Name: From PXM to AI-Ready Data: How MCR Safety and ACR Are Driving AI Product Discovery
Uploaded: 2026-09-17T16:31:24.132Z
Duration: 59 min 44 s
Description: From PXM to AI-Ready Data: How MCR Safety and ACR Are Driving AI Product Discovery

Transcript for "From PXM to AI-Ready Data: How MCR Safety and ACR Are Driving AI Product Discovery": Alright. Thank you guys for joining us. We'll get started in just a few minutes. If you're in the room and feeling like chatting, feel free to go to that little purple chat window and, let us know where you're from or at least where you're calling in calling from, Zooming in from today. I personally am outside of Philadelphia. We're gonna have some shy people today. That's okay. Thank you, Ryan. Buffalo, New York. Excellent. I will give it one more minute, and then we're gonna get started because we have a lot to cover. Alright. Everyone hear me okay? What's up? Alright. Well, welcome, and thank you guys for giving us your time today. Mandy Earl. I am a head of industrial market at Full Supply. At CelsoStar, we help manufacturers and distributors put their product data to work. What that means is, bringing together the data, improving it through governance and workflows, and then getting it out to anywhere it needs to go, whether that's a system, a channel partner, or really anyone or anywhere else. I'm really excited to be here today. I think, you know, AI is obviously a very hot topic. It's been for a long time. But what I'm really excited about is to have, both Brett and AJ for both sides of the supply chain here to represent, you know, a manufacturer and a distributor point of view, as well as our, really awesome partners, Impacts, with with Troy and Bob. We'll get to introductions in a second, but first, I wanted to do a little bit of housekeeping. So be sure to use the q and a tab on the top right if you wanna submit questions throughout the session. We'll hold the questions to the end, and address them altogether, but we definitely want, to provide space for this to be interactive. Secondly, if you want closed captioning, it is available in a variety of languages. Simply select your language and turn on closed captioning using the tools in the bottom right hand corner of your screen. And if at any point you have, anything you wanna say or questions, you can use the messages, or the chat, I'm sorry, to, interact with our our our backseat moderator. Alright. So I've introduced myself. I'd love to to get through it. Right. Why don't you tell me a little bit about yourself, about MCR Safety Group, and and sort of what does product data and content mean to you? Sure. Thanks, Mandy. So we have Brett Lipscomb here. Appreciate, the opportunity to be here with everybody, and it's, you know, a pleasure to to kinda share our story and kind of everything that we've got going on. So I'm the director of IT with MCR Safety Group. We're a global manufacturer of PPE, so personal protective equipment, safety glasses, garments, hard hats, gloves, boots. You pretty much you name it. If it if if it's a wearable, we're a manufacturer of it. We are a global company, and, you know, we get our data out for since we're b two b, getting our data out to our distributors is is key. Right? So that's where. we partnered with Salsify, found that as a a great tool to standardize our data, all in one place, product data, chemical permeation data, basically all the. metadata that our distributors are asking for to fulfill their end user requirements. Right? So, Yep. I think we'll talk a lot about that today, you know, in terms of, like, our customers are our distributors, but the end users, you know, from AJ's perspective, he's gonna have that. You know, the questions that they're asking, they're asking the distributors or trying to provide those answers to the data, that we're providing. Yeah. So really excited to to kinda share our story, and it's a little bit about us. Excellent. AJ, same question. Awesome. Thank you. My name is AJ Brenner. So I'm the director of ecommerce in digital air industrial. So my responsibilities are full digital channel end to end. It's the website, product content, any integrations. ARG is a 100% employee owned industrial distributor. We are headquartered out of Anchorage, Alaska. We mainly specialize in hose fittings as well as lifting and rigging. We serve a large variety of industries, anything from, oil and gas to marine, timber, agriculture. Product data is super important for us. You know, the application for our customers really matter, and, you know, they're looking for products in a pinch and, you know, that's where we come in as the, the solution. Yeah. Thank you for sharing that. Troy, let's let's talk to you for a minute. Tell us a little bit about you and about impacts. I know you're still a little bit of your of the impacts thunder from Bob, but you're next in line. let I'll let Bob give a more in-depth, commentary on Impact since he is the founder and CEO, but I'm the vice president of the product experience practice here at Impact. I've worked for, a couple of pin companies over the years, Salsify included. Here at Impact, we are well over 50 successful Salsify implementations and heading closer and closer to that, triple digit mark. So, I'll turn it over to Bob to give you a little bit more about our domain expertise in this particular segment, which, like to call industrial manufacturing and distribution. Awesome. Thanks, Troy. So I am probably the old guy in the room when it comes to product content. So I founded a PIM company in 1998 before it was even the PIM term was coined, and worked almost exclusively in the distribution space. So I've spent the better part of my career working with distributors and manufacturers trying to solve this. product content problem and have definitely seen, how both the the scope of the problem and just the approach to the problem has changed over the years. So I know we're gonna talk a little bit about that, in in a few minutes here. As far as impacts, so impacts was created as a digital services agency. And we, again, we focus exclusively in this, wholesale distribution supply chain. Both sides of the equation, working with manufacturers and distributors to just. help them, evolve, as the the digital landscape evolves and help them solve challenges, particularly around product content that we're gonna talk about here today. Thank you, Bob. Your experience is very welcome. I think a lot of us are still always trying to figure out what happens next and, obviously, how things have evolved is a is a is a good indicator of some of the things that we can do in the future. So with that, I'd love if you could maybe tell us a little bit about that evolution and where you see things going. Yeah. So this is a this is a great sort of illustration that I think kinda captures where we've all the the journey that we've been on from a product content perspective and and how that really gets significantly more challenging, with, you know, autonomous agents kinda getting into the the purchasing role. If you look way back when when when we were first kinda getting going, product documentation was in paper documents. And a lot of distributors actually had big racks of books where, you know, they had their their product information, and they were literally, you know, plugging it into their ERP, and their customers were leveraging that information in the field. So in that that first example there, you know, I've got a problem in the field. I'm looking for a repair part for this thing. I'm probably pulling out a manual, Yeah. or a document. And that really sort of that evolved to ecommerce, which, you know, really started to drive the value of product content. I remember working with a distributor in the sporting goods industry, and he he told me that, fundamentally, in the last couple years, we've shifted from being a, you know, a product company to now we're a data company that happens to ship products out the back door. And he couldn't have been more right. And if you think about it, you know, the the distributor was really challenged with aggregating all that that product data and and getting it online for their customers, but you had a fallback. Even even today, when it comes to ecommerce, you have a fallback. I'm in the field and looking for a particular part of valve, let's say. I go to the product detail page, and I'm missing a spec or something like that that I I just need to confirm. I'm gonna call someone. I'm gonna call my sales rep. I'm gonna call the counterperson. I'm gonna have the opportunity to ask that question. What's really sort of kind of highlighting, that fallback position is where things are going with autonomous agents. So you can easily see where, you know, large customers are going to leverage autonomous agents, particularly on things that they buy on a regular basis. So they have, you know, just repeat buys that they do on a regular basis, or, you know, you you've got, you know, something monitoring a a particular part in the field and then identifies it's nearing its end of life. Now the agent gets into the mix. And so the agent goes out and you know, Right. to three distributors and says, you'll I need this part. That's fact that was missing on the product detail page, it isn't gonna call anyone. So it it's you're gone at that point. So the the the importance of product data has has really never been at a higher level than it is today. Yeah. And that that's that's a theme I've been hearing in all sorts of events that I've attended. It's it's kind of at the forefront. And I know this this kind of I was seems so wacky and futuristic. Right? But but the data is showing that this is real. When a when a contractor, a facility director, a procurement manager, they're starting to use AI for research. And and I think maybe what a lot of people don't realize is that AI already decides what you're gonna see first. They decide what they're gonna show you back on that request. And so they're they're gonna scan public information. They're gonna create that shortlist, and they're gonna present it back to the user. If you're not showing up to or if you're not meeting the AI's criteria, you might as well not exist. Right? They're never gonna see your product detail page or your product listing page or or however you have it. You know, AI referred traffic over the last year, has grown about 623%. And automated agent navigation has grown over, like like, basically, ADX, in 2025. Cloudfare projects that agent traffic will surpass human web traffic, and partly that's because an agent can visit, like, a thousand pages in the time that a human takes to read one. You might be asking yourself, well, my data doesn't really show that. Like, I'm how do you know? The problem is is that standard analytics tools like like Google Analytics or Adobe Analytics often filter out crawler traffic as spam. They've been designed to do that on purpose. So it creates a blind spot for manufacturers and distributors who are trying to actually understand what that trap you know, what what is happening traffic wise and if they're actually getting agents that evaluate or bounce off your catalog. You really can only see it if you look in, like, your raw server logs or or the security network reports, But the blind spot is really, really real. So how do you avoid how do you avoid getting bounced, or how do you avoid not showing up in the shortlist? Troy, I think you have a a thought, and tell me when you wanna move forward. Yeah. Yeah. This this is good right here. Thank you. So this is normally the part where we all talk about how AI is gonna make our life easier. But this presentation or this webinar today is a little bit different because, actually, Mhmm. this is the part where, AI is gonna create some challenges and opportunities for you. So, we if we look at what we're providing for today, we've always had, as Bob was talking about, the human lens. So this, you know, as far as I can tell, we're still going to have, humans out at trade shows that need to look at your product sheets. Mhmm. You're still going to have humans on your website looking at it and that needs to look good to the human eye. But as we transitioned into the digital age with marketplaces and digital become becoming more important to distributors, Yeah. we started getting used to, you know, SEO and SEM and metadata and tags. So, some people still are not completely ready for that. Just keeping up with that and the images, is is a challenge even today, and we saw that during COVID. Yep. But, the agent, well, it's just, it's a horse of a different color. And remember, while we're do while we're getting ready for the agents, we have to maintain our our competitive lead on the the other two as we keep going. So if you. go forward to the next slide, let's talk a little bit about what makes agents a little different, than the, the digital age and the the human age of our of our product content. The first thing that, that an agent needs is, that they're not they really like to read JSON LD, unlike humans. I don't know about you. I'm not a big fan of reading JSON LD. I have it on my Kindle. we really want you're not not a big fan of it. It's not very readable, to the human. But to an agent, it's going to be the the best way to deliver the content. We need to do that today on a website, but this whole area is moving so rapidly. While your website is more important this year than it was last year, there. are rapidly developing channels with chat and Google with Gemini. We're starting to need to syndicate to catalog. So, it's really important that we stay on top of things and have the tools to be able to deliver, very quickly. So the the next thing that then starts to happen is, you know, an agent is doing something on behalf of a human, so they need the context of what they're looking for, kind of the the who, what, and why of that request that they're making. So our data models need to take that into account. And, you know, we're having to build new data models to make sure we provide that efficiently, and we'll cover that here in a second. The next thing on that list is since I'm an agent and I'm doing it on behalf of a human that asked me to go to go do something, I wanna make sure the recommendation I'm giving you is the right recommendation. So typically, most people think of reviews, and, of course, an agent can read all those reviews very quickly. I'm sure we've all seen those on our searches, but it's different in the industrial world. So we'll talk about that here in a second. And then the last thing, and this is something, you know, we I talked a little bit about how quickly things are changing, but velocity, you're gonna have to pivot probably quarterly, if not quicker than that in this world. And we implement a lot of Salsify because of its record based way of dealing with product data versus entity based data. I'm not gonna get too much into the weeds on that, but you could maybe ask Claude to explain the difference between those two. What I will tell you is a records based way of handling product data allows you to respond more rapidly than the entity method. It's a more responsive way to make changes on the file quickly. So if you go to the next slide, we'll start to really, this is where I think it gets interesting because a lot of AI webinars that I've been on start to talk in theory, and we're gonna start talking about how we really start to address how to tackle one agent's needs so that we're prepared. And, just to start off, you think about what a typical search might have been two years ago for Brett's. So in Brett's world, which is someone might go in and look for a PVC coated chemical glove. Maybe they they would type that in the search bar. But now it's probably more likely that they're looking for a glove for acetone. drum handling, and they need to know that they're in a waste disposal facility. That's the the type of search that we need now. So in order to do that, what we've done and you'll see that we have, and I'm gonna run through what is a what I'm gonna call the the the way we start looking at this with a client. And then, Brett, if you could pay attention because as we know from the way we started the data modeling, there's, we start off with a mold and NCR gets to break it for us, and we'll adapt the mold to the NCR way. So as I start to run through this, think a little bit of how this works in your in your world. But, you know, the industry is kind of the where or the who. So we might use that to start to build the context. And it might be, in Brett's world, it might be chemical manufacturing, oil and gas, automotive, food handling, medical, any number of things that might drive that, the the who part of it. But then as we start to get into the task, we start to look at, you know, what is it that we're doing here? It could be something like solvent transfers, spill response, parts cleaning, materials handling, and then that starts to drive things. And then the agent will will then want some structured data that will ask, well, why is what you're doing? So the why might be chemical resistance or breakthrough time or permeation data, any number of things that that gives the why. So if if you then look at that and then the the, that that will build us this particular model that we can structure within a tool like Salsify. And then, you know, we've gotta build that content. And a lot of this content won't reside in, standard areas like ERP and a dam that we're used to looking in. It might it might reside in, say, some channel training guides. Mhmm. And so that's where things like, AI tools like Intelligence Suite at Salsify might help us go mine for that data if it's not readily available to us and put it into a structured place like Salsify. And the last thing that I'll say is that we talked a little bit about building authority, and how reviews are the typical way that we're used to seeing an agent do that. But in our world, there's not just a girth of reviews on, industrial equipment. It's not Amazon. It's a totally different world. Yeah. So what we like to do is is is build q and a, but not just any q and a. It has to be the right voice and right tone. It can't be, sales editorial q and a. It needs to be something that's driving to the task, and why it's right. So we would do that either at the product family level or the SKU level in a structured way in Salsify. So that that's how we build it. Brett, you wanna talk a little bit about how you might bend that, when we're implementing it for you? Yeah. Absolutely. So, you know, I love the the taxonomy bridge there. You know, Troy, the industry, the task, the hazard, the where, what, the why. For MCR, you know, historically, focusing on industry first, honestly, again, I'll be completely transparent, became a little bit, difficult to manage even with tools because a lot of our products, if you take a a safety glove, for example, it applies to so many industries. And for us really to say that this glove is in a very specific industry, say welding or oil and gas or automotive, it it almost becomes limited. Right? Because then if we build the structure around that industry, then this is kind of prior to the AI models. But, you know, even searches on our website, show me automotive gloves or show me, show me medical gloves even, for example. Well, maybe we didn't tag that one, a very specific glove that could be in that industry because it has the same features. It has the necessary safety features for that industry, but no one checked the box. So well, we actually kinda took a step back. So instead of focusing on industry first from our data models, we look at, you know, the the what and the why. So what is the person doing that needs the particular equipment? Are they welding? Are they pipe fitting? You know, what, you know, what are they doing? And then why do they need it? Right? Obviously, it's a protection, but the why is the hazard. So between the combination of, you know, what are they needing it for, what's their task, what are the hazards, then we we can identify based on our product attributes exactly what products that they need. And then the industry becomes more of a metadata of sorts. So it must almost a statistic to show, well, heck. I mean, we saw someone use our this glove because it had the right protection for their needs, and it it fit what they needed for their environment in a completely random industry. Now we're starting to actually build our own analytics to say, hey. That might be an industry that we wasn't even on our radar, but now we're given to sales. I'm saying, hey. Let let's look at companies in this industry because we're seeing that tick up because of the combination of the task and the hazard. So we're building around our structured data, around attributes, not just size and colors, but into materials, surfaces, polymers, you know, all of the features that really make up the safety of a product that are specific to a particular hazard or, you know, a task, chemical permutations, breakthrough times, compliance testing, all of those things that an AI tool or even, you know, a spec sheet can drive, you know, to be able to recommend the safest product, you know, for that. So that's a lot of the ways that we've kinda twisted it as as Troy, as you mentioned. Still considering that industry, you know, but, you know, for us, you know, like I said, it's more of a a statistic giving us a a a sales edge into where we could potentially further target. I like how you are getting signal from the market, like signal from your end customers based on something that, you know, sort of happened. Like, instead of having to put things in a box, the flexibility, right, actually gave you something more than you expected, and now you're capitalizing on it. Yep. That's that's really cool. absolutely. And that's all based on how we were you know, we thought about, you know, looking at the task or the hazard first, and letting the end user tell us what their needs are. And then, organically, you know, it drives those additional data points that we can take action on. Super interesting. Let's, flip over to the other side for a sec. And Bob and Ajay, I'd love to hear a little bit from you on sort of how you're thinking about navigating the shift from a distribution perspective. Yeah. And and I'll I'll kinda start, and we'll we'll we'll reach out to AJ and kinda get their real world perspective in terms of where they are right now. I think what's really interesting to me is just the pace of change. And when you look at a distributor and a distributor who's dealing with, you know, several 100 manufacturers from a product content perspective, the pace of change just gets multiplied. I mean, literally, ChatGPT was released November, 2022. So we're we're coming up on its fourth birthday. And we're still trying to, you know, fill attributes and things like that in our product content for for SEO. And we're actually now talking about autonomous agents that are gonna come in and shop. So we're really still very much at that top level there, people first revenue reality. It's still largely true in most distribution organizations. We'll let AJ talk to to their organization. But the majority of the sales still happen with a level of human interaction, whether it's asking a question, do you have it in stock? Can I come pick it up? You know, what whatever it happens to be there. And and, likewise, in the middle there, we're still chasing just filling the core product data. Yeah. And and we've got sort of this this thing coming, where we have to randomly answer questions to to agents, which is really a challenge. So so, AJ, I'll kinda kick it back to you to just and kinda talk about those dimensions and, you know, sort of how that's where you guys sit right now and and how that's impacting you. Yeah. Absolutely, Bob. You know, like for most distributors, you know, right now, people still carry the majority of the sale. You know, if our data on our site doesn't answer a question, the customer still calls their branch contact or their salesman to close that, you know, knowledge gap. But it really goes well beyond the data. You know, if the site says there's nothing available at their branch and they need it today, you know, they're gonna go and call that branch rep and that relationship still has that value where they're gonna know that we're gonna still fulfill that need, whether it's, you know, transferring from the branch, drop shipping it from the manufacturer. So that counter person, that salesperson has, you know, a ton of value. And and we put a lot of work into our product data, and it was really compiled and built for, you know, one primary reason, which was, you know, site search. That was the, you know, the main focus, you know, two, three years ago before this entire shift to, you know, agentic search and agentic purchasing. And when we had a search, it would be, let's say, like a two inch suction hose, for example. And we put a lot of work into our attributions that's returning great options for our customers, but the reasoning in application is a whole different side of it. Yeah. You know, the real ask is they're saying, hey. I need a, you know, 12 feet of the two inch hose to replace a suction line, for example, on the North Slope in Alaska. And that kind of application has, you know, very specific requirements, needs to be, you know, be able to go under, negative 30 degrees Fahrenheit. You don't unpack what PSI needs to handle. And, you know, that really becomes more complex. And the the site search itself does not handle that at this current point. And, you know, the cost of, you know, an agent coming into the site and going to that same process, if they don't see on a product detail page the information that explains the application, explains, you know, that it can handle a, you know, partake, temperature rating, then, you know, an agent's just gonna move on from the site altogether, and they're not going to call out to, you know, one of their reps or the salesman to, get that last bit of information. They're gonna move on to the next site that gives them more information that, you know, speaks to, you know, they can handle that application and can fulfill that job. So really, you know, it's the cost of not answering at that point. So, you know, the. value proposition that ARG always has has been the knowledge. And today, with how things are changing on the agent front, you know, we're starting to change our data and how it's structured so that, it could answer questions outside of just attributes alone. And it's a big shift and it's gonna take a lot of work on our end, but in the end, our industry is slowly moving towards that in the next few years and that's gonna be larger companies that push their purchasing towards this agentic version. Yeah. We we talk about it a lot. It's also about the the cost of invisibility, right, as well as the cost of inaccuracy. I I think, you know, if you if you wanna serve your customer, you really need to serve them any way that they want to interact or to get the answers that they need. And, you know, sort of this becomes an and problem, not an or. Right? It doesn't go from human to agentic. You've got to kind of be ready for all of that no matter how they're they're interacting with you. That's that's really fascinating. Any more to add Bob or or AJ on this before we the the one thing I would add, and and you made this point earlier, but I think it's, it's a 100% true. You know, right now, when we look at, analytics on a site, we can see that, you know, Bob was on there. He searched for this term. He didn't get any results. So we can follow-up with him. We we can reach out to him. Here, that agent comes, and it just leaves. There's no bounce. There's no abandoned cart. There's nobody to retarget. And so that blind spot that you were talking about ear earlier becomes a a really a significant issue. Very true. 100%. Not just for that sale, but also the relationship that you have with that account or that user. Yep. Yeah. So we're gonna have kinda shift into sort of more of a more of a panel discussion, get some get some q and a, some answers from you guys. Let me start with this. I I think AJ and Brett, you both kind of touched on this. But when you look across those three lenses that we met we we talked about, the human, the sort of algorithm or the search agent, how would you kind of self assess your your readiness of your content? And how are those influencing your biggest content priorities right now? Sure. So, you know, in in terms of a little bit by the numbers, you know, we MCR Group has about 3,000 different styles. You know, a company has about five to six different brands, about 17,000 SKUs. And I would say, on a high note, you know, 40%, maybe 50% is is getting to a point where we feel comfortable that we're about ready, you know, for that. And and that may have sound like a high number, but, you know, at the end of the day, it's it's really not. Right? It it goes back to that comment you made, Mandy, about invisibility, Yeah. or, actually worse, you know, incorrect information. Right? In the. safety world, I think incorrect information is worse than than no information. You you have that liability out there. You know, if someone. can't find your product, okay, you lost the sale. If they did find the wrong product because of wrong information, you know, that can lead to a liability issue. So, you know, making sure that our data is accurate is is key. So it's for us, it is more of a accuracy as opposed to speed. Right? We wanna make sure. that our data is accurate. So we are we are getting there. It is definitely in the past, I would say, even relatively soon or recent six to twelve months in the past, really since we kinda started hearing from Troy for Ed Impacts and working on our Salsify project really got our our teams thinking in that direction a lot sooner, than we had before. So we're making strides towards, you know, getting our data into that standardized structure, put some some good efforts towards it. You know, it it has definitely become a shift internally in terms of getting things ready, you know, for our our next phase, in product syndication and and not not just delivering to our distributors, you know, our customers, but, you know, for for those AI agents. So we got a long way to go, but, you know, like I said, we're getting there, and, accuracy is is key more than speed at least for us right now. And. right know, it's actually distribution side, yep. it's it's pretty similar to Brett. You know, accuracy is, you know, paramount to to speed at this point. You know, I think we've done a great job on the attribution side of our data, but, you know, where we're thin still is the application and compatibility. That's the biggest part is, you know, we have jobs that, you know, need equipment and need hose and fittings in a fast manner and, you know, equipment is down and that cost companies, you know, tons of money and time. Right. And so, you know, we're really working on structuring our data so that it can give them that value in a quick manner. And if it's whether it's a human going through that, whether it's an agent going through our product detail pages and giving them the configurations through custom assemblies and configurations, replacement parts so that whenever they get to that site and they get to that product page, they can get all the information they need in one source. But as Brett mentioned, he has about 15,000. We have around 40,000 products and a little over 200 different manufacturers we work with with all different types of data. So, we're still focused on that accuracy but with that large of a SKU count, you know, speed is also relevant there too. So, you know, we're working as hard as possible to get to that point where we're happy with our data but, you know, it's a it's an ongoing uphill battle. I think I personally think Jada improvement is always gonna be kind of a journey. Like Bob mentioned, like, the the evolution of what you need is going to change as, you know, who buys your products and why they buy it and and just digital shifts. But you mentioned something interesting. I think you said two two over 200 suppliers. Right? If you're the AIG is like many distributors I've spoken to. 200 suppliers are probably not giving you all their data in the same format, in the same completeness. You know, what what are the challenges in the business impact that that you guys have to go through? Because you you are dependent on the suppliers to provide you with the data. It's not something that you can do in and of itself to to get to where you wanna be. Yeah. It it is a very complex issue. You have some manufacturers, you know, like Brett, you know, is very data focused and data first and provides, you know, formats that make it very easy on our end to enrich and provide content to, you know, our customers that, you know, gives them value in a very easy manner and other manufacturers that are still a little bit behind the curve and are still working on, you know, building that data foundation to provide from us. So, you know, depending on the manufacturer, it can take, you know, a couple of days to be enriched and enabled or, you know, several weeks to find the right content, you know, talk to the experts that can, you know, give us attributes, application, and information that's going to be valued to a customer. So, you know, it is a very complex process and, you know, I think manufacturers are really starting to see that shift now. I was just at a conference actually a couple days ago and there was a lot of talks about how product data is, you know, vital and that, you know, that manufacturer specifically was investing in their product data too to provide that. for, you know, distributors across the nation. So I think that shift is starting to happen as well where manufacturers are starting to see the importance so that their products are also represented online properly and, you know, the correct application for the job. Mhmm. Yeah. It's it's interesting because as we you know, you can do the application and compatibility and try to drive, you know, improved agent performance or or human, you know, sort of performance in terms of the the end customer. But you you can only go so far because you you need that partnership. So when you go to the other side, like Brett from the manufacturer side, right, like, how do you collaborate with your partners to ensure that the data that you do create gets gets to them and actually shows up right where you're where their end the the end users or their customers are are looking. Yep. So, yeah, as as AJ mentioned, you know, their challenge is all these different formats coming in from, you know, their customers, which are the manufacturers, and we have the opposite. Right? We manufacture, and we're trying to push this data to our distributors, which they have their own formats and their way of doing things. So, you know, we we're kinda both fighting the same battle, just kinda. different different angles at it. You know, and then we also mentioned, like, the gaps in, you know, the the analytics. A big gap that we have is whenever we provide data to our customer, you know, are they using it to the the best of their ability? Right? Are they using it all? Right? Are they misrepresenting, you know, our products? You know? Hopefully not. We haven't really seen that. But, you know, it's possible because that's where our gap as a manufacturer lies. You know, we're building this data pool. We're we're slowly building our syndication channels out being able to to standardize some of this going out. You know, we still have that lack of visibility into what are our distributors doing with it? Right? We're giving them. this data for their ecommerce, for their customers, and trying to help them along as much as we can. And then, you know, from there, it's like, okay. It's on you guys, but, you know, that's that's where we have to be collaborative with our our, you know, a a key set, right, our our top distributors to make sure that, you know, we're providing not only the data to them in a in a way that makes sense for them to consume it, but also providing a a post data service. Right? Helping them. understand what our data means. Right? Understanding how these attributes or these data points line up to the questions that their customers are actually asking. So that's an initiative that we're kinda working on internally as well, working, you know, with our our top distributors and and making sure that they're getting the answers getting the answers to the questions that they're being asked and making sure that our data can answer those questions, in as much of an automated fashion as we can. Yes. It. sounds like from go ahead, Bob. I wanted. to add something. Yeah. I wanted to add something to that because I think you made a really great point, Brett, and and there's definitely a trend that we're seeing, kind of in the industry recently. Surely nobody is is no distributor is trying to, you know, is is attempting to misrepresent the product, but I see it all the time over the last couple years underrepresenting the product. So you have spent a tremendous amount of time and effort and money, building out the the data to help support a rich experience for the distributors' customers, but not everybody is. So as a result, the distributor kind of retreats to a minimal viable, you know, lowest common denominator. Here's the basic information. So all that extra information you've provided, doesn't make it there. But I can tell you in the last two years, we have definitely seen a significant ramp on the distribution side to really proactively curate the the product content and the product data. We're seeing more distributors, actually asking us about pin solutions and taking a proactive approach to how they're managing that content so that they can, you know, drive more of your content online at a structured way. Yep. I I like that term underrepresentation of our our product. That's that that's exactly, you know, where our gap is is is is seeing that and and understanding, Yeah. and and that gives us an opportunity if we have visibility into underrepresentation. Now we can help that distributor, you know, boost, you know, hey. Exactly. We're giving you this data. Here's, you know, kinda what that means and and help them boost that in in whatever means necessary. And to your point, Bob, we're we're seeing a shift in that as well. We're seeing an increase in. distributors asking for that more enhanced metadata, you know, not just a single primary image. Right? We we we need a a three d model or a 360 degree frame, you know, of your product because, you know, we have to have that. We because that's you know, people wanna see all the sides of your product or, you know, the chemical guides and just all the different applications. So we're seeing these onboarding templates and the data being requested of us, just continuously grow and grow. And that's actually right there is the the main reason we we said we need a solution. And, you know, of course, Salsify, you know, solves that problem. But, you know, that's what drove us to it, and and we're but we're seeing more and more of that. And it's clearly, it's only gonna continue to increase. So I think. the the saying is, like, the rising tide lifts all boats. Right? Like, it it sounds like, you know, distributors, it it's not that they don't necessarily want to have all of that, but if, you know, the FMCR is the only one of their suppliers is doing it, You know, you go back to like you said, Bob, go back to it. But I guess the other question is, you know, doesn't don't distribute yourself, like, 25 people just sitting around waiting to play with the data and make it good. Like, you know, you guys have, you know what what is the reality of how you're actually working on on this data? Is it is it someone's, like, full time job for you know? What are you guys facing? You know, on on the distribution side, you know, it is small teams that are handling, you know, two, three hundred different manufacturers and and like Brett and everyone else has said that the data is, you know, structured very different depending on the manufacturer. So, you know, it's one of those challenges where we're we're taking in large, you know, bits of data and we're trying to take the most important pieces on that. And, you know, as Brett mentioned, there's a lot of underrepresented, data in our website sometimes because we're trying to balance so much data coming from the manufacturer side as we challenge them to provide more enriched data, better application, better images. And so, you know, it it's really based on us trying to, you know, prioritize, you know, larger manufacturers, you know, prioritize that data into the website first and then, you Right. try to get to the rest of our manufacturers as quickly as possible. But, you know, in the distribution model, it is, you know, thin team. So that becomes, you know, more and more difficult as, the data becomes larger and larger through time. Yep. And, AJ, just to kinda talk about that too, you mentioned small teams. I mean, that that's the same even for your larger organizations or I mean, I don't know how large you guys are compared to even us, but, you know, we have small teams as well. And, you know, product management or product data management, is even a smaller team than, you know, product management. Right? Everybody wants to to create the new product. Everybody wants to be, you know, fulfilling the product and and working with our vendors, but no one wants to manage and, you know, ensure that the data stays accurate in the system. You know, we call it, you know, master data, you know, within our ERP. And we have a really small team that does that, and and and they're overwhelmed just, you know, just in in general, always constantly busy. So I think the the challenge there, you know, as AJ kinda alluded to a little bit, is is still time and and personnel because it's Yeah. you know, we're providing this data to these agents, these AI agents or the websites or, you know, whatever the channel is. But it it's still people that have to ultimately start you know, it'll start with people, Yeah. you know, to to obtain that data, at least at least right now, anyway. Yeah. It's just to add on to that as well, yeah. you know, and one thing that I think is kind of a misunderstanding on the distributor side is that, you know, once you've taken that data from manufacturer, you know, the job is done. It's a check mark. and you're complete from that point on, but it's a constant cycle. We're constantly going to go back and reiterate data, you know, improve, you know, and ask the manufacturer to continue to work on their data. It's not just a check mark. It's something that has to constantly improve over time so that, you know, as things progress into, you know, agentic purchasing, that we're ready for that, the data's ready for that, and that, you know, our manufacturers are also pushing in that direction so we don't stay stagnant there too. That's a great point. And, you know, Bob Troy, you guys work, obviously, with both manufacturers and distributors across. What is the biggest disconnect between what a manufacturer thinks a distributor needs versus what a distributor actually needs? And when you work with manufacturers or distributors, how are they what are the best ones doing in terms of helping sell in the value of that data to their organization or working with their channel partners, to collaborate. I think the answer to that question is, that they're they're being proactive. So when we go to the industrial manufacturing space versus when we talk to folks in the retail space, I'd say if we talk to 10 customers, you you know, in the retail space, probably nine of them have an have a have a PIN. Now. when we go to the industrial space, all 10 of them have a PIN. Unfortunately, four to five of them in any given sector sector are using, Excel as as their pin, which is, you know, difficult. And this is, as we've said many times in this webinar, it's an ever changing world. I think some of that is business as usual, and websites. You know, yes, Yeah. they power digital sales, but I think we sometimes forget that that was a branch sales tool as well. But I think as AI starts to evolve, it's gonna be a more powerful sales tool for. a distributor. And if you're not providing that content, it's it's, it's it's just gonna get rougher for the manufacturer to to to, you know, to to well, it's good. You're gonna run the risk of not of of being underrepresented. So, we have a a tool called, AI tech assist, which is basically taking all the product documentation and putting it into an LLM and answering questions so the client doesn't have to go in there and download and read it. And so you can ask any number of questions about a specific product, say, you know, whether a valve is marine grade and get an answer. So what we found out is that the biggest user of that is, believe it or not, the the branch salesperson. And and I think this could also be a manufacturer's website. If branch salespeople know that there's a good place to go get answers on PPE or any number of of other categories, I think those branch salespeople want quick answers at the desk for their customers. Yeah. I think that's the biggest thing that that people are missing is this it's a sales aid. It's not just a digital sales aid, and and and we've gotta provide more and more data as we've as we've talked about to to help those tools be. more useful on Powerpoint. Yeah. I think that's a that's a great point. And probably, I I love our tech assistant tool, but you it creates a ton of data requirements. Right? So if you think about it and, Brett, I'm gonna give you a a nightmare for tonight. So what we're doing is, you know, those PDF documents contain all the secret information that didn't make its way into an attribute. Right? So that that one that we were talking about earlier, that that scenario that AJ was talking about where, you know, I'm on the product detail page, and I wanna ask the question, will this work in a high pressure environment at minus 30 degrees? There in the there isn't a spec there, but we're answering that question based on the documentation. Here's a scenario. The manufacturer realizes that there's a mistake in their documentation, so they, yes. supersede that manual with a new manual. You know, getting that out in a timely manner to the to the distributor so they can update what what's happening on their side is even that much more more critical. So on the one hand, it adds tremendous value. On the other side, it creates, you know, a need for a faster flow of of data, which I think is what we're gonna continue to see. And I think what we see and and, again, I I do see this evolving very rapidly on both sides of the equation. The the the level of content or data provided and the level of content or data, you know, leveraged on the distribution side, really, there's a broad spectrum of that. There's just a you know, both on the manufacturer and distributor side, there's such a broad spectrum in terms of maturity and understanding of the value, and what that data represents in their business, but that is accelerating quickly. I think everybody that that analogy that the the sporting goods distributor said to me where he realized, and he's the CEOs of the company, that they fundamentally have become a data company that happens to ship products out the back door. That's new for people to start kind of thinking that way, but we're we're definitely seeing it happening in the in the industry. You actually, got a question. How how do you, like, you know, if you're in an organization where maybe maybe you're a little bit behind on that that realization that you just brought up, Bob, about the the criticality of data, how do you how do you convey like, how do you convince your leadership that that the the this data matters? It really matters, and it matters enough that you, you know, need to to start treating or working with your channel partners in a different way. I don't know, Brad or AJ, if you've got some like, what worked for you guys to get that buy in or if Bob or Troy, you have examples, but, you know, usually, that's a first step is is the buy in. Yeah. For us, I mean, it it it almost kinda organ I don't know if this is a a good answer, you know, to to help, you know, whomever asked that question. But for us, it kinda organically happened. You know, we started seeing more and more need from our distributors, you know, more and more requests for the data. And that actually led us to do some research. Like, okay. Why are we why are, you know, why are we being asked of all of this stuff? At the time, this was, you know, a few years ago before we you know, almost prior to AI for sure. We we started seeing that influx of certain key distributors, right, the the large ones that everyone's familiar with in the industry. I feel like they saw the vision a little bit ahead of even us, and the buy in kinda naturally happened. Meaning, like, okay. We're starting to see this, and then marketing was tasked with going and seeing, like, okay. Why what are the needs? Like, why are we being asked about this data? Oh, look what's coming down the pipe. Like, really look at technology in terms of what's coming. Yeah. ChattGPT hit the ground running. Like, holy cow. Soon as ChattGPT kinda launched, you know, that was is the most amazing thing at the time. And we immediately were like, holy cow. We're we're already behind. We're we're we're already behind. Right? Our data doesn't do anything into this. Right? And so. the buy in, you know, kinda didn't actually happen almost as I'm saying not not like a scared tactic, but, I guess, seeing the vision. So get and I guess, you know, kind of a a tip there. Getting the buy in is, you know, understanding where you want to kind of take your company, like, where the company is gonna go, what is the vision for being able to solve your customer's problems. And I think that's the biggest thing. Right? Your products, whatever they are, solve a problem. And your your customer needs to be the hero of their own story, and you're the guide. Like, your company and your product is. the guide to help them succeed in their day to day. So you've gotta find the questions and the answers or the answers to their questions that help them solve their problems. And then if you can see that and then get ahead of you know, if if your data doesn't answer those questions that your customers have, I think that's an easy buy in. It's like, hey. This is coming. They're asking these questions through technology, and we can't provide that. So they're just gonna go to the next competitor that can, Yeah. and and you're gonna kinda become irrelevant in the space. So I think to me, the the buy in a little bit becomes, I know, trying to get ahead of that. And and, like I said, I don't know if it's a scare tactic or, you know, whatever you wanna call it, but that was, you know, kinda how how we kinda developed and evolved. Sounds like you're more just listening and and because of the care that you have for for who your end customers are and users are, you know, that it's it's really it's really it's it's it's almost like AI is ironically more humanizing because people are treating with it, like, as they would conversationally like a human versus being trained to, you know, search for a specific keyword or browse a a category tree in a in a way that is just not very natural. It's more natural to say, can you help me do this job, Yep. and what do I need to do that? Yeah. Digital digital's, one step above when you're trying to guess the right keyword to find something. It's it's just one step above trying to press the right IVR button when you make a phone call. So Exactly. Right? But don't get anybody started. on that one. Repeat Yeah. that again. are. Repeat that again. yeah. One of the things take we've to that level of hell right there, but one of the things that we've started doing with, with working with our distributors is, you know, in in most distributor organizations today, particularly, I'm gonna focus on mid market distribution, the value of the product content is getting attributed to the website. So in other words, that's why we're getting all this content. So how is the site performing? How how is ecommerce performing? And when you look at it and somebody said CEO says, we only do 2% of our sales, you know, online or or whatever it is. This can't be that important to us. One of the things that we've started working with our distributors on is attributed sales, which have actually started to re reveal quite a bunch of interesting information and helping them to educate their senior management on the value of the content. So there, what we're doing is we're merging the analytics with ERP order data. So we're able to show that, this customer was on the site on this product detail page, and then four days later, they bought that product at one of the branch locations. And so now you're able to say, okay. We did a million dollars online last month, but the site actually influenced $5,000,000 worth of sales, because that's where people were looking. to figure out if this thing was the right thing for them to purchase. And that really sort of puts a different lens on, you know, the value of the content and its contribution You know, it's interesting. That's been a long standing challenge in the retail space on the consumer side, to attribution. Right? When someone's in store versus browsing online versus. in their phone, where the purchase happens or how the purchase happens is often disconnected by the influence or the consideration that went into it. Yep. I I've I've heard said, like, understanding, you know, what what you said, Brad, but understanding the customer the end user and customer's journey and and how they actually navigate and interact or the touch points with the data that allows them to make the right decision to complete the job that they have. I think that becomes a little bit more critical as you as you start to really explore that those kind of data and and how that stuff comes together. Super interesting. While we're waiting to see if anyone has any other questions, you know, we've got three things to remember, just, you know, quick quick tidbits from the conversation today. Three lenses at once, human, algorithm, and agent that you're gonna have to be serving, and AI is differently needy. It needs structure, context, authority, and velocity, which often comes with having, you know, a structured place to govern and and evolve and take action on your data. You know, a a bias PIM PIM PXM like Salsify, but also better together. Right? Like, how do you actually work together and collaborate with your channel partner to make sure that you're getting the best data out to the market so that those end users, those end customers are being safe with accurate data, but also choosing your products or choosing to buy through you because of, you know, you're you're showing up. So if you're if you're interested in learning more or or kind of figuring out where you stand, there's a QR code down there. Please feel free to to to scan it. But, as we wrap up this webinar today, Bob, AJ, Brett, Troy, any closing thoughts or any tips for folks that are thinking about starting on this journey, or or thinking about where or how they can serve these three lenses? Yeah. The only, kinda comment I have is, you know, if you're just getting started from the data perspective, it can definitely feel overwhelming. You. know, what we have is not perfect, but what we have at MCR has been built over time. This isn't yes. We we recently, you know, kinda launched Salsify, and it has helped streamlining and making our processes more efficient and and syndication out and and all the the great benefits. But from a data standpoint, you know, it's it's been multiple years kinda in the making and making sure that we've got the right information to be able to feed out. So don't try to tag that all at once for sure. It can definitely feel overwhelming. They I almost tell my kids the same thing. Just start. Right? Find find. a screen point and just start, looking at, you know, what questions you're trying to answer, you. know, where the main gaps are in your product data, and then kinda start from there. And, you'll see that it'll evolve, and then, it'll it'll grow and and get to a point, faster than you may think. Hello, Beth. Yeah. Yeah. Just to add on to Brett, I would you know, go ahead, AJ. sorry. Yeah. Just down to Brett, you know, product data is what what I kinda think of as like an ultra marathon. It's not a it's not a sprint race. Mhmm. It's something that continuously it has to be worked over years and years. Like, At AirG, this is probably our seventh or eighth year of working on our product data and getting to a point where we're seen as a leader in product data in the distribution space. So if you haven't started by now, it's not too late. It never will be too late, but something that is not gonna happen overnight. It's gonna be a continuous journey, a continuous investment, but over time, that value will come back into the business. Yeah. That's great. And the the one point I was gonna make is, yeah, if if to to Brad's point, we've gotta get started. Yeah. E ecommerce sort of triggered this conversation around product content. That was you know, we've had a decade. or more that we've really been actively working on that on in the distribution side, and you heard today that we're still working on it. The pace of adoption around AI, like I I mentioned earlier, none of us heard of chat GPT four years ago. So the pace is just gonna overrun us. So getting started, now now is the time to get started and and to get proactive about product content. No. I just wanna say thank you guys so much for joining us today, and thank thank you to everyone who who is here and listened. This has been a really interesting discussion, and I'm just sort of excited to see where everything takes us, but also to continue to watch all the incredible things that, you know, MCR safety group and, ARG are doing, as well as, you know, continue to have the great partnership that that impacts Springs to Salsify. So, if you want to learn more, like I said, scan the QR code. There's some, resources on the docs link there, but really, really appreciate it. I'll end with I I think it's great. Just start. Let's just start. So thank you all. Thanks for having us over today, Mandy. Thanks, Yep. Thank you so much. you. Of course. care, everybody. Mandy. Bye, guys. Bye. Bye.