and say, okay, well how much, how many times does a person go out and do stock checking? I think you everybody, I dunno if you remember the, the robots that used to roam around Walmart.

[00:53:42] and they were doing real time, you know, scanning every single shelf. You know, Boston Nova Robots is what they were called. Sorry, it took me a second to be able to draw that out. They went from, you know, one store to a hundred stores and then they killed the entire thing. Cause the ROI didn't make sense.

[00:53:55] Well those are robots that are going up and down the, the line does make sense that it, it didn't do do

[00:54:01] Matthew Moreno: t c o cause it's, it's linear. It's doing one thing. Exactly. That's key. Right. And if I can, I mean, I'll end it there on my own purpose. Sure. Never let the cameras just do one thing. Get as w get as much value as you can out of them instead of the old, you know, sticking it on a wall and being done with it.

[00:54:18] Right. Re mentality. Great cameras of course. But really try to exploit the data as much as you.

[00:54:25] Jacque Brittain: So if you looked at this thing from a holistic standpoint, you're gonna have leaders here that are gonna wanna, they're gonna have to present a business plan on why they should be able to use this solution.

[00:54:41] Yep. If you had to come up with a few bullet points, that should be part of that. Just real quick. Sure. What do you think they would be?

[00:54:49] Matthew Moreno: I mean, obviously life. Okay. I, I, I can speak on behalf of the hardware piece and then I'll let Raffi talk on the, on the software piece. But on the hardware piece, you, you have to look at, you know, the lifecycle management of your existing physical security system.

[00:55:04] Right. And what are the pieces that we can. Better, right? Like as a measurement format. Is it quick to share video, you know, from an efficiency perspective, is it better, more quality video from, you know, a solving of a, of a case that's easier for you, or reducing risk in general? Right? And of course, you know, the overall component is I, I need new cameras where I'm expanding my operation and things like that.

[00:55:30] So we can make the the business case for that really, really simple, right? The idea of actually buying new cameras. Now, adding the use cases is what we need back from the customers, right? We can identify a hundred use cases, but that's way too many. And it would probably cause, you know, a nightmare of, of, of an idea to go out and d you know, to deploy these.

[00:55:48] But identifying like one or two use cases. Then I, we insert Mel and Raffi in there, and then what?

[00:55:54] Raffi Vartian: So we're, we're kind of the, imagine if you could part of the equation, right? Because we're, you know, it's a non-linear a return on investment. So if you're already making the investment through the Meraki technology and we're in application that sits on top of it, we've got ROIs that are a hundred to one, right?

[00:56:11] Where we're into automation and certain environments, and we're looking at, okay, do we have to staff up people and labor's always gonna get more expensive, right? Or can we actually utilize software and technology? It's not gonna be a hundred to one ROI in every single environment, but even if it was a conversation of every dollar he gave me, I gave you 10 back, how many dollars could you find?

[00:56:31] That's overly simplistic. Of course, Jacque, but I'm, you know, I can do this for half an hour.

[00:56:36] Matthew Moreno: We're, we're the ones that are gonna help you with that. Like I said, the, the four pieces, right. Obviously the customers first. Second is the hardware piece, which is the infrastructure and parts and pieces of the network.

[00:56:47] That's where I come in. Right. And then of course, the use cases and the solutions, the business model around the Raffi is, and then of course the channel part people are actually installed. They have to be, they're, they're part of this is key. But that's the fun, right? That's where we live for, that's where we own it.

[00:57:01] That's where we take the, the, the, the path of, of the customer journey to get to something that's, you know, a identifiable business case that we've helped make For sure. You raise our hand for that one, . Okay.

[00:57:13] Jacque Brittain: Looks like we have one last question here. John, thank you for this question. It's one that I can answer.

[00:57:18] He's, he unfortunately was only able to catch part of our presentation today. The presentation itself will be available on demand. Through the LP Media Group website at www.losspreventionmedia.com. You'll be able to access that both through the invitation, the U R L O that you got when you signed up, or you can go directly to the website and be able to get that information.

[00:57:44] It'll be available for at least 90 days. And we definitely encourage you to take another look at it. Gentlemen, I certainly appreciate your time today. I think it was a great conversation. Thank you for your time today. I wanna thank everybody out there for attending today's webinar. And I would also like to thank our sponsor Cisco Meraki.

[00:58:03] It was a great presentation today. Thank everybody for their time today. Stay safe and have a great day. Thank you.

[00:58:10] Raffi Vartian: Thanks, Jacque. Thanks Matt. Bye.

[00:58:12] Thanks, Jacque. Thanks for hosting.

Latest from meldCX

meldCX and Signagelive team up to offer AI data-driven digital signage

by
meldCX
Jan 24, 2024
3 minutes
Leveraging AI to deliver targeted digital signage content and in-depth anonymous viewer analytics

Get the latest meldCX news and insights right to your inbox!