From RFID to Physical AI: Giving Everyday Assets a Digital Identity

[00:00:00] As we worked with a lot of the industry analysts, how they were positioning the magic quadrants and where they were going, they calibrated us a little bit and said, well, look, ambient IoT is nice from a Williott standpoint of maybe how the tags are being harvested. But in reality, the definition of what you're doing today is physical AI. Hello and welcome to the Next Level Supply Chain with GS1 US, a podcast in which we explore the mind bending world of global supply chains, covering topics such as automation, innovation,

[00:00:30] unique identity and more. I'm your co-host Reed. And I'm Liz. And welcome to the show. Hello, everyone. I hope that you're having a great day. We really appreciate you joining us. We have a great discussion lined up about RFID, AI and BLE. That is Bluetooth Low Energy. Our guest today is Amir Kashniyadi. He's the vice president of Williott and they are a solution provider in this space. And we're going to be talking about

[00:01:00] about physical AI and the impacts it's about physical AI and the impacts it's having on our supply chains. Unfortunately for today, Liz won't be with us. She's traveling and got caught up in the snafu of flights and so forth and was unable to make the schedule. But it's still a great conversation and we'll have Liz the next time. Hey, Amir, welcome to the show. Thank you for having me, Reed. Happy to be here.

[00:01:25] Yeah, we're happy to have you. We're delighted to make this work. And so before we jump into our conversation, if you wouldn't mind, just for the audience, just give us a little background on who you are, what your role at Williott is, and just kind of set the stage for our guests.

[00:01:44] Sure. So my name is Amir Kashniyadi. I'm the vice president at Williott. I handle a handful of different responsibilities, everything from the marketing to the channel business development and also cross-functional into the strategy of the business. Spent the better half of my career really on the RFID side. And then prior to that, a handful of activities with software within the Microsoft channel, power system analysis software within oil and gas.

[00:02:10] So I've kind of evolved over the years into the RFID arena. And I had history with this business about seven years ago, right around the pre-Gen1 times when SmartTrack, our RFID business was acquired by Avery Dennison.

[00:02:23] One of my pet projects within the umbrella was actually Williott as a startup and early investment. So I've had the privilege of knowing this business pre-Gen1 and then during my time at Identiv, actually being part of the forefront of producing the tags up to about 55 million units of the pixels. So being able to actually have hands-on experience on the production side. And then the best of all worlds came together when I actually joined the business officially and the curtain was open.

[00:02:52] So I've seen the company and the product from infancy all the way to where it is. And it's a privileged position now to be working with big companies such as Walmart and many others in the industry and being able to be a part of this journey from the ground up. I've watched this from afar and I'm excited for us to kind of continue this conversation here talking a little bit more about physical AI, which I'm going to get to in a second.

[00:03:14] But before I get to that question for our audience, how would you describe Williott? I mean, what is the tagline? What is the company provide? From a tagline perspective, very simple. We're physical AI for everyday assets. Our vision really is to turn everyday items into digital assets that can live and breathe and you can track them throughout the supply chain with no brownouts, with full visibility. As a company, we are a physical AI platform. We are a platform company.

[00:03:42] And the way that the technology comes together is we've enabled a complete ecosystem that starts with everything from our IP around the physical tags. We enable all the manufacturers in the industry that produce RFID tags with the same equipment to have an ability to produce these tags. So they have a commercial model to go to market. These tags are enabled by network devices, everyday type devices that are in the market today.

[00:04:08] They energize and harvest energy in a battery free format and they transmit signals to these bridges via 2.4 gigahertz, which is a Bluetooth low energy protocol. The information that's transferred basically through these bridges is everything from the item location, motion, temperature, humidity, even exposure to lights. So if it's a tamper situation, a package is open, we have insights into that. So the Williott platform as an umbrella sits over all of these assets.

[00:04:37] It's able to decrypt this information, become a source of truth. And then it is a one-stop shop to relay any kind of reporting via the AI features. Or that data could be relayed to any ERP WMS out there. And you can utilize investments you've made into third-party softwares. And we can just relay that clean source data from the tags. That was a lot to take.

[00:05:03] So let me start with where I was going in the beginning is we're hearing this term physical AI more and more and more. What does that really mean? And how is it having an impact on supply chains? Sure. So I believe that even taking a step back from physical AI, we've went through quite a bit of a journey over the years.

[00:05:25] And there's many milestones within that journey, positive or negative, such as COVID, that have surfaced up and things have changed in the ecosystem. So stepping into it, there was a term around Internet of Things. Right. And it was very prominent. And the reality around IoT was that you have all these living, breathing assets in the market. The reality was that wasn't the case at the time. It was more around the devices because everything was going mobile first, cloud first. Many of us have multiple cell phones.

[00:05:56] We have tablets. We have laptops. These devices were becoming prominent for applications that were utilized in the cloud. But the theory of actually being able to tag assets and give them digital identities was still in infancy. It wasn't new. So it was a concept that was being born, but the devices were the enablers. Then we fast forward. And what I really call from Internet of Devices, we evolved truly to what Internet of Things was. And a lot of that actually was a byproduct of COVID.

[00:06:25] So we saw that everything was going contactless from credit cards, the payments went through your phone. QR codes were prominent in all the menus. So the devices now were becoming the byproduct of the readers that interacted with these everyday items that now started to become digitized. So Internet of Things started to pick up.

[00:06:48] And then there was a fuzzy period where between Internet of Things and where we are with physical AI, there was a coined term called ambient IoT that was mixed in. And ambient IoT was really a stamp that Williott was pushing forward because they were ambient waves that were energizing the tap.

[00:07:06] But as we worked with a lot of the industry analysts, how they were positioning the magic quadrants and where they were going, they calibrated us a little bit and said, well, look, ambient IoT is nice from a Williott standpoint of maybe how the tags are being harvested. But in reality, the definition of what you're doing today is physical AI.

[00:07:25] And although historically it's been a robotics term, the reality is all these years of devices into things have now segmented into what we are layering in as the physical world going digital at the asset level. And AI is now prominent.

[00:08:09] Specific asset, that digital identity that's been created. And we can make sense of it with our platform layer and the AI that we're putting forward. So our customers can make better predictive assumptions on what their assets are doing through the supply chain. I really appreciate you kind of backing us up and taking that through because I've heard IoT for well over a decade now. I mean, way going back to the John Chambers days of Cisco systems when edge computing and all this other stuff.

[00:08:39] But we were waiting for 5G and other things to happen. But you discussed the Bluetooth low energy, the BLE kind of space and that ambient IoT. I had an old boss that, well, personal wasn't old. It was just a boss in my previous life that called it the fidgetal. Yes. Digital. The fidgetal. So let's talk a little bit about this.

[00:09:04] Physical AI, it really depends on understanding what's happening in the real world, right? Like it's out there. It's out. Let's just take a fictitious scenario. But it's on a truck. It's in a warehouse. It's on a shelf. It's lost. We don't know where it is. But when we find it, it's kind of like a black box. I want to know what happened to this when it was lost in this period.

[00:09:32] So it's providing you all are providing this visibility, as you said earlier, into like, I'd like to understand this a little bit more temperature, light, you know, what can we get from your tags that we might not get from others that are complementary to the visibility and traceability of logistics and filling some data blind spots, if you will? Absolutely.

[00:10:02] I think it's one understanding from a capability perspective where the limitations of other technology historically has been. And there's a way that everything can coexist. It's not a matter of always being a binary, either they select Williith or they go with a different technology. It's really around the use case and the ROI that we're creating for the customer. And I think that's fundamental to every prospective discussion we have, any kind of customer that we have today that has invested in legacy technologies.

[00:10:31] It's not about us stepping into it and saying, you're going to rip and replace everything overnight. It's more around understanding what those pain points are and how Williith fits into the equation. We've taken a lot of time to go through analysis around where the limitations are first within supply chains. And we've seen that there's been strong limitations when you look at traceability of an asset in various choke points within the supply chain, what we call brownout spots.

[00:11:00] And so what I mean by that is that you might be able to track an asset within four walls or within leaving a facility through a dock door. But from that time where it leaves the manufacturing facility, as an example, and it gets loaded on a trailer. And then during the transit time from point A from the manufacturing through the trailer process and the transportation towards a distribution center.

[00:11:26] Historically, that has been a challenge to be able to, at all times, at the item level, understand where your assets are, what condition they are in. And when there's a handshake between the manufacturer and distributor, whether it's under the same umbrella or 3PL or something in between, can you validate that those assets went from point A to point B in the time that they were supposed to? And were they delivered in the condition that they were supposed to be?

[00:11:51] And if for some reason that broke down any part through that process, can you pinpoint exactly where it broke down? So we understand why and we don't repeat the same mistake. And then you can follow that same methodology from a distribution center to a back of store. When that item is delivered to the back of the store, was it dwelling too long on a dock door? Was it a refrigeration freezer issue that it was loaded in, taken out, and then it went through temperature volatility?

[00:12:19] So we kind of unpacked this and we said, okay, the first starting point is in transit, we've seen issues. We've seen that the handshake when something is delivered, sometimes it's not clear when things need to, especially on the perishable side, be refrigerated, be frozen in a timely manner. And the solution that we're putting forward covers not only, it's really five main use cases. It's when the asset leaves, when the asset is received, the condition of that asset, exactly

[00:12:49] where that asset is, cycle count within the facility. And if it's being transported by any kind of reusable container, RPC, rolling cage, where is that physical container at all times? And is the assets part of it or is it not? And as we've unpacked these five use cases, we've started to make sense and start to land and expand with customers. But this starting point is really around these five.

[00:13:12] And when you look at legacy technologies, we've seen that the capital costs to go in and install heavy readers, doctor readers at a facility with 300 different doctors or within 300 doctors with 300 trailers that pull in at different times, the cost becomes very capital intensive. And then at a certain point, scalability gets hindered.

[00:13:36] So for us with very low cost bridges, gateways, we can get in there, many cases right around 50 to $150, install the hardware and then get full visibility through that process in a way that legacy technologies, machine vision with QR codes, RFID with very expensive gates or manual processes that are error prone, we can automate that. And so that's been our starting point to look at really those five select use cases and then

[00:14:06] start to dig in based on the requirements of the customers. That really paints a clear picture for me. I was going to be asking about use cases. I know that you guys have done some work with Auburn University and you just laid out these five. Are they cross industry or are we seeing more in one industry than another? Cause two come to mind right away for me, which is certain pharmaceuticals and food.

[00:14:32] So whether it's food service or grocery, but are we missing out on other things that you're doing within the industries around apparel, general merchandise, or even I'm thinking. DoD inventories, you know, department of defense. Great question. And right now we are cross functional across these industries. So I would say it's agnostic. It's for us, it's more around the prioritization exercise.

[00:14:58] So some of the DoDs on the compliance side, it's a little bit more difficult. These are longer sales cycles. We can address the pain points, but we know that it is working with the government. It does take time to get involved through that process. Even in healthcare with pharmaceuticals, these are very long sales cycles as well. But the three PLs that work with them, we can get started today and start to build, whether it's midstream, upstream value, or downstream value, a certain node within the supply chain

[00:15:26] that then we can start to carry through. But the prominent ones that we see today as fast moving and really repeatable is under the retail umbrella, food and grocery. You can also kind of have a cross segment there with quick service restaurants because they all work with food processors and they need to understand the source of the different types of food that come into their facilities. We have automotive. This is a lot and heavy on the RPC side and very high value cost of items.

[00:15:54] And you also need to validate exactly where they are. We have CPG, which is a major one, especially with our partnership with GS1. And then we have basically a general category under logistics as well. So we've announced Royal Mail. We have also a couple other logistics providers that we're starting early stage deployments with. And we see major value on the logistics side because that has a major also mini supply chain within itself because these logistics providers also coexist.

[00:16:23] So if you look at the USPSs of the world and how they work with Amazon, we see future potentials around that. Royal Mail, for example, the UK has a very broad ecosystem of partners that they work with in their logistics umbrella. So that also segment, we see major potential and we have a very good use case both at the roll cage RPC level, but also at the disposable level. When you look at cardboard containers, gaylords, we can tag those and we can give full visibility

[00:16:51] to every asset that's sitting within those. And also call out if an asset is moving to the wrong trailer. Is it loaded wrong? Can you offload that? Can you save all the costs that goes into sending that to the wrong location and then having to revert it and then all the downstream issues that come with it with customer complaints? Wow. Okay. I'm starting to, the pictures are starting to come together. It kind of hit on some things for me. There is reusable, but there's also disposable.

[00:17:17] And again, everything has its ROI justifications and reasons why people are leveraging one versus another. I'd like to bring us back to some of the stuff you mentioned in the very beginning. You're leveraging Bluetooth to connect physical products and assets to your digital system. Talk to me a little bit about the complimentary integrations with other existing systems, whether

[00:17:45] it be QR codes on your tags, other RFID systems that are already in place. One thing that's jumping off for me is, are your RFID tags compatible with other RFID readers and Bluetooth hubs? Can I use both? Can it interject into both systems? Those are just some things top of mind. Maybe you have some other examples. Sure, sure. And that's also a very great, great point here. When we started, the foundation was really, and I'll show it again, our own pixels.

[00:18:13] So our own tags being the means for all these different capabilities. That vision and that journey hasn't changed. So we still are using our pixels. The RFID manufacturers that are qualified with Williard are producing Williard tags, the Avery Denizens of the world, the Tagioses of the world. They produce these at scale for customers and they work directly with the customers to supply the pixels themselves. But we do see a future because the pixels are all encompassing.

[00:18:40] They have the asset tracking capability and the condition monitoring. That there's certain use cases that pixels are not ideal. So when I look at a conveyor belt, and this is again, a delineation of the different technologies to use at the right time, is that if we're approached by a customer and that customer is specifically looking at fast moving good, a CPG on a conveyor belt, let's say 200 items per minute. The speed is very important. The millisecond reads through that terminal are one of the foundational requirements.

[00:19:10] We may pause them and say, okay, if it's one choke point that it's going through, it doesn't really require manual intervention. And it is super fast. You're talking millisecond reads. Bluetooth low energy may not be the right solution. Their RFID might be the right solution. Now you can argue maybe you could put two tags because it's a perishable and you need to get visibility from a passive and low cost solution through the supply chain.

[00:19:35] But that specific speed on the conveyor belt and that portion of the supply chain is a problem and RFID can solve it. We can coexist in that type of arena. Or if it's just one specific use case, that's one that we might defer and say, okay, RFID is a better use case. But from a platform perspective, we've seen value that our roadmap is now going towards being able to bring in other data sources outside of just Willian.

[00:20:03] And so we're now starting to march with certain customers in a format that they have access to Willian's data from our pixels, but also they have access if they'd like. And that tag has, let's say, a barcode that information could be ingested. Or if it's something that has to do with RFID and they're using it again with that conveyor belt example, we can take the SKU information tied to the TID or UID of the RFID tag associated

[00:20:29] and then bring it into the platform so you have a one-stop shop for all your data. So from a tag perspective, technology perspective, the platform has now became agnostic that we can bring in data and give our customers more options. And then when you look at an infrastructure, because the tag is one part of the equation, it's the infrastructure also that needs to be coupled with it. We have a very robust ecosystem of bridges. Bridges, what I mean by bridges is any kind of beacon reader that's out there that's a

[00:20:59] prominent name that can already be cross-qualified to read Willian tags, even if they're doing RFID from an existing standpoint. So many of the prominent names there, I'll save some names so I don't call out some and forget others. But all the major players out there have had experiments with us or are in live deployments with us and they have experience with legacy technologies as well with the RFID or with the machine vision.

[00:21:26] And we have found a way to get certified both from their expectations and also from Willian's expectations so that we can coexist in those environments. And that piece of the puzzle has been so important is when we walk into a new engagement and somebody is interested around the technology for pain points, one of the first questions we will ask them is, do you have any legacy investments that you've made? And when we learn they've done RFID, we might learn there's a zebra reader in there.

[00:21:55] We might learn there's an energist, rigato, whatever it might be. And we have now an inroad to say, we've worked with them before and here's what we can do. So it's more around getting more runway around the investments they've made versus having them rip and replace and start fresh with one or two vendors. That's really encouraging to hear coming from the IT world and thinking of this physical to digital.

[00:22:18] We see a lot of solutions out there where we're looking for interoperability based off of standards. And GS1 US and GS1 in general globally is a standards body for item identification, right? We created a syntax so that people know, hey, when you receive this, whether it's in the R code or the RFID tag, or if it's even in, you know, a blockchain environment, you know, this syntax

[00:22:46] is a GS1 global item identifier, right? It's a GTIN. And it's understood through a set of standards. And it seems like what you're driving at is, let's leverage investments that were made so that we can be interoperable. We can have our own secret sauce in certain areas, but we want to be complementary to the system. Am I hearing that correctly?

[00:23:14] Because that's what it sounds like. Absolutely. Absolutely. Anytime I think a new technology is layered in, you're going through an era of digital transformation and that you can argue the tech stack is continuously evolving at rapid paces. You don't want to stop the ecosystem and the market from the traction that it's had historically. You want to coexist. And that's a discussion we have with partners every day is that let's look at the ROI that we're driving with customers.

[00:23:41] Let's look at the use cases and let's make sure that collectively we approach customers with the best solution at all times. And if there's areas where they can get more value by consolidation, we should look at that. And this is a perfect example when you look at like a barcode or a QR code or RFID or Williit or even an active logger, what are the right use cases behind those? And if there's already a deployment behind it, but they still have the pain point, the ROI is very, very clear.

[00:24:11] How do we do it in a way that it doesn't become a bottleneck in your supply chain and your cost of ownership is low and you're seeing that ROI? Yeah, I always referred to it as a competition. We're competing and cooperating at the same time to betterment. I loved your term of the value stack and the digital stack. Like we keep layering on that makes a lot of sense to me. And it's tough for companies. Some of these companies are big conglomerates and multi-billion dollar global folks, but some

[00:24:40] of them are small mom and pop 3PLs that need to enable their distribution warehouses to incorporate in to support these larger conglomerates. And they just don't have the wherewithal to upgrade technical investments every year, let alone every three or every five. Exactly. Exactly. Yeah. I mean, it gets really, really challenging. That's super encouraging. Let's look forward a little bit.

[00:25:08] Looking ahead, where do you see supply chains in the next five years? Are we going to have a massive evolution, revolution, or are we going to just keep building and have some compound wins here and there? But subtle, do you see a big change coming? Well, I think it's my subjective opinion on kind of where it's going. I think it's going to come much faster.

[00:25:35] When you start to drive value, especially at the enterprise levels that we work with, you get a lot of individuals that were waiting. They were looking over their left or right shoulder. They see big companies moving forward. They probably see a lot of them as partners as well. So it might be upstream, downstream partners that are gaining advantages and they want to be part of it. So they're going to start to shoulder in. So I really think we're in a transformational time.

[00:26:00] If I make a forward-looking statement between 18 to 24 to 36 months, we are going to be accelerating with many more deployments that are going to coexist within the value chain and how they work together. I made the logistics example earlier. I would make another prediction really around food processors and how they work with the QSRs because this was an area within RFID that we were always challenged with was how do we build the value, not only for the visibility of these perishables through the supply chain,

[00:26:29] but also where do we build value that the condition of these assets are being tracked in the right format and without jeopardizing the tag. So the tag would always need to have a change in the IC. The readers would need a different type of calibration. You're driving higher costs, both on the capital side with the infrastructure, but also the item level costs with the tag. And it wasn't a one size fits all with all these use cases.

[00:26:56] And so when you look at the capability metrics, I think Williott is very well positioned. That acceleration year and a half to three years out is going to be matching between a lot of the individuals that have had these limitations with RFID. And then if I look beyond that, I would say that the technology is going to go through evolutions. So we're not only improving, we're in Gen 3 right now, we'll be on track with Gen 4 and Gen 5. So you're looking at reductions of cost of the tag. We're going at roughly about 10 cents at scale to about 5 cents.

[00:27:26] And you can make the assumption that we would probably go another 50% less on that. So the cost of having a tag with all these capabilities being in the same arena as RFID, you have a no brainer now to go with this technology over RFID. So you're getting more deployments out there. You're getting more tags. There's the economy of scale. And then I think the capabilities of the tags will evolve. So we will see a future. It's kind of scary, but that every asset could potentially be talking to each other.

[00:27:54] And then when it is summoned up back to one node, you can get visibility of everything within the ecosystem or swarm of that asset. So if you're in a retail environment and you'd like to buy a garment, perhaps that garment already from a digital identity is linked to a pair of jeans. So when you lift that garment and you get visibility that it went to a dressing room, you will get a notification to go pick up those jeans and get a discount because these items are talking to each other.

[00:28:22] So it's going to evolve, in my opinion, much further than a supply chain and become a really a customer lifetime value discussion. And more marketing teams are going to be involved. They're going to try to get into more promotions. And we love marketing teams because typically they have the biggest budgets. So they want to explore. They want to get more visibility on customers. So that could be definitely a forward-looking potential that this could evolve to.

[00:28:49] That is really thought-provoking. When you were talking, I was thinking immediately of packages talking to each other. Like, what if I received a pallet of bananas or tomatoes, a perishable fruit, food? And this pallet has, I don't know, let's just, for argument's sake, it has, you know, 25 boxes of whatever it's on it.

[00:29:16] Could I plausibly pick up a box and it says, I'm one of 25? And the next box says, I'm 10 of 25. Meaning, one should be sold first. 10 should be sold tenth. Like, so that I know, hey, it's not first in, first out. Because first in, first out works in some cases. But in other cases, hey, I might be more ripe. You should sell me today. Or I have one more day of shelf life left.

[00:29:42] And if you don't sell me, you probably want to discard me or discount me right away. Like, I'm thinking of that type of interaction. Is that a plausible future we can see? Absolutely. And I think that's even sooner than the examples that I was discussing were brainstorming for the future is that reality is partially there today. It's more around the discussion. Can you get the full visibility from upstream all the way to the back of the store? But it's not only just the perishable viewpoint of when it was processed, what the shelf life

[00:30:11] looks like, when did it come in? Inventory right now reads can somehow make sense of that. And with AI, they can make sense of what came in, what needs to be shelved forward, what needs to be shelved in the back. And how does that process get recycled? But what it's missing today is that did any of those perishables ever go through any level of volatility through the supply chain? Did those strawberries, for example, did they get exposed to some weird temperature where

[00:30:37] they picked up some humidity and then it became warm and then it became again humid. And then you're starting to grow all kind of mold in different things and you're not catching that and you're selling it. But if you're picking up these abnormalities through the supply chain and somebody goes and picks up those strawberries or picks up that salmon and asks it, are you safe to eat? And it can stop you before you even go pay for it and say, no, I'm not safe to eat because I went through three temperature changes.

[00:31:07] And through this process, one of them was 20 degrees volatile and outside of the condition of that I'm supposed to be, you now stopped your, you saved the money, you saved the bigger problem, your health. So you're able to catch that. And we're not far from that today, especially with the AI models that we've built on. It's just a matter of the implementation through the supply chain and you can ask those questions. This is awesome. I mean, there's so many topics from, you know, general merchandise to food to pharma.

[00:31:36] Well, I can't thank you enough for spending the time. We're about to wrap up here. I do have two questions left for you, but they're not really about, you know, Willie, they're more about you. So if you don't mind, I want to wrap up with, with these two and I'll go with the first one, which I'm eager to hear about. And whether it's your personal life or your professional life, what's your favorite technology that you're using right now? I mean, I'm a techie at the end of the day, so I don't think there's any limitations, but

[00:32:04] I would say definitely Claude. Claude, I would say has been transformational. We have, I think all of us to some degree, we're using ChatGPT in different frameworks, but the level of processing Claude has, especially from a application layer and being able to spin out technicalities, even when you get into like spreadsheets and complex formulas, unbelievable. So I would say at the moment, that's my favorite one. Awesome. Awesome.

[00:32:33] Now, what is something that you would like to learn in the near future in the next six, six months to a year? What's something you want to learn? I'd like to learn more definitely around the pain point of manufacturers that are within the food processing arena. I think there's certain things in our industry that are very clear and there's certain things that have been gated.

[00:33:00] And my dream scenario would be over the next 12 months to get a lot of them together and have the right forum. So we can, we can help them address the right challenges because the problem starts at the top of the supply chain. And if you can start to solve some of those issues there of what they see and where the fingers are pointed always back at them, I think you're going to have much more fluid processes through, through that framework. So I think that maybe on the business side, that would be the first one.

[00:33:27] And then I would say on the personal side, I definitely want to get a little bit more integrated into some of the AI capabilities and better wrap my head around solving some of the applications and different things. So that's on the fun side on outside of work. Well, listen, Amir, thank you so much for joining us today and sharing more about Williott and what you all do in this new technology and where it's going. This was a thrilling conversation for me and we really appreciate the time. I really appreciate it. Thank you, Reid.

[00:33:57] Thank you for joining the Next Level Supply Chain with GS1US. If you enjoyed today's show, you can subscribe to our feed or explore more great episodes wherever you get your podcasts. Don't forget to share and follow us on social media. Thanks again, and we'll see you next time.