To Operationalize AI, Take the Blue Pill and Enter the MAITRIX, with Chris Perry and Oskar Kaszubski, Co-Founders of firstmovr

Transcript

Our transcripts are generated by AI. Please excuse any typos and if you have any specific questions please email info@digitalshelfinstitute.org.

Peter Crosby:
Welcome to Unpacking the Digital Shelf where we explore brand manufacturing in the digital age. Hey everyone. Peter Crosby here from the Digital Shelf Institute. We are definitely in test and learn mode in ai, and many organizations are struggling with how to structure and focus their experiments to move past the hype towards impact while struggle no longer or at least struggle less thanks to the learned minds at first mover who have taken the fruits of their AI strategy engagements with their clients and turned them into a framework called the MAITRIX, M-A-I-T-R-I-X, clever Chris Perry and Oskar Kaszubski rejoin Lauren Livak Gilbert and me to offer you the blue pill that will help you make order from chaos and align your organization on a meaningful AI path. Chris and Oskar welcome back to the podcast. We're so excited to dive into the MAITRIX with you.
 
Chris Perry:
Thank you for having us. It's exciting. Thank
 
Peter Crosby:
You. I should probably explain why the MAITRIX. So operationalizing AI is not for the faint of heart. It is definitely an adventure and people are still figuring it out while AI continues to evolve at this incredible pace, and so we know so many people in the community are struggling with it, scared of it, trying it, and it's just going to be test and learn for a while and test and learning really requires a framework. It really does, so that people can approach this in a very cross-functional, careful measurement based way. And you guys had to go all the way to an alternate reality to come up with a framework that would work with this crazy world. Kiana Reeves might show up because it's called the MAITRIX Framework. It's spelled cleverly, M-A-I-T-R-I-X. First question in this scenario, are we choosing the red pill or the blue pill?
 
Chris Perry:
Definitely the blue pill.
 
Peter Crosby:
The blue pill, okay. All right.
 
Chris Perry:
There are people that might take a red pill and then again, they'll wake up as if nothing ever happened and they'll just go about their life. But those who are boldly the first movers out there, the early movers in this space who bravely take the blue pill, whether it be for e-com or Omni or AI or whatever comes next, they will now be forced to go down that rabbit hole. We won't spoil the movie for you if you haven't watched it. It's been out for a while, but it's definitely worth going back through the whole series to watch the progression, and we're going to play to that storyline here.
 
Peter Crosby:
I love the way you always spice these themes up with a memorable theme. So let's dig into it. Tell us what was the impetus for you creating a framework at all, and what is the MAITRIX framework?
 
Chris Perry:
So for a while now, again, we work with a ton of brands, again, whether it be on trainings or advisory or capability work or tech stacking. And a few years ago, I think we even shared it on our podcast here, we had put together that trillion value scape because there are a lot of these landscape visuals to show all the different players in media, or in our case, all the different digital commerce tech stack partners and capabilities. Not again, to push anybody towards one company over another, but just to give everybody an idea of how they all fit into that ecosystem. And increasingly when we put that together, AI was just kind of a fleeting thought. It was starting to be integrated early days, but it wasn't as hyped as it is today. And we realized there was an opportunity as people are now focused on what's the really, really about AI to lay out all the different players in an organized fashion onto an AI landscape.
 
Chris Perry:
And so Oscar and I getting to work with so many different companies on the brand side of starting to implement it with different test partners, all the different capability solution partners who are starting to implement this in to their solutions, we realized we could start laying this out again in an unbiased independent perspective way. And we laid them out kind of lots of retailers and tech companies and then obviously solution providers and data companies onto this kind of MAITRIX as it ended up looking like. And then that obviously as soon as you said the word MAITRIX and we thought ai, we immediately paired together the movie, which was a very convenient connection. But we had that visual and anybody listening can, I know it's hard to hear a visual, but we have all of this is available for you to tap into after today's recording for free, just for democratizing great insight here.
 
Chris Perry:
But we realized that laying it out just on kind of a map doesn't necessarily show you how to integrate it. It shows the big picture, but it's not the process. And so we kind of took then that movie theme, and if you've watched the movie, this doesn't s spoil the end, or if you haven't, it won't s spoil the end for you. But there's a little boy in the movie who in the real world can bend the spoon. He's called spoon boy, it's very appropriate name. And then when he was bending the spoon, it kind of created this infinity loop visual, which then we took towards this go-to market strategy loop, and we realized that there were actually eight Ps, if you will, that, and this isn't to try to overcomplicate the four Ps of sales and marketing, but there were eight Ps that all of those different capabilities and use cases really fall into that actually does help us mentally see how to tie them all together and start to integrate them into our real go-to-market processes.
 
Chris Perry:
And so we obviously won't have time to double click into all of these Ps, but some of them are more important and maybe lower hanging fruit than others in the short term. But basically we start kind of on the loop and you'll be able to see this too. There's platform which is really the through what or through whom do I sell what I offer as a CPG, as a consumer goods organization. That would be the retailers or any platform I might sell, a device that I might sell through in the future proposition is the what I sell. So it's my portfolio, it's my availability, it's my assortment mix, it's my pricing. Then we get to presence, which is how I show up, it's my content, it's my search placement on that digital shelf or other shelves, other placements. Then we get into promotion, which would include obviously traditional trade promo, but more importantly media, retail media, paid search from that end.
 
Chris Perry:
And then we would get into performance, which underpins all of this, the actual metrics and measurement, and then another thing that kind of underpins, but on the visual, it's the other upper part of the loop as it comes back around. We kind of lumped people process and partnerships together. They are separate Ps, but they're not for all intense purposes. We are all the people in these organizations who need to leverage AI within new processes and often need to be tapping partners, not just our retailers, but the actual third party solution providers because we don't produce AI tools. We produce snacks or laundry detergent or beauty products. So we probably really do need to outsource some of that at least in the early days, to someone else that can plug into some other partners along a process. And this again, keeps feeding itself. So this isn't to be a high level theory. There are a lot of use cases as well within each of these that brands are actively testing today and we can obviously tap into them as we go through our discussion.
 
Peter Crosby:
And tell me, because you've been rolling this out now for a couple of months
 
Chris Perry:
And we've been doing it for a while before we actually put a MAITRIX name on it
 
Peter Crosby:
Before you actually, yeah. And so when you introduce this to the teams around the table, how do you find them engaging with it? I'm just wondering whether having this as really providing them with sort of the guardrails, the kind of map that they feel like they've been missing, what are the constituencies thinking about where this might take them?
 
Chris Perry:
I would say Oskar I'd love your perspective too on this. I would say I think the first part is with any new area of change, it's kind of like that original Microsoft screensaver of stars coming at you, right? It's a lot of things. It's all over the place. It's hype, it's headlines, it's real use cases, it's real challenges. And again, for that first mover leader of change out there who's been tasked to do AI overall or it's an AI czar or it's a content lead or digital shelf lead that is actually tapping into ai, maybe that's one of the low hanging fruits. It's a lot to try to manage. So the MAITRIX framework isn't intended to stay theory, it's supposed to get actual, but it is a way to organize it into what is arguably a go-to market strategy loop that we all already had, but those eight Ps were already there.
 
Chris Perry:
It's just what we use in those eight Ps to spin this flywheel or loop as we call it. And I think obviously, I think by making it simple and kind of a fun catchy way, it's helping create a storyline to share with executives or other stakeholders internally to get them to buy in. But more importantly, when we double click into each of those ps, that's where the rubber meets the road. I mean, that's where it's like let's talk about presence and how AI can help from an end-to-end content perspective, from creating it to housing IT and organizing it to syndicating it, to measuring its effectiveness, to optimizing yourself for search, obviously within the textual content as well. So before we even get to the measurement part of all of the digital shelf. So that's where I think that's going to be the greater values that double click in, but the first part is trying to organize it. But Oskar what is your take on that? But
 
Oskar Kaszubski:
I think you're absolutely right. The problem that people have is with AI is that there is so much hype and it's really hard to distill what the truth is. So in a way, that's why MAITRIX does work very well because we are talking about the truth and we are trying to basically break down the hype from the reality, and they also allow the people to really segment it and they say it's like, okay, I really understand maybe the presence, but I really need to think through about platforms because platforms might be harder for me to understand because I'm still stuck in a cell phone based app based world that I don't think beyond it in terms of all of it. So I think it just helps the people to navigate it, especially if they see some of the new vendors because we had a lot of the brand leaders kind of reach out to us and it's like, you know what?
 
Oskar Kaszubski:
I had just new company popping up. I couldn't understand what's their value proposition, but I immediately started to put them in a little bit of your AI box in those eight factors in terms of all of it. So I think that just helps people to see it this way and understand a little bit better what's the value proposition of a lot of the solutions. But I think we had this conversation a few years back when we talking about 20 years ago when digital started, we had this wild race into building digital products. Everybody wanted to be the second Amazon. We see the same thing with AI at the moment. There is a lot of software companies coming up with the product. There is a lot of venture capitals investing in AI based product, and it's not necessarily all of them are valid. Some of them are just purely a marketing hype. So being able to see that building AI company, it's all about building a company that can actually fit in into ecosystem of brand world that helps to also distill what's true, what's not true, what's useful, what's not, what's useful, what's not useful.
 
Lauren Livak Gilbert:
And this is hard for brands adopting anything new at a brand, especially a large brand can be difficult. So as you're going through this framework and you're working with brands, Oskar what do you think or what are you seeing as the big challenges, whether it's legal or IT or data? What are the big things that are coming up and how are you helping brands to navigate that?
 
Oskar Kaszubski:
The biggest challenge that AI currently have is that AI basically is a little bit overhyped, right? We are basically a little bit, a little bit you can even see behind me is going to flash the NVIDIA stock, which is getting a little bit of a beating in the last few weeks. But the problem, what it is, is we almost want to live in 20 30, 20 40, but we are actually in 2024 and there is some limitations to everything what we have. So I think the hype is, thank God, the hype on, for example, the mainstream media on CNBC where they were basically talking about that AI agents will be here tomorrow and it is going to fundamentally change the world. That's a little bit dying down. I think honestly, I was so grateful. Open AI actually kind of released their five step pathway to artificial general intelligence, which is basically everything what we need to build in ai.
 
Oskar Kaszubski:
And we are at the step number one and AI agents under step number three. So I think they are basically resetting the expectations, what's actually doable today. I think the last expectation we need to actually reset a little bit is that AI can today replace majority of jobs out there because we see a lot of brands that are basically just restructuring the team. They're thinking that Elon Musk, they can basically run entire company with only 20% of the people. We starting to see is that the reality is actually hitting a lot of the C-suite leaders that it's not going to be as simple, it's not going to be as quick. But I think in general, the funny thing about AI and what's happening at the moment is that it's also revalidating the IT teams and a lot of the IT teams ironically are actually with ai. And the reason for it is because they are conditioned to launch capability as a checkbox and AI actually requires a lot of hands-on testing, and they are not actually ready for that hands-on testing and hands-on application. So AI is not something like you can just launch and have a checkbox and basically just forget it. It constantly needs to be iterated
 
Peter Crosby:
Tuned, tuned and sort of monitored and improved over time. That's really interesting. Essentially the company of your name is First mover and you, you've been talking about that a lot. I had to sort of dive in on that because first mover, sometimes there are a lot of companies that don't want to be the first mover. They're like, I'll let somebody else figure out and then I'll dive in and sort of be the second round. So there's what kind of companies should be digging into this framework and do they have to consider themselves first movers to do it? And then secondly, do you view this, you're talking about the opportunity for it, getting good at this is a career making opportunity I would think, but so I'd love your impression on that sort of what companies, what do you find the mindset is at the places that you're going in and the qualities of their thinking about it that makes them willing to dive in with this framework? What needs to be in place?
 
Chris Perry:
I would throw in, so we like to talk about first mover advantage, right? So we don't talk about ourselves in the third person. We actually just happen to call our company the group of people that we represent that we like to be ambassadors for. But I always like to say, I mean obviously a really large enterprise CPG is going to be a different type of first mover in some cases than the challenger brand who only goes to market in e-commerce on day one, pre-acquisition or scales it and has a much more, I do this, this is literally how I grow, versus we're dabbling in this and it's still kind of a sidecar effort and we're trying to figure out how to integrate it in. So I don't think there's any one profile per se. I do think smaller companies, medium companies, especially medium ones that have resources but aren't so bureaucratic yet that they've got the agility and the resources.
 
Chris Perry:
Typically you see them pop up and they're big enough a name as a company that they speak at events, and so you hear about the things that they're testing. They're also a little more willing to be the case study also for a discount or deals or value exchange with some of those solution providers. So you see some of that. I did want to Peter, that made me think of one story as just a short aside. I've heard from a leader before, they did not want to, this was way before first mover became a community, but from a leader that I was worried when I said, don't we want to have a first mover advantage? He said, I never want to be number one, I want to be number four. And I'm like, whoa, that's unusually specific. Why number four? Because the first person is the disruptor, but they also break the rules and they often either get blamed for the negative part of it and aren't necessarily seen for the positive.
 
Chris Perry:
The second one looks like someone trying to sneak in the third also, but by four it's mainstream and it's okay. We can say, well, others did it. You let them do it. And they had this very quick rationale for it, which so I don't want to say I don't want to rename us fourth mover, that just feels silly to me. But it was that idea of you don't always want to be first, but my point is, if nothing else, I like the iPhone and I like getting the new iPhone, but I'm not going to sleep outside the store to get it the day it comes out or the day before it comes out. I can wait two weeks and get it the same way and I'll be just fine unless my phone was broken, I might then more urgently go in. So I would say first mover can be stretched to an early mover versus an early adopter comes a little bit after that. But again, some of it is like early mover could be, I tested it in one of those PS a few times. It doesn't have to be I completely scaled my business
 
Chris Perry:
By adopting a completely new process holistically. And again, to Oscar's point, gutted everyone and reoptimize everybody to be automated. So I think part of it though is, and we have our whole change formula, which I know we've talked about on the series before, our sheared eight factors of change. But honestly, those factors of change aren't just for e-commerce and Omni. They're also for AI or any area of change. And so I know you mentioned the IT team specifically, but SharePoint, all functions are starting to realize that as E becomes silent in e-commerce and AI is actually underpinning all potential areas of commerce and go to market strategy, everyone has a chance. It's no longer some brave sales or marketer sales leader marketer who said, I volunteer is tribute to go into e-commerce. Everyone has a chance to be a part of this change and figure out how to implement it for their company in their area.
 
Chris Perry:
And honestly, they'll be heralded as heroes because they tried it out. If you're category management, maybe there's a way with the data partner to leverage AI to better analyze the data and find those insights. If you're it, it might be helping be the connective tissue between some of the capabilities of not tech stacking, but I think there isn't one profile from a size perspective or a function. I think it's just really being, as a leader, it's getting alignment to actually test and learn something on a small scale that won't cause any problems, get the use case study that worked, then ask, can we scale this a little bit bigger and then so on. And so I think that doesn't sound as sexy, but that's how you build change over time.
 
Lauren Livak Gilbert:
And that's how e-commerce really kind of came about too, right? It is just the way that you should approach all things. And I think with this, not forgetting the level of change management and education, I mentioned legal, right? Legal needs to educate the org on what you should be asking or shouldn't be asking when you're talking to vendors. And then the marketing team needs to know the same questions and needs to understand how AI is used. So I love that it's a simple formula that you just outlined, Chris, that's like start small test and learn how we like to approach many, many things in e-commerce. It doesn't change when it's this new and big and shiny object. You should approach it the same exact way.
 
Oskar Kaszubski:
But let me actually, let's talk about something kind of related to it. It's like how people actually approaching the change based on what we are actually seeing is one of the challenges that we are basically seeing is that the teams are basically, and it could be marketing, is they're not taking enough time to really understand this, how to work with ai. Basically, it's a little bit like this. Imagine if you took from 19th century just random people and put them in a 20th century, CPG, that they never work in organization. They never got onboarded. They didn't allow them to make mistakes and make full of themselves because we live in a very impatient kind of a culture. Everything needs to be, now I have to be an expert. Now, no matter if I'm going to say the truth of not truth, I'm just going to keep on going, right?
 
Oskar Kaszubski:
We're talking about as long as you have a conviction in what you're saying, you can actually succeed. But the problem, what it is is with AI is as opposed, for example, to e-commerce, AI takes a lot of more iterative work that basically needs to happen. Even if you look at the image generation, oh my God, in the last two, three years, we went in from what Google was showcasing with Image generation into now we have eight to 16 XG options, different options for image generation with all of them pros and cons and competing with each other, midjourney, grog, Dali, et cetera. And it's constantly changing and we have to keep on figuring out what's the latest, what's the best, because it's also a massive competition at the moment without a lot of investment coming in into ai. So everybody's trying to one up each other.
 
Oskar Kaszubski:
During the e-commerce, you kind of had to wrap your head around it. You had to have this aha moment that you basically can buy something online and it can be delivered to you as opposed to going to the store. For me, it was like a very linear process. AI is a little bit more iterative where we actually have to take everything under consideration. And the one thing is that we haven't even touched upon it is what type of a social change AI will actually introduce to our world. I'll give you an example. I think next month there is a product coming out for $99, which is basically AI friend. And what AI friend is, it's an on device AI that you can basically almost interact the same way as interacted with the movie in the movie her. So you can have a conversation, ask for advice, ask for encouragement, et cetera. We don't know. We're
 
Lauren Livak Gilbert:
Not ready for that
 
Oskar Kaszubski:
For about anybody else. But the point of that is we don't even conceive what type of a change that's going to actually introduce to potential shoppers, to potential consumers. We are not even thinking about this because we are going to probably see this within the next four or five years in terms of the social fabric. How is it actually changing? So I think we need to be a little bit more patient and a little bit more hands-on this famous thing is sometimes within CPGs we have too many people that want to lead, but nobody really wants to be doing the grassroots grassroots work. That's where we are actually finding up. We almost have to step back from the leadership role and it's like, I'm just going to sit down today and just experiment to get better to understand this, how to put those different blocks together. So I think that's missing at the moment.
 
Chris Perry:
And I
 
Peter Crosby:
Think that kind of takes, go ahead.
 
Chris Perry:
Sorry. I was going to say, there's one saying that came back to me recently that represents what Oscar just said. It's kind of an aggressive one, but I think I kind of say it because then once you hear it, you can't, am
 
Peter Crosby:
I going to have to bleep you?
 
Chris Perry:
No, you won't. No, no.
 
Chris Perry:
There's that saying that kind of famous quote that everyone wants to go to heaven, but nobody wants to die. And in today's world, everyone wants the heavens benefits or whatever heaven you believe in or don't. They want the end goal and not all of that iterative work. And honestly, even when Oscar and I are testing out, developing our own, just playing around with mid journey and creating our own imagery, there's a frustration. It's like, oh, I didn't prompt it right? Or Oh, I didn't put that in. And even as we're getting a little better at it, we're not experts perfectly yet as no one is, but as you're doing it, you're like, crap, that didn't work that well, or, okay, I'm going to have to do this and do that. But now the good thing is on a personal level, you can fail as many times as you want.
 
Chris Perry:
Nobody knows until you show them the final good. But we're so used to just jumping from do the proven task, get the proven result that test and learn, or again, fail fast. Those are nice things to say. Companies don't always really want to let you do that. So that's why honestly, some of this could be test some of this as you can, not with confidential company data or anything, but test some of this on your own personal tasks, home care, planning a trip just so that you get really comfortable with it so that you already have the synapses connecting when you have to apply some of that type of stuff internally.
 
Peter Crosby:
So the process of doing, I mean, part of this in order to do the test and learn here is some of it costs money. What do you see companies doing to fund this kind of, and it not only costs money, but it costs time. As you're saying, iterating and testing takes up time. It's not just, I'm going to do my thing, as you were saying. So how are you seeing companies figure out their correct investment level to operate these experiments in a way that sort of makes sense? Do you have any examples of that?
 
Chris Perry:
Yeah, so I won't name direct names, of course. Yeah. I'll say, but this goes a little bit back to, and again, some of this sounds like brilliant basics and it's like, yeah, yeah. But again, the brilliant basics work. That's why 50,000 New York Times bestsellers have just reiterated the same leadership principles in 70 different books that have all become the top seller, all the same stuff that Dale Carnegie told us almost a hundred years ago. It is the same stuff, but always worked for all humans. Back to Socrates and beyond. It's like, but that being the case, one of the things is we are kind of hitting, and actually one of the things Oscar and I also are working on that's parallel to AI pulling is bringing Catman back into e-commerce. E-commerce to date has been focused on pull the levers, measure the leading metrics, and you're like, got it.
 
Chris Perry:
Content search, search placement, media, traffic generation, availability, reviews, all important things. But that's just how you grow. That's not why you grow, right? That's, I did content and media and assortment to drive bigger baskets as a way of incrementality or to drive more trips because I'm a snack brand and I want to create new occasions. So it's all those traditional catman principles that we applied to a mature in-store business and often then declining as things shifted online that we're not yet fully applying to the online business, that is actually also starting to mature for some, it's still growing, but starting to mature. So we have to be a lot more creative. So to your question about where do you find this funding, how do you get started? Actually in some cases, brands almost have to, because like what I've been doing isn't getting me what I want anymore.
 
Chris Perry:
I've gotten all this funding before. I've gotten all these resources, all these capabilities. It's not that it's not delivering growth, it's not delivering as much growth the way I've designed it. So thus I need to squeeze something else out different. I need to do a little something a little differently. What might do it? Well, I know the hype is telling me ai, how do I actually get past the hype to how I could test it? And then that's where, again, how do we do we use the AI tools out there that are helping me, again, if I use content, every brand is trying to get accurate, complete content. And generally it looks really awesome because now we've gotten brands involved and there's designers and we've created these beautiful PDPs and beautiful brand stores, but have we really measured them for effectiveness? Since most retailers don't give you an AV testing capability?
 
Chris Perry:
How do I look at the effectiveness of it, right? Walmart target club, they'll measure your content completeness on their scorecard, but they're not telling you how effective it is. And all I can normally look at is what my ROAS is or what my sales look like as a result of an update to my content. So tools that help use AI to measure effectiveness become a way to get more out of the thing I'm already doing that I need to show more results from. So there's a little bit of that. I think that tension to squeeze more juice out of the thing that used to generate more juice, if we use that analogy. But there's also an interesting thing too, I think of one use case. I know a number of brands have been kind of doing, I was just talking to Oscar about this earlier, but not maybe formally or publicly, is reviewing, I would say auditing the reviews, ratings and reviews for insight, not just for content, but for your assortment.
 
Chris Perry:
What should I do with my renovation next based on what people, and we've had this for a long time that I almost think it went dormant, right? We've had the bizarre voices of the world, the power reviews, we've collected all these reviews. We had this database of insight. Did someone use it to get more incrementality? Not always actually when asking around. So what's interesting now is the e-comm team or the omni team is now looking at AI is like, Hey, we got some pretty, when using AI to look at what I could control, we got some other insights on what the product could be. And so now it's like I don't have the funding to do that. And if I'm an e-comm leader, I don't necessarily have the purview to change my assortment or my innovation pipeline, but I can go back to the brand and say, this is real consumer feedback.
 
Chris Perry:
It's been analyzed at a level that we wouldn't have been able to analyze it, look at the insight, and we can connect this potential sales opportunity to your brand pipeline development opportunity. Boom. Now there's an opportunity to test something out through one of the quicker routes to market some of the companies have. And if you're a smaller company, you'd probably have a more agile way of going to market with something, a test product. So we're seeing a lot of those. I would say it's where you're getting pressure to show more results from the resources that have already been given to you because everyone's coming back around with their 10 cup, they want the money back. If it's not going to deliver extra growth, this is a way to potentially derive extra growth and people are willing to see if it is because we think AI will bring that advantage.
 
Peter Crosby:
Sorry, I was just trying to figure out what question to ask next.
 
Lauren Livak Gilbert:
Yeah, same here. So we have the budget question, which I think you covered, but I guess the one piece was you can't afford not to do anything. If you could touch on that a bit from the budget standpoint. And then the other question we had after that was examples of the brands driving real impact. You started to go into that with the ratings and reviews and things like that, but did you have any other specific examples you wanted to go into for that one?
 
Peter Crosby:
Yeah, if we could close with some use cases that people, just to get them noodling, that might be good. So in terms of, Lauren, do you want to just have Chris or Oscar continue? Just add to that prior
 
Chris Perry:
Oscar could build on that? I'm sorry, I
 
Peter Crosby:
Didn't mean
 
Chris Perry:
That
 
Peter Crosby:
To No, that's okay. No,
 
Chris Perry:
That's
 
Lauren Livak Gilbert:
Okay. I just think that's an important statement to say because it'll be a good, you can't
 
Peter Crosby:
Afford not to statement.
 
Lauren Livak Gilbert:
Yeah.
 
Peter Crosby:
So Oskar are you comfortable just jumping in with that
 
Chris Perry:
Jump in like you were building off of what I just finished?
 
Oskar Kaszubski:
Sure. Okay. So yeah, Chris, I mean, the other thing I would want actually talk about is it's also gen ai, right? And what's really happening with all of the different brands looking to implement gen AI within their strategies. And what we are basically seeing is that especially for content completeness, especially for creating content sets for not only the top retailers like Walmart, target, Amazon, Kroger, et cetera, people are starting basically build content sets for other retailers. They're actually starting to address also the brands that they haven't had the love, but because they didn't have the budgets and really it's being actually look upon as a way to scale it. The challenges that we have is that from a tax asset perspective, we can actually get it done pretty well. Of course, there are going to be some issues, for example, with legal reviews because AI is very creative, but it's not necessarily very well adopt yet to basically create legal approved claims that are very specific, right?
 
Oskar Kaszubski:
Think about it is we almost asking AI to do two things to be creative and rigid at the same time, which that often creates a lot of those issues with the text assets, the images. On the flip side, on the product sides, to be frank, based on where we are with all of the different image generation generators, what we really are doing is we are basically creating different layers for those images. We are not able to create a perfect prompt for a perfect product image because AI has a lot of limitations, like for example, understanding the three dimensional relationship between objects, and it's very hard to generate the exact copy of your actual product. The text generation got a lot better versus what we've seen in the past, but it's still those type of image production requires a lot of handling by the designers and our directors to be able to actually get it to the right level of production.
 
Oskar Kaszubski:
But it can speed up things like, for example, texture generation, background generation, some of the setting generation, people generation, because AI is very well versed into generating a lot of those human assets. So it does help, but I think fundamentally where we can actually see the most successes to taking an asset that maybe had zero content and suddenly create at least the full-fledged content that we can actually push through Salsify, for example, into the retail sites. The question that I will have is what's going to happen, for example, where the retailers will actually get their generative AI at scale, that they will say it's like, you know what, we can actually produce those content assets using AI and maybe create a situation that's very similar to what Chewy is doing, that all of those content assets are basically thematically basically fit in the store on message, on brand, et cetera. So the ification of the content, is that going to become, that Walmart's going to be doing, Amazon's going to be doing, et cetera. So that's the question where the content gen AI is going to lay. Is it going to be brand driven, retailer driven? We already seeing that, for example, Amazon is actually using some of the gen AI into addressing some of the content, especially on the third part, marketplace items, et cetera. So it's an interesting kind of a tug of war in terms of what's happening at the moment.
 
Chris Perry:
And I think a core message here is that this isn't a time to just run into the wild and spend everything and hope for the best, but you can't afford not to do anything. You have to do something, right? So pick a P that's important to your business. One maybe you've already been investing on, but without AI yet. So you already know you've got some tailwinds boosting you and that you bought yourself a little bit of margin to try something different, which undoubtedly will help you if you're getting some efficiency or some more effectiveness out of it. And try get some additional stakeholders internally. I'm not trying to promise this on behalf of the solution provider community, but I bet money, I bet my own money on the fact that some of them would be happy to do a case study with somebody if in exchange, as long as you could share it with them at an event or share a public case study or parts of it.
 
Chris Perry:
Because again, they too want the proof points of the tech that they're developing. So you might be able to actually get a better deal. And you should ask your own partners, what are you doing with ai? Can we test something? Even if you get one soundbite, one proof point on one small sub P of these eight Ps, you can bring that back to buy yourself another P, right? Or buy another sub P. I'm making up all these, but it's about that as Lauren as you said, it's doing exactly what we did in e-commerce and Omni a little bit faster because we've already been through it and we know it worked. Once again, I know that stinks have to do it all over again, but we have the playbook, follow the plays, and just be that leader of change because no one else is doing it.
 
Chris Perry:
So you're still going to be, you're going to get all the benefits of the success and the long-term impact of learning, failing, optimizing, doing better, but because no one else is going to do this instead of you. But if you don't do it, someone might at some point take that lead role. So I think it's find a stakeholder, find a sponsor, a patron internally, find a partner that you're already working with or a new one, willing to try something, get alignment that you can share some of that data as a case study. Obviously that has some personal benefit you might get out there with your own personal brand as a leader at that company. It helps with recruiting and thought leadership with retailers too, that you're doing that kind of work. So start in one place. You don't have to boil the ocean to start with AI on day one.
 
Oskar Kaszubski:
But the one thing that we actually love about AI is that because AI touches every single part of the organization, it also uncovers so many basically rotten foundation that people don't have the right data, don't have the right processes, they have too many silos within the organization. So those AI projects are also a massive tweak to the way the organization is functioning. And basically it's like your annual checkup where it's like, Hey, your cholesterol is too high, maybe your weight is a little bit too low, whatever it is. So those AI projects are really uncovering some of the underlying causes, and especially I'm very happy about it as a massive data gig that those inefficiency and the lack of data is being exposed.
 
Peter Crosby:
So AI will make you feel embarrassed about, but that all good IT projects do that. It always exposes where things need to be filled in. And I was thinking of that as you guys were talking. I was thinking once again, it comes down to our audience, to our listeners, to our community of DSI and first mover to kind of be the ones to chart the course here, which as I was talking about earlier, can also be an extremely career defining journey. So what certainly I would leave everyone with is go to first mover.com. So first mover, MOVR, no e.com/MAITRIX, M-A-I-T-R-I-X. So you got to get all those spellings, right? Or if you want to, you of course can always go to LinkedIn. Oscar and Chris are very active on there, and you can say, how the hell do you spell that again? But certainly first mover.com/MAITRIX and this framework, all of it's there for free you to take advantage of. And we are so grateful for both of you, one, for doing the hard thinking around how to make things like this approachable and organized and productive. Taking what you learned from your clients and putting it into action, and then of course sharing it with the DSI. We're really grateful.
 
Chris Perry:
Thank you for having us. Thank you.
 
Lauren Livak Gilbert:
Thanks.
 
Peter Crosby:
Thanks again to Chris and Oscar for the trip through the MAITRIX. Before you take your blue pill, go to digitalshelfinstitute.org and become a member. It will be helpful on your journey. Thanks for being part of our community.