Jay Roxe with Inriver

SUMMARY KEYWORDS

Product Information, Data Management, Inriver, Marketing Strategy, Manufacturing, Digital Economy, Single Source, Supply Chain, Cloud Native, AI Readiness, Product Experience, Automation, Syndication Distribution, Enterprise Software, Channel Growth

00:00

05:13

We do do a great job to make you sound and look like a hero. You are. You deserve that attention because what you're doing is a vital not just not just in your it's vital globally because we need to have these conversations and never ever stop and then take these conversations and market them properly. Market them get them out, rolling fast. Publish, boom, go. PR, you want to be on other podcasts? Well, I'm in the middle of it, so we can definitely help with that. Help with that, and and get that marketed as well. It's it's a one stop shop. All you have to do is go out to Industrial Talk, reach out. We do so much more. You need to know about that. Anyway, all good Inriver, the product information management system. We talk a lot about hey, what do we do? How do we succeed? What are the steps? Right in this conversation, Jay does not disappoint. Enjoy the conversation, Jay. Welcome to Industrial Talk. How are you doing today,

06:26

Scott? It's great to see you. It's a fine day here in Boston, and I hope you're doing well, also.

06:33

I can't complain. We we had a conversation offline, and yes, it's a spectacular day, and I'm still glowing, basking in the glow of of my son proposing and California. So I'm all happy as a clam.

06:49

It sounds like it was a fantastic whole family event. It was.

06:55

We drank too much. We ate too much. I'm I'm in detox now. I'm detoxing, working out. it.

07:03

It's worth it's worthwhile celebrating. Yeah, coming out at the end of summer, going, ooh, maybe I needed to add a little bit of detox coming out.

07:13

Yeah, yeah, do. All right, listeners, we're going to be talking a little bit about Inriver. That's the company we're going to be talking to Jay, but Jay, I have a I have a question. This is this is sort of cool. You know what I did for the first time, and I had the equipment, I had the gear. So in the back of my head, I've been sort of mulling over. I said, you know, wonder if I can ground my own beef. And so I'm thinking to myself, and it just happened over the weekend, and I'm looking at some patties, and they were a combination of Chuck and and another beef. Anyway, and I said, "I've got this piece of equipment, which is a KitchenAid thing. He got an adapter, which is you know he's sticking on there, and you know what? It's a game changer. It's completely different.

08:01

What did you end up mixing together? Do you remember?

08:04

08:23

Now, I guess I guess the question becomes: Once you're done with that, are you gas grill, charcoal grill, or smoker?

08:32

This is all I did. I made hamburgers. Made hamburgers. I had over seven pounds of ground beef, and I said, "Okay, here it is. I get ground beef. I know I make some hamburgers, smash burgers. Right? It is a smash burger, and I don't care in the kitchen. All good. Just throwing it out. Salt and pepper, nothing fancy. They are completely different. The flavor's phenomenal. It's tender. It holds together. You slap some cheese on it. You're riding high. You're in Flavor Town in a big way.

09:04

I'm going to give you another one to try for Flavor Town as we're coming into winter. Yeah, what's that? Evermade bison chili.

09:13

No, no, no, no. But it's very lean, right? It

09:16

is very lean. It is very flavorful. You do it with some chili, and it holds together really well. It's even slightly sweet, so it's a yeah, it's a family favorite.

09:31

Yeah, see, I'm all into the protein. I'm going to eat like that. I'm a I'm a carnivore at heart, and so I'm going to. And I just just dawned on me. I said I can up my game, I can up my game, and it worked. It was phenomenal. Anyway, it

09:45

sounds like a great experiment.

09:46

Yes, and now I'm I'm cursed for the rest of my life. I'm not going to buy anything, and it and it and it cooked differently. It just everything was like what this is. Anyway, that's it. I'm sorry. I digress. I just wanted to see if you ground your own beef. Then I was in, wow.

10:04

I do not grind my own beef, but I have been known to try different ways of mixing it up for some of the family favorites.

10:12

There it is, man. Bison. I I I have that. All right. Before we get into the conversation about Inriver, we need to know a little bit more about Jay. Tell us who you are.

10:24

So, I'm hi guys. Jay rocks. It's not just a name; it's a sentence. And the took you a minute there, Scott. So, and I run marketing for Inriver. By way of background, I'm a geek. Started my career as a software developer, and the have built product and marketing teams since then. Live outside of Boston, three kids, and my lovely wife, and sort of recovering and aspiring runner. Let

11:01

me ask you this: How does somebody get being, you know, all into technology and and then sort of navigate to marketing? What what what draw drew you there?

11:13

So started my career long time ago as a developer at Microsoft. Did five years as a hardcore systems dev, probably still running on half your machines. Sorry about that.

11:26

Then

11:27

got the MBA and ended up doing the product management and product marketing side of it, and then just took advantage of a few career opportunities. And look, it's very useful to be if you're marketing a technical product to be able to speak geek.

11:45

It's it's spectacular. I hate. I came to the conclusion. I ran a couple of companies. It's all industrial, right? And I would outsource or you know hire some people to do marketing or whatever. So I did that. Here, go do it. You know, get some eyeballs, ears on it, whatever it might be. Then I realized that I was lazy, and then I said, "I need to know how to do this. I need to understand I'm being lazy. And so I ended up saying, "I started out with doing, you know, podcasting because I said I got to tell my story. Then I got to market it. Anyway, it was it was a phenomenal transition. Just because it's one thing to have the technology, or what I did is like an industrial maintenance company. It's another thing to be able to tell that story and to create those relationships and to create the buzz. And that's a whole different. It's it's a good thing.

12:41

At the end of the day, a lot of what we do, either for marketing or our jobs or even like meeting each other, life is storytelling, and it's the question of how do you best tell that story and make it resonate with the person you're talking to. And

13:00

I think it's imperative more than ever to to inspire that next generation. You got to tell your story,

13:06

and more importantly, teach them how to tell their story. And

13:10

yeah,

13:11

like all my

13:12

and you're just so funny. All my kids, they they're all into media. They tell their story because I pushed them. It's like got to do it. Yeah, I'm. I'm writing some stuff down. By the way, there are uncomfortable pauses in the conversation just because I can't. Apparently, I can't talk and write at the same time or spell. I can talk, but the spelling is absolutely awful.

13:37

I don't. I don't think anybody's seeing what you're writing. So the spelling, your second grade teacher is not going to come double check you on that one. No,

13:46

my wife will. What are you doing there? All right, what? What? Tell us a little bit about Inriver. What is that company all about?

13:57

So Inriver does product information management. That's one of those and product experience management. That's one of those things where you're like, what is that? But at the end of the day, I'll keep it in the category we were talking about. Products have stories, like you. And in order for peep somebody to know about your product, they need to know the story. And the easiest, the way I explain this to my mom, what I do is when you go to Amazon, and you are looking at the information, and it's got a product description, and it's got the size, and it's got the bullets telling you where it was made, and it's got the images. All of that information has to live somewhere. In this case, it lives in a product information management system, and if you're just doing it to Amazon, that's easy. When you're controlling information that has to go out to 100 different endpoints, and you're working with Granger and distributors and lots of other places, that's when you really start to need more sophisticated. And more orchestration.

15:02

So let me paint this picture. I'm a company. Let's say I'm a manufacturer. I'm a manufacturer, and I have these relationships with various vendors. Yep, do because I've got this thing and that thing, and and it and it's all over whatever, depending on my size. It's everywhere.

15:17

Yep. Are

15:19

you saying that that Inriver, if I if a manufacturer uses the platform, it it normalizes that all that information into a way of being able to understand it because right now it's not. I'll tell you that right now.

15:36

Look, most people need a single source of truth, Scott. Yeah, if you're a manufacturer, you're pulling it out of your ERP. You're which may be you're pulling out of your PLM. You're pulling it out of all sorts of places. Realistically, the significant number of people we talk to are on Excel. They've just built a sheet, and you you would be shocked.

16:00

No, no, it's it's been that way forever. Even when the systems I were deploying back way back when, that it was always. I can't believe Excel has done a great job at marketing that. I'm telling you, everybody's just jeez.

16:16

Excel. It is a fair statement that Excel is the second best tool for almost anything.

16:24

It is

16:24

the but when you start getting to a certain scale and you're having to move with a high degree of automation.

16:32

Yeah.

16:33

Look, the let's take a look at what life is like for a manufacturer today or a brand. Most of them have seen the number of channels that they that they have to distribute to go up by about 140% since. When you say

16:49

channels, explain what that chat. What what is that channel? If

16:53

you are, it can be you're getting it out to somebody else's website. You're getting out to your own website. You're getting it out to one of your distributors, you're getting it out to a partner, and the number of places that people are selling through now just went up vastly during COVID, and has continued to grow. And equally importantly, the number of things you need to know about each product has gone way up, and depending on where you are, that may be different things. Like it, if you are working with regulated products in Europe, you may have to store different information about where it was made, the what type of batteries it has, other information for a digital product passport. You may need to be able to come up with a bundle of products or understand the relationship between some of the products. So there's a lot of information that needs to get stored, optimized, translated, distributed out, and then you need to make sure that it actually got there properly, so evaluating it because about 40% of information that's online is wrong. About

18:10

no way,

18:11

it's really well. It it's people, and it may not be wrong in big ways. It may be something introduced misspelling. It may mean floated an image. It got rotated 90 degrees.

18:26

Yeah, is it?

18:28

19:22

pilot. The props.

19:25

It's far more context, and so we're seeing that degree of product information being required for the way people are finding products today.

19:36

See, but interesting when you start to talk about LMM Large language models. I can't say LLMs. Clearly, I can't. It's it's again. If 40% of the information that exists out there is just sort of inaccurate, that's coming in to the results as a result of that prompt. It's like. And everybody thinks it's gospel, but I didn't know that.

20:03

Well, and some of it's not wrong in major ways. Like if the picture of your lawnmower got rotated by 90 degrees, that's not the end of the world. There's some stuff though where you see people with more significant errors, and it's where are the LLMs? See, I got it on that one. Getting getting getting their information from, and how do you make sure that's accurate? And the other side is as people are trying to take something to market, how fast do they have to do it? Like the trying having a nine month release cycle is done. It's people are trying to take products to market faster and faster and faster.

20:52

Okay, so we've we've pretty much hammered on the challenges that exist out there in the market today.

21:00

Yep,

21:00

but Inriver has a solution. This product information management platform take take us through it. It's like okay, we got all this in the data. It it's everywhere. Bad data is everywhere. It just is. And I'm not cleaning it up. I don't want to. Been down that road, and nobody ever cleans it up. Never nobody does. It just stays out there. So how do how do we deal with that?

21:24

So let's start at the beginning. It's how do you onboard your data, and it's everything from what's the information you get from your suppliers, and how do you use AI to clean that up. It's how do you integrate in with your other upstream systems, one or more ERPs that are going in. Are there just spreadsheets you need to upload and things that people know? But even once you have the information in the system, Scott, it's what do you do with it? How do you make sure you've got the right copy? How do you use AI or manually or AI to enrich it? Make sure it's consistent. It's got the right categories. You've got a good long description and a short description. You've put it in the right number of languages, and it's gone through the workflow of all of the people who need people or agents who need to be involved in doing this work, so how do you make sure you've orchestrated that process of getting the data into the right shape?

22:34

But how? This is always the the you know million dollar question. Is your yours yours is amped up in complexity because you're just you got all these channels you got stuff. It just you you can't go out to Acme and say hey I guess you could but Acme your your data's I need that but your data's terrible and and how I don't I don't even know where you begin how I mean, back to how.

23:03

Sure. You you go back to how, and it's a process when you begin. I mean, how do you understand where you're getting? Look, people are selling through these channels mostly through those channels today. The question

23:21

is:

23:22

How do you reduce the cost of doing that? How do you make it faster, easier, and enable them to go out through more channels? So people have some set of data today, but it's the you can build your connection into your ERP system or your other upstream systems, and we have the connectors pre-built that enable you to do that. You can onboard data from the other vendors that you work with. You can upload the Excel sheet and then use AI to clean it, or you can take data from any of your suppliers and whatever format it's in, clean it up and make it usable, and that can be both human-assisted, but also applying some applying LLM and learning across it to make sure it's consistent and appropriate.

24:21

See, and then when you get to the point where you got all this data, let's say it's look if you're out on video, I have my hands hold. Here's the data. It's, but really, I don't need that. I just need as I hold my hand up. It's a small amount. This stuff takes up room, and this is stuff that is just noise, and I don't want it. Versus something that is, yeah, that's relevant, and I want it. Does does your platform just you know get rid of that? I'll take that. That's what's real important for my business.

24:50

Well, I think it's the question of you create your single source of truth. So it's a question of you bring in the data and. Do the work to clean it and optimize it, but then you have to know as you're going out to the various channels you're sending data to, Grainger may ask for something different than Amazon may ask more than Home Depot than somebody else. So that process of how you syndicate it-it's called syndication distribution. Make sure you're getting the right information out there. That's the next part of the process of taking the large amount of data you've got, figuring out what needs to go to each channel, and how do you send it in the format that they want to see it,

25:42

I'm telling you, my ears are bleeding, my eyeball is bleeding. Just FYI, with the level of complexity and the level of challenges, the the the Inriver platform, the solution itself. How long has it been around?

25:56

Well, the we are a technically Swedish company. We were founded in Malmo, Sweden, and next year will be our 20th anniversary. So always in

26:07

that space.

26:09

We were we were so we were one of the first of. There's a people have been solving this for longer, but we were one of the first cloud-based, truly cloud-native platforms around this. And again,

26:25

it's it's it's exploded. I I mean, just the amount. And here's the here's the consumer side, right? I go out to Amazon. I'm searching around for whatever thing I apparently need, which I don't really want, but I gotta get it for something, and it's out there, and that information is out there. I'm researching it. I'm looking at all the information that never had, you know, you know, go back 20 years, 30 years, whatever. That that wasn't, I didn't have that information in front of me now, I do, and so the consumer is just more, just have to look at stuff and get that information, and then you, the company delivering that, has to be able to provide that too as well.

27:14

Well,

27:15

it's crazy

27:16

nailing product information. You've defined a great scenario, Scott, because nailing product information, particular like it's not just the first thing you buy; it's all of the replacement parts and the follow-on and the service parts that you need to go buy. So I have a challenge that I'll give you that'll annoy you and your wife. Go go into your refrigerator tonight. Yeah, break one of those little clips that holds the shelf to, and try and find the replacement part. I've done this. You can end up pretty deep down some exploded parts diagrams, and if you do this on a your refrigerator, it's inconvenient. If you have a multi-million-dollar carrier HVAC unit or reactor that's down in the field. You'd better be able to find exactly the part you need and know that the right one is going to show. And that's a very important market to the actual brands themselves. Yeah, we were talking with one customer that had a several 100 million dollar gray market parts problem, as in people were having trouble finding the genuine replacement part, so they were going to gray market alternatives, and that. So it's a real bottom line issue that hits many of these companies.

28:42

Yeah, I'm I'm I'm an eBay guy. If I can't find it on eBay, it doesn't exist. I don't know why it was eBay and YouTube. There you go. Place anything.

28:55

You'll find pretty much anything once you start going down those routes.

29:02

So, as a company, if I if I go and I use this platform, and I say, okay, I'm getting my I'm getting my act in in order here. I'm

29:10

you know

29:11

diligently doing my stuff. Type of bottom line value can I see?

29:16

Yeah, I mean, I think we are seeing. Look, we help people measure it across a couple of different angles. It's how are you looking at faster time to market, and we're seeing customers pretty routinely report 30% 30 to 40% faster time to market. You're seeing wow, and maybe for individual parts. You're seeing people reporting 20, 25 percent less work associated with maintaining the product data and taking it in, and then you start getting things like people, like what happens when you go purchase some. And it shows up, and they measure it in inches instead of centimeters, or the color isn't right. You send it back, and in after a couple

30:10

of colorful words,

30:12

I'll leave that to I'll leave that to you to figure out. Yeah, when you start sending it back, like that's really expensive for the organization, and it may not even be possible in some manufacturing scenarios.

30:27

Yeah.

30:28

So that's the type of thing that we want to help people with. But I want to level set for people for a second, like where people are on this journey because Scott, we talked about bringing the product data on and like where it is. We interviewed about 400 heads of marketing, heads of data from industrial manufacturing companies larger than about $250 million.

31:02

Okay.

31:03

and the big journey that a lot of people want to be on is how do you get to saying you're AI ready? Like you can't, you probably can't do a podcast right now without the letters AI coming up. But everybody wants to say they're agent ready, they're AI ready. We of the 405, 117 of them said they were AI ready when we actually tested it and said, how many of you have end-to-end automation? How many of you are monitoring your answer engines? How many of you understand the number of channels you're managing? Zero.

31:47

The old goose egg. The old

31:48

goose egg. So it's interesting to me. You go out and you look at LinkedIn or YouTube or any of those, and it sounds like it always looks like everybody's like kicking back. They've got their coffee and they've got this swarm of agents that's doing their job for them. It's not reality. And the important thing for brands and manufacturers to know is, like people that are saying they've totally moved to AI right now, that's not reality either. Like a lot of people are at various stages journey.

32:26

Yeah, I still think we're just scratching the surface, and I think there's this journey that I think that there will be a craving for that human component, that you know connection, that human connection. Because I, you you brought up LinkedIn. You could tell the posts that are AI generated. It's like they have no soul. You can just tell they might be flashy and all this stuff, but you can tell, and it's over and over and over again. And I think I look at AI, I look at its application, I see it as a tool, and and that's great, but I believe that in the long run, it'll still be humans.

33:07

I think you're I think you're right, and but you're putting your finger on something interesting, Scott, which is content generation right now is free, like it

33:17

is

33:17

creating content and that is on brand, underwritable, meets your legal requirements.

33:27

Yeah,

33:27

can be easily translated. That's not free, and that's when you start entering into wanting to have specialized tools, and importantly, having humans in the loop, having people that say, "Okay, I understand that's been generated. I want to review that and make sure I actually trust it, and it's how we're going to take things to market.

33:52

So, as we wrap this up, I am intrigued by one thing, and that one thing is, this is great. It seems like it's super needed. The one, if I contacted you, Jay, and I said, "Hey, Scott Mackenzie wants to get his, you know, act in in in order here. How long is sort of a deployment?

34:12

Look, we've got company. We've got customers that depends as always on where it fits in the organization, we've got customers that have gone live in 30 to 60 days, and the but it depends what use case you're taking live.

34:32

Yeah,

34:32

a 90 day quick start is pretty common, and then you've got if you're doing an enterprise grade transformation and you're doing integration with your SAP system, yeah, more downstream. Then you're into an enterprise grade deployment. Yeah.

34:47

Okay. So you also answered that question by saying, "Hey, it can, it can connect with my existing technology stack. I've got a gazillion dollar ERP here. I've got this other. I got, I got this technology stack, and we just need to figure figure out where.

35:06

One, you know, one of the things where like we do a lot of work with ERP, and we built the only certified connector with their BTP, and so it's like having those connections both upstream, like where you're getting the data from, and downstream, where you're sending it to, super important.

35:27

Yeah. No. Oh, it's important. All right. Anyway, now one last question.

35:34

Yes, sir.

35:35

What are the roadblocks? I got it. I need it. It makes sense. It's getting far more complex out here. I have a solution I need to deploy. It can connect. Why am I saying no sometimes?

35:49

Good question, Scott. I think it is like with everything in enterprise software and all of the solutions. There's always competing priorities, so it's understanding within your organization what's going to give you the fastest value and the fastest value for your buck. The so it's not that people are saying no to PIM. It's we want to help them build the business case and see the return that they get, and that's what can make it an obvious investment. It

36:25

seems, quite frankly, a no-brainer in the sense, and I understand what you're saying. There's a certain a decision tree that takes place from a technology perspective. However, as as this world, this market becomes far more complex in the demands of the market and all of this stuff. I just see that it's you better better get going, better do something. I

36:48

think we're I think we're seeing manufacturers make the same transition to a digital economy now that retailers had to make as we all turned to online shopping and then got supercharged during COVID.

37:06

Yeah, no, I'm good. All right, as we wrap up, we need to know how to get a hold of Jay. Really, a great guy. Jay needs to tell us where to get a hold of him.

37:18

Really easy. Go to www.Inriver.com or just drop me an email at j a y dot r o x e@Inriver.com.

37:27

I should have coached you on the beginning. Are you out on LinkedIn?

37:32

Absolutely.

37:34

I I'm lazy. I I grab the URL and then I paste it.

37:42

On LinkedIn,

37:44

because that's where everybody goes. It was Jay. I'll go find him on LinkedIn. Okay. Lives in Barston. Yeah. Okay.

37:53

And oddly enough, Scott, there are not very many people with the last name ROXE on LinkedIn, and they think I'm related. Isn't that

38:00

amazing? So that's pretty doggone cool.

38:04

So,

38:04

given there's a gazillion people on LinkedIn, of which, whatever I don't know. Anyway, you are fantastic, Jay. I've been looking forward to this conversation in a big way.

38:14

Scott, really nice to talk to you today.

38:16

Outstanding. All right, listeners, we're going to have all the contact information for Jay out on Industrial Talk. Stay tuned. We will be right back.

38:26

You're listening to the Industrial Talk Podcast Network.

38:32