What Enterprise AI Transformation Actually Requires

Speaker A: Foreign.

Speaker B: Edition of the Pure Signal podcast. You have myself, Ryan Medine, a manager in our marketing solutions practice. It's the podcast that gives you all the latest and greatest on AI, uh, machine learning, transformation and more. And of course I'm not doing it alone. I'm with my trusty co host, one of two co hosts that we normally have are North America CEO uh, Kevin Erickson. And joining us on the Pure Signal pod, we brought you some amazing guests this year and this is no different as we welcome in Beth Ann Wilhelm and Aaron Gratzinger to the Pure Signal podcast. First off, thank you both from joining us in your respective cities and super excited to have you to talk some big transformation and orchestration talk here. It's a lot more exciting than it sounds, I promise you that. And uh, to do so, uh, Bethann, I'd love to start with you for you to just give an intro and talk about what it is you do for Cordera and where your expertise lies and then Aaron, we'll, we'll go to you.

Speaker A: I came into Credera through the legacy Leap Point uh, acquisition. Uh, I was actually one of the founders of Leap Point so been there quite a long time, excited to be part of the Credera family. Now I serve as a chief architect over Adobe content supply chain so designing uh, implementations of Adobe products like Workfront aem, um, integrations using fusion with various Adobe and third party platforms and I just get know, really into designing and implementing Adobe. That's my specialty. And uh, yeah I've also mostly specialize in regulated industries, financial services and healthcare. And so now I'm uh, as part of Cordera I also uh lead our pre sales practice for um, the Omnicom Adobe practice, the new newly formed Omnicom Adobe practice. I lead the pre sales function there and I'm also enterprise ah architect over a new investment that Adobe um, has made within Omnicom for the Agentic AI operating model.

Speaker C: That's a. She's a busy, busy, busy person.

Speaker B: Yeah. Well thank you.

Speaker C: Yeah. So thank you for carving out time to talk to us. I mean it's a lot of roles so um, really appreciate that.

Speaker D: Two primary roles that I have at Credera, one is overseeing sales and delivery for energy industry sector and then two is supporting Bethann and others in our Adobe channel partnerships and go to market motions really uh, bringing agentic marketing solutions to life. So those two things keep me busy as opposed to Bethann's you know, list of five or six. But uh, good to know. Uh, Ryan and Kevin that were just the, uh, poor man's Jake. So, you know, thanks very well.

Speaker C: I've been saying that for about 12 years now, so why should we stop now?

Speaker B: No, I think it's, uh, first off, yeah, thank you both for. I know you both are extremely busy for taking the time. I think it's super interesting to. To have you both here today because the themes that we've been talking about this year, Kevin, uh, it's kind of been all over the place, but really they've kind of honed in on, like, companies are really going through transformation and how to orchestrate that transformation and, and bring it to life. Which is exactly why I think it's super fitting that we have you both here. And I think from the lens of when we looked at like last year, I think a lot of companies were kind of focused on the initial, like, Enterprise Wide IT Data foundation and getting that in order in this world of being ready to use agentic and plug in agents. And then the immediate shift after kind of that was settled was, okay, where do we find the most roi and where's like, most of our money being spent? And where can we find efficiencies and, um, cohesiveness? And that's where all eyes have kind of turned into this, like, marketing transformation and orchestration and looking at, you know, the tooling that they've had for forever. And is that the right tooling? Do we have too many tools? Too little tools and all these types of conversations that we're hearing, and I'm sure y' all are hearing as well. And so, you know. You know, Bethan, I would love to start with you just from your perspective. In the last three to five years, are you seeing what I. I'm seeing, or am I a little off base

Speaker A: there in terms of, like, too many tools? I think we'll be.

Speaker B: The focus has been. Yeah, like where the focus of these CMOs and these companies have been.

Speaker A: Yeah, I think what we've seen is like. And even going further back, like 10 years, we used to have very, like, large solutions, very large solutions that covered many things and that didn't work very well for people. And so then they started to branch out and get specific solutions to fill specific needs. But then they didn't talk to each other. Now we're seeing the rise of all of these integrations and really getting everything integrated and talking to each other, which is the right thing to do. Uh, but now we layer on AI on top of that, and now people are just scrambling to try to figure that out. So it really has been a progression where we started with very, very large solutions that covered a lot of things but didn't do any of it very well. Then we have solutions that do specific things well, but they don't talk to each other. Now how can we integrate it, keep everyone not having to manually re enter things, integrations, and then how can we speed up what people are doing through leveraging AI? It's an interesting evolution and it's something that we have to stay in front of, which has been also fun but challenging. But yeah, we're definitely seeing that progression specifically around the integrations and AI.

Speaker C: Yeah, I think one of the, I think one of the things I'm excited about this topic today is obviously we've been talking about, uh, the different components of the elements where it's going to go. Whereas AI, agency, workforce, and we've been talking about a lot of the human side of that. I'm really interesting, Aaron and Bethann, from your perspectives of what is coming from a hyperscaler, what is coming from a platform? How does a platform evolved into this new world order? How do we help our clients and just really the businesses where they think about, hey, what does need to be done where, you know, and again, that builds on Bethany and I think the comment around kind of integrations and all that element. But I think it's a fascinating question of, you know, I'm sure we have some of our listeners out there that potentially are being like, well, I can just go use a genetic workforce, I can create myself or I'm going to pull from here, Do I even need these products anymore? And um, you know, obviously we are, uh, in the middle of all that, helping our clients think about different things and when you should do it. So I'd love to get some perspective on just like, when do we think, think about it, where do we build our agency workforce from? You know, what would be different considerations that a client or a company would use? Because again, they have many options and not certain there's a right one or a wrong one, but there's many of them. I'd love to think about how you guys are thinking about that.

Speaker A: Right. So I would say like everyone does want to use AI, but AI is only going to be as good as the info we're feeding into it and the info that it has available to it. So having a really good underlying data foundation is going to be key. And that's also where a lot of customers tend to struggle as well as can they get a good data foundation in place. Um, so that's going to be super important for integrations and for AI in general. Um, so I would say it's two things like making sure we have a really solid underlying data foundation. That way the agents have access to the right information for it to be presenting those recommendations or those insights. Uh, but then additionally, before we start kind of putting in different tools. So let's say if a customer does want to get away from a tool or go into a different tool, um, this isn't so much a technology answer, but you have to understand your processes first. We need to know who's involved and we need to know what the overall process is and try to streamline that process. Then agents can then be woven into that process to help streamline things. Maybe what was being done by a person could be done by an agent in some cases. Not in all. Uh, but I think understanding the process is going to be pretty key for these technology implementations. Everyone wants to spend bring into, let's just implement this new tool and maybe it'll save the day. And that's very rarely what happens. Uh, uh, at the legacy Leap Point company Nick Debened at our CEO used to say, um, if you build it, they won't come. So kind of off of a field of dreams. You could build the best solution with the best agents. But if you're not getting the people involved in making sure that there's change management and effective process, people aren't going to use it. So I think underlying, making sure we have the processes identified and weaving in the appropriate people and agents along that process and then obviously the underlying data foundation needs to be sound.

Speaker B: Yeah, uh, Aaron, I'd love your opinion on this because I think it's so interesting that in this age of technology is stronger than it's ever been, the conversations centered around establishing your data foundation and then the people are like the next big thing you have to conquer in the sense of like process and, and governance and all of those things. Aaron, from your perspective, is that been kind of fascinating to you how we were kind of moving away from the people element when machine learning started to take hold, but now that AI is here, it's like, oh wait, no, we got to go back to the basics of like, get your people ready for this, get them ready for change. How are they going to react to that change? And then also that underlying data foundation. Because I just from past experience with projects, you know when you tell somebody that, hey, you're going have to spend all this money to redo a data foundation, oh, and then you're also going to get your people on board. Before AI, that was like the hardest sell in the business, I feel like.

Speaker D: I mean I want to double click into a couple of things, Ryan, that you said and said. I would say my thing with AI is that the technology is advancing faster than people organizations can catch up to. And I think your leading companies are at best still two to three steps behind the solutions being rolled out. And I would say the majority of companies are probably five to six steps behind. Right. Just, they just got to digitizing an offline process and now all of a sudden I've got to start thinking about AI and I think you're ones that are more the tip of the spear have been looking at AI as hey, help me do individual tasks more quickly. And now these solutions are saying, hey, we can actually sync up the entire workflow. You can think about agents as additional team members on your team still have the human in the loop, but how do you bring them together? How do you onboard them in a meaningful way? How do you empower them to make decisions just like you would others on your team? And then how do you reserve certain things for the human to still be in the loop to make, make that happen? And so as it relates back to platform, I think Kevin, you asked this question, it's something for us to consider is it goes back to I think process and workflow, right? Creatives still have a backlog of creative work that they need to work on. Marketers are still looking for strategic insights they want to cascade into campaigns. And so if they are thinking about things in a campaign centric world or creative workfront or work workflow world, um, we probably don't want to upend that right now. We want to look at that process end to end and determine how we can truly make agents and humans work together. Once we start to see that working to the full amount of efficiency that we can gain and see the incremental lift from that, uh, we can start to think through maybe splicing together types of work that weren't uh, coming together before. Right. Maybe campaign creative are more intertwined, maybe the customer experience. People that have largely been separated on managing that evergreen digital experience now get intertwined with campaigns because we've got kind of an always on uh, data enrichment piece that's sending out signals. And so I think to go back to, I think your question for Kevin, why the platform still remain relevant is because they are the owners still of uh, the workflow and kind of a grounding force that you can still gain a lot of lift over what I would consider to be the next several years and then we'll have to really test and learn to see how people come together, how they leverage agents to the fullest extent and then what the technology will become to where we can make even bigger shifts is sort of bringing everything together to deliver the great experiences we want for customers and driving that top line and bottom line growth. So a number of responses to. I think the different things you said but um, that's kind of where I see is we're just fundamentally behind and catching up.

Speaker C: Yeah Aaron, I think that makes a lot of sense. I'm actually this is. I've got for both of you. I'm curious as I've listened to you makes sense. I'm curious around even the construct of we're you think about AI driving task acceleration now AI driving really rebundling of how things get done. So we're using an agentic workforce and it's been synonymous with really human capital replacement. But I am curious now when we think about a platform, where is it actually replacing a human element and that's why we call it an agent versus it's just functionality that's in the platform. And I can see then the question for me then will become I know there's multiple questions here is like where is that different? But then also how do you price it? Like I understand I have to have a license to use a platform. I understand how that thinking about that. But now how do I think about an agencic workforce? And you know, am I buying an agent? Am I buying an outcome? Am I buying a piece of access to that? So I would love to know how. Because how we're starting to think about. We know we're in a lot more conversations today. We've been talking about the last couple episodes where you know, I think our clients now are starting to have a better understanding of what it cost. You know, what is the cost of an agent. I can actually compare that token that, that CPU cost to what a human resource would be, you know, different productivities but you can actually start to have that. I'm curious how that will play out also with the platforms to where you know, I mean I don't think Adobe's in the uh, business of giving anything for free. I've never heard them be accused of that. You know, so it's like, you know, I'm curious how you guys are thinking about that.

Speaker A: I do think that at least in the, for the especially certain industries like regulated Industries we still will need like human in the loop. Um, but I do think we can see some efficiencies with agents kind of drafting the initial content. So something we're building as part of an existing project is can that agent kind of give me an initial campaign brief based off of some insights or recommendations that I'm seeing. So can I give me that first draft and then I can work with the agent to refine it? So instead of me sitting and trying to write it all from scratch, can the agent be drafting things, getting things updated more quickly than it would have taken a human. But a human still needs to look at it. Specifically in regulated industries. I think over time in non regulated industries, we're going to see the agents doing a lot more of the workflows and then there'd be some key approvals. But at least in regulated industries, everyone's going to, they're going to have to see things along the way, um, just because of their underlying, uh, compliance regulations. But in terms of how we price that, this is something we're actually trying to figure out at this time. Uh, because you can get out of control with the cost, the cost of the agents running, the use of the tokens that you could go and spend way more than what it would have cost for just having a resource in place to do some of that work. I think in terms of pricing it, there likely have to be some limits put in place for how much individual users could be leveraging those agents and if they hit some type of max. So I know if you have like your own personal accounts for like Claude or, or um, Chad GPT, like they kind of max out. You can max out on those things if you've done too much in a certain period of time. We might be looking at something like that. Um, but then would that start to become disruptive as well? Because if you need people to be continuing to generate those things, if, if people hit limits, is that going to cause more bottlenecks in the process? So it's an interesting, it's an interesting thing that we'll need to think through and figure out. Uh, and I think we're just starting to see how this is looking like in the marketplace and everyone's trying to figure it out right now. But Aaron, I would love to hear your thoughts on it.

Speaker D: Yeah, no, I think those are great points. I think from the pricing standpoint, um, I think you have to start with what are the outcomes you want to achieve and then what are the right inputs to driving those outcomes. And I think the struggle right now is a model is going to go to the full extent often of giving you something that's probably way more uh, analysis and energy and usage intensive than maybe you need to for that task. And so we've got to set that right context to where the cost of delivering that output is right size. To Beth Ann's point, I don't think yet we've done a great job and I think everyone's still trying to figure out how do we do that right, mapping the technology available to us today and then figure out then what is the appropriate value to put on that impact. Part of why I think is because we are not really feeling the true cost of using these tools. Like I think they're artificially deflated as the scalars have gone out. And at some point there will become a, hey, the price to do this all the way up through this vertical integrated chain is going to be 10x 15x 20x of what you've been paying today. And so we will need to start to do that mapping and that right sizing of things to then give people the approximate amount of value, um, uh, of what the, what the true kind of landed, uh, solution is going to be. So I think we're still in the middle of that um, I would say as it relates back to like human ingenuity. Part of what we're trying to say is that hey, humans are going to be more on the strategic inputs and trying to determine what is most relevant, what is most impactful and giving that guidance to agents, um, because they're still kind of operating off of that input. And I think we've got to figure out what is the value of that to an organization particularly I guess in terms of shaping that operational cost of those AI tools. Uh, just like cloud has financial ops and there's a lot of rules and systems in place and uh, wisdom and how to deploy that tool effectively. We will need to come up with a similar finops model for AI, um, and address that on an ongoing basis.

Speaker A: And in terms of kind of the comment about Adobe pricing as well, um, right before summit I was in the um, the customer advisory board sessions and they actually asked the group I was in, how would you like us to potentially price these agents? They actually didn't have like, they were trying to elicit input from the customers to see like, what would be like a more attractive pricing model for you if we were to start to sell these agents.

Speaker C: You know, I think what's interesting and I don't know if there's A answer to this question today, but I am m intrigued about the different parts of the value chain. Right? You've got, you know, where you're, where you're the pipes, you've got the energy, the token consumption, you've got the actual platform, the access. And so I think it's gonna be really intriguing. Bethan I would have really liked to have been in that session. I would have probably had a snarky comment or two if you asked me what my preference was. But like, but like, but I think it's actually, I do think it's gonna be an interesting component because the true total cost of ownership could be interesting because you may be having to pay for access to something through a platform. You might be having to pay for consumption on a multiple different basis and you may have to. Then there's going to be an accuracy element too. You know, I think that's. And how you have to continue to refine. So which I think is kind of actually some of the fun of this to me to where it really kind of starts to create a bit of a new, you know, new business model. And I think even like if you're. Yeah, I think it's going to be really intriguing about like some of the work that we're doing here is, you know, what's going to be appended, where, you know, what will be within. I may be thinking about the architecture wrong. That's why we got an expert on. But thinking about that, how that ties back into specifically an Adobe product, how could that be in my own environment where I've got agents that I have from different vendors, different sources, things I've home created. Ryan might have his own personal agent that's trying to figure out how he gets a quarterback for his college football team. Uh, but that's also running uh, at the same time as he's trying to do his day job. And I think that that's uh, an interesting element of where this is going, going to go. And you know, to Aaron, to your point right now, I do think, I think the true cost of all that is probably being uh, shielded a little bit. And it's going to be interesting how that starts to become more known and how companies then react to that.

Speaker A: And actually Adobe is doing something so they have something called their Adobe, um, Agent Orchestrator, which is supposed to coordinate various agents that do different things. So basically an agent to organize all the other agents. And I think what we are also going to be doing as part of our agentic AI operating model project is we are Also going to create, we're calling it a coordinator instead of an orchestrator because we don't want the terminology to be confusing with Adobe, but our own agent, uh, to coordinate agents and we'll be able to coordinate some of the Adobe agents. So there is a way for our coordinator to coordinate some of the Adobe agents as well as some of the agents we will be bringing as part of what we already do at Cordera and even some of the um, Omni AI agents as well.

Speaker D: Yeah, I think the opportunity Kevin, for our clients and for us really is how can we articulate a very complex total cost of ownership that has multiple vendors, multiple platforms, multiple inputs that could widely vary, even going down to the source energy, feeding those um, data centers and then cascade that into a very simple pro forma that a CFO can look at.

Speaker C: Right.

Speaker D: Because I think even if the complexities are exponentially greater, the roll up view needs to be understood enough that an executive and a board level can know okay, this is worth us doing. And there's very simple like what if scenarios that we have. And so I think that's the opportunity that I think um, and kind of the complexity that no other solution has had because you used to have one SaaS calculator, one kind of cloud calculator and now it's you know, three or four things that come together to make.

Speaker C: But well I think that's part what's kind of, that's what's the interesting element of like do I got some AI, do I use the AI now it's going to be a different element. We've been talking about the importance of the cfo, uh, more coming into this scenario here because of just the competing cost structures that are associated with that. So I really actually like this next stage that we're evolving. I don't know who the, it's going to be interesting to see who the winners and losers are. Particularly as we're obviously in our moment of tremendous IPO elements of that. But I think you know, certain platforms and the hyperscalers, I mean and then really just individual other products that are associated with those ecosystems. I mean it's going to be really fascinating to help evaluate how do we help a client or a company, how do you evaluate where do you want that? You know, where is that append. So again if I'm m a marketing use case, Adobe is a logical place where you would continue to build off that investments you've made. Well, what if, you know, what about find my broader elements of that? I mean there's going to be, you know, automation and agencic workflows that are going to happen all over the place. And this is what we're talking about today. You know, again, I think this is going to continue to go around brand new business processes, brand new way things get done. Um, so it's going to be a fascinating time but I think definitely continues to bring us to a different question that companies are asking. I would have thought, you know, in my own simple way a few months ago that we had been focusing more around the process reengineering side of the equation, uh, a little bit more before we get to the cost side of it. But we've really quickly gone now to the cost side of it. Where is it going to be? How do we work through that? And you know, in some ways how are you going to put the genie back in the bottle a little bit? Right? I mean you're going to have a lot of activity going on, uh, through companies. And you know, I think Aaron, your point before about just even thinking about how do you limit the response and energy or really the amount of time of processing that it goes. I don't, you know, your typical early stage chatgpt where I get, I answer uh, a 32 bit question and I get you know, War and Peace back. You know, it's like how do we limit that? It's going to be really important.

Speaker D: Yeah, for sure. I think the viral story of the week, which we'd have to go double check the source but you know, the whole Starbucks rolling out AI to go count milk jugs in the stores and it's like that wasn't, we didn't need to use an AI solution for that. And how like what is the fit for purpose solution? Uh, which again comes back to evaluating the work to be done, what type of solution is best and then weaving that back out into ultimately like what are we trying to solve for our customers. And one other point I would make too is that whole work around your addressable market, right, the tam, the total responsible market. What do we think we can have? What's the incremental market share? Where is the opportunity being really um, upfront with that and even leveraging AI to have even better data and insights, but then knowing where to place those bets and increasing those bets versus understanding, hey, this is a market that there's not really that much incremental share and it's probably not worth um, going down further modernization. So I think all those things become key in making the right decisions for where you drive Impact.

Speaker B: Yeah.

Speaker A: The only thing I was going to add on to that was also understanding maybe you don't need to use AI for every single little thing. Like you can still use, like we leverage, um, work from Fusion for a lot of integration between tools within the Adobe stack. Maybe an agent could replace some of that, but also that would be increasing the cost of your agents as well, whereas a Fusion scenario could do that just fine. Actually, as part of what we are building for this agentic AI operating model is we're assessing where does it make sense for the agent to do something where it needs to take large data sets, lots of context and summarize it versus something that Fusion can just take from here and put over here, or generate very easily and continue to use an integration platform to do those things instead of replacing it with an agent because there's no reason to do that.

Speaker C: Yeah, well then I add that I don't have any answer to that question. But I go back to where do our interests become so much more in the agency type idea? Right. I want a natural language, just go do this. Where Adobe would actually take that functionality, it would be done and workfront Fusion and actually disagintify it. Yeah, it's still being done. Again, I don't necessarily envision what that means, but I can just see a world where you're going to deconstruct some of the stuff that is built into our platforms today because the preferred user interface is going to be much more just natural language and expect things to get done. And I don't have. I know you know, this is your world. I have no idea what I'm even saying from a cost perspective what that means or how you would deconstruct that. But I got to believe the licensing model that's associated platforms is going to radically change. And there could be real advantages too. I mean it could be more, could make more money, it could be more on demand. I mean we all know how much we pay for stuff and we use a fraction of the elements that we want. So it is possible that you get to choose your adventure a lot more, which could be really interesting. Put a bunch of software engineers out creating custom development, but we already know that ship's already sailing. Right. So it's going to be interesting how it plays out.

Speaker B: Yeah, I think one of the things I want to talk about, and I think there's been hints of this discussed and in my head I'm kind of wrestling with, we kicked off this, talking about do I have too many platforms, do I have not enough platforms. Uh, Bethan, Aaron, if y' all could kind of help me shape my thoughts on this in the sense of when we look at platforms and how many platforms we have, it seems like the next natural thing in this kind of AI, uh, world is you kind of look at how many agents you have. And. And Bethan, you mentioned having the coordinator, um, and then these relationships and this integration between agents and platform. How do you look at those relationships between platforms and agents? Because obviously platforms are going to start coming up with their own agents. But do you have to look at that separately of, like, do I need platform and agent or agent or platform? I, uh, just. How do you look at that at a granular level when you're talking about orchestration?

Speaker A: And, uh, a lot of it goes back to the process. So what is the process that we need to accomplish? And where do we have tools that can support that? And where do we need either integrations or fusion, uh, I'm sorry, integrations or, um, AI intervention to help streamline and drive that process? So some of the things that we're designing now is. Well, we know that, you know, a lot of marketing, um, and creative groups that they work in workfront. So we want, whenever we put together a brief, we want those details to drop into workfront. So when people pick up their tasks to do their work, they can see all of those details. Now, you could argue, oh, an agent can summarize all that information for you, and it can maybe even help you draft stuff. And I think that's the directly, the direction we're going. But the way we're designing this, um, solution is leveraging the tools for what they do best. Instead of trying to rebuild that in an agent or custom build that, the tools already do those things, certain things pretty effectively. So leveraging the tools for their strengths and then looking at the process to see where do the agents overlay on that and then where do the humans overlay on it, uh, to make things really streamlined. So that's how we're taking the approach. And then there is a lot of conversation around an agent does this, and then how does it get over to the person? Is that an integration or is the agent doing it? And so that goes back to the. The agent versus an integration and how. And sometimes you have both, sometimes you have both in the mix, um, driving things. Now, over time, yes, that could become much more identified. It could all move to agents. Um, but then, like you said, if we end up having all these agents that are doing all of these things that could become quite costly in the long run. And then how do you make sure they're not overlapping? So we're proposing also creating like, an agent registry that would have each agent kind of registered. We would know what they are doing, we know what their skill sets are, and we would have governance over the agents, kind of like you would have governance over the platforms. You also now have to have governance over all of these agents to make sure you're not duplicating efforts. And, um, also keeping track of the volume going through them. So you can also try to help control cost.

Speaker D: I was going to add to that, I think. Kevin, you mentioned something. I would view the world as like, there's two distinct parallel paths that I would say, if you're a intelligent SaaS, uh, platform provider, you'd be doing. One is in the shorter term, you look at a platform and as Bethann said, platforms bring relationships between data. They store the data, they've got security, they've got consent permissions, they have a baseline workflow, um, and they have like a UI that allows people to interact with that. Where they lack is that they don't necessarily intuitively know exactly how each user spends their day, what is their priority work, what are the questions being asked of their management team? I think the space that we are able to play in is how do we layer on top of that the agents and intelligence to go answer the questions of the business, answer the questions of what's needed, and then cascade that down into the workflows that the system does really well, and so allow the platform to know connections with the data, how to optimize the business rules and the workflows as created in that SaaS platform today. But we would kind of sit on top and say, hey, we know that you're trying to grow market share of this product, and so we're going to orchestrate all of this to go help you answer those questions for the report that's due to your boss and your boss's boss in the next one to two days. Right. The system's not set up to do that. We can work on a UI and an interface and a conversation and a set of agents. I can go answer that question. Um, similar thing for, for sales. Right. A CRM is mainly a front end to a relational database, but it's not working deals.

Speaker A: Right.

Speaker D: And sales is thinking through how do I optimize deals? And so how do we go create a deal view of how you can move that deal through, um, end to end, and remove the friction points to Go drive that. So I think that's the space that in the short term we'd want to do and work with the platforms. But then in parallel, my hope and my assumption, Kevin, is that there's other platform providers that are rethinking. How would we remove the friction to just go and create that custom workflow and allow a user to say, this is how I spend my day and they go and rebuild those data connections in an intuitive pane of glass that they can go, go make. I don't think the platforms are ready for that, but I think that would be my assumption, something that they're working to do in a more dynamic basis. But right now I think there's a lot of room for opportunity in the next one to two years to go optimize this other thing. And the lessons learned from building that UI layer at the top that intercedes with the workflows will inform probably the better, like fully dynamic experience that people are hoping that they get. Bracing.

Speaker C: Yeah, that, that brings up. That's a really good, interesting perspective and probably one that I hadn't thought a ton about in the context of this because I think I'm probably more naturally in the camp of like, you know, where do platforms exist right. In the future? And again, just to be very blunt, and I think that's an interesting element of like, we don't know the answer to that question. There's definitely competitive advantages and different elements, but there is definitely a path today where you can really continue to use and leverage because, you know, I mean, I hope that the platform companies and are asking that question here and that you're thinking about. But uh, we absolutely know that companies are asking that question. It goes back to even Bethann, what you talked about at the, at the end. Oh, it's like, you know, it's really around the data and it's really who rings fences the data the best. Is that going to be within a platform and they're going to be able to create guardrails that allows you and forces you to use that. Is it going to be, you know, the data bricks of the that tier layer? Is it going to be, you know, different elements of that. It's going to be companies. Right. And I think that's the million dollar question for me is really around, you know, I, I fully understand advising our companies or, you know, running my own company in the sense that we're gonna have to pay for these services and we have to pay for the functionality. But you know, hopefully, like, I would love to have A vision where you just described it, where I can just have, you know, I can have all kinds of different platforms, I can have all kinds of data, different locations. We can run that, that query, we can run that process and not have to be going through these stage gates or toll gates of sorts between different elements. And I don't know, I mean, I think that's the problem. That's maybe that's the dream of what we're seeing right now is we can, you know, you know, I'm an old guy. I grew up coming up through SAP. I still have nightmares thinking about it, you know, and it's like, you know, and I think, it's like, I think. But you know, can we, can we work in a world that's efficient without that? And I, I don't know if we know the answer to those questions. I don't.

Speaker D: Yeah, and I don't think we've really touched on the slow. The laggard in this probably right, is regulations and accounting rules and public disclosures and there's certain amount of auditability and traceability that's needed. And so how do you mimic that in an agentic world where you have all those right checks and balances and there was nothing audited or creative in your audit trail for the last seven years that wasn't made up. Right. And so fixing like making sure that we're aligned in that then allows these other workloads that aren't as regulated to be, um, you know, modified more quickly. But then still that chart of accounts, um, we have, we have a way of demonstrating all the checks and balances that are necessary. So will be interesting to see.

Speaker B: We have roughly about 10 minutes left here and I want to make sure we kind of start focusing on kind of like what the road ahead looks like. Obviously you're talking about all these things that we'd like to see happen or we think might be happening. But um, before we do that, I just kind of have a question I want to pop to. But you both, Aaron and Bethan, of just of all these things that we're talking about in your eyes, and feel free to not use specific examples, but just in general, what, what are some of the mistakes that y' all are seeing out there in the market with these, these enterprise companies or even, you know, mid market size companies that, you know, hey, I've been seeing a lot of this, you know, maybe stray away from that or. I mean, we've talked about some of the issues, but just from your opinions, what, what, what are you seeing out there that is not clicking with some companies that are, are either trying to move too fast or, you know, give me all that AI you mentioned, Aaron, the Starbucks example of uh, counting the milk jugs. But, uh, maybe if there's some other examples that you can just think from, like an enterprise grade, uh, level that you could uh, shout out.

Speaker D: I've got one thing that comes top to mind. Uh, so I would just say a lot of what last year was, was I have these license tools. I need to roll them out as quickly as possible to my people and I want them just to start using them for different tasks to learn. And I think that that's not a bad thing. But I also believe that there was the promises made to executive teams and board teams of uh, really big, you know, whether it's efficiency gains or increased output or even like cost savings that they were putting out there. And what wasn't happening was going back to the people in the workflow side of getting people to say, hey, you traditionally owned this task of the process or these two tasks. Now we're wanting you to really own this entire set of capabilities and the value is going to come with you knowing what's coming up upstream and downstream, giving the right context to your tools and, and really generating an output that was more comprehensive and complete. That had taken three to four different teams coming together and now you can largely do that on their own. I saw teams starting to roll out the task optimizations and sort of got people anchored to that without looking at, okay, now we've actually got to rethink how the teams come together and how do we more quickly move through to the output that we're actually trying to deliver for our end clients or for our teams or for our leadership. And so I think that's been the. Maybe the pain is how much cost was incurred with just doing that rollout to then see limits of benefits. And then now you've got to kind of go back and re architect thinking of how people leverage these tools and even the goals are seeking to achieve. Bethann, what would you say?

Speaker A: Uh, similar, um, and not even. This has actually been happening for many years as long as I've been doing implementations. But the thing people and organizations tend to not invest enough time or energy in is change management. Um, we can develop the tools, the integrations, and even some of the process work. Not everyone does that either. But the process and change management tend to be things that are not emphasized enough. But in terms of really capturing up front, here's what it costs and the time it takes and what it costs to do this now, and if we switch to a new tool, what are the time and cost savings there? But then also making sure people are brought along with the process. Um, because of the proliferation of tools that we are seeing now, you know, people are having to use a new tool every other month or take a training on a new tool and they just get very frustrated. And so especially as a company announces a new, we're rolling out, you know, doing a whole new technology innovation, you know, people tend to just get very frightened or upset or just exhausted with those things. And so making sure that the, you know, people are brought into the process early on, that they're communicated with and that they have input into what these eventual solutions look like. That's where I've seen things be most successful. When organizations actually invest that amount of time and energy into change management and not enough organizations do. Um, one of the best implementations I ever worked on was they made a whole campaign around this technology implementation. They did a naming contest, they had logos submitted, people got really into it and the implementation was hugely successful. The other thing is the partnership between business and it was very good there, um, like the marketing. Org and it sometimes we see some, um, friction there and that's why you end up seeing marketing groups going out and getting all their own tools instead of aligning with what the organization has. So really if we can make sure that change management is done well and that the organizations are working together collaboratively instead of kind of having friction, uh, that would tend to be more successful in the long run.

Speaker B: It's amazing out of all the change people process technology, just as much change as there is, some things stay the same, right? I mean all these new tools. But at the end of the day it comes down to kind of those three, those three pillars.

Speaker C: Yeah, well, just it's work. It gets to me, it gets exciting. Not only for what we get to continue to go do work with our clients and also kind of the, you know, the classic line of like, you know, consulting's gonna die, their death is greatly exaggerated or whatever that line is, but like I think that there's going to be role for us to continue to work through that. There's going to be tremendous amount of change. You know, I'm intrigued. I mean there's going to be winners and losers and I don't know, you know, that that's, that's the reality of any type of innovation. And I think we're seeing that and you're seeing that with terms of the investments that are being made. And I think it's going to be a, uh, bit of a confusing world. Uh, just, you know, again, as we continue to think about the acceleration of these elements of that and you know, it's going to be shadow, it shadow agency elements like that are going to be really proliferated and how do we just do that? And uh, Aaron and Bethan, you both talked about regulation and regulatory environments, data security, privacy, those components are huge concerns. I can see how that might be a place to where the platforms actually have a leg up, where that could be. One of the elements that they can offer is that, hey, if you play in this ecosystem with this data and these elements of that, we can guarantee that in a way that's better than you just doing that with, uh, just a combination of different sources. But I don't think these problems are new. They're just happening at a faster pace and uh, require a greater degree of nimbleness. And then it goes back. I think Aaron, you said it really well earlier. There has to go back to a business reason. There's gotta be an outcome that's desired. What are we trying to accomplish and how do we continue to work through that. But it is absolutely fun times and it's an interesting element for, uh, what was once a client, what was once a vendor, what once was a partner, are all kind of blurring. Those lines are all blurring right now.

Speaker D: So yeah, I just got to say, Kevin, uh, I just like, I think the winners and the losers are those that can articulate the most meaningful vision that excites their employees, their team members and their customers. And the vision of, hey, do more with less is not a compelling one. Right. And so how do you talk through, hey, we have data now at our fingertips where we can surprise and delight our customers and personalize the experience. And they in now in ways that they didn't even know what's possible. Right. They'll go, how did you know that this is the car I wanted to buy in the landscape? And you've almost intuitively understood why I want to go buy that thing. And you've made it super easy for me. Right. Or something in the retail space. But I think that coming up with, hey, we can now do these things, let's go and grab the data, let's go act on the data, do the things we always wish we could, those that make that accessible to their full team, uh, make that real, uh, and can demonstrate, hey, this is like the actual moment we'll be able to Deliver will be great. And I do think too often it's just go do more with less, find a way to be efficient. And again, for any sort of person working on a team that can be deflating and almost um, and just make you tired from the get go, I think those that can put that vision together are going to win. And it's exciting to be able to partner with clients to do that.

Speaker A: And reframing it more as the agents can kind of help you reduce a lot of the manual like burdensome administrative things maybe that you were doing where it can take some insights, present them to you and you can kind of pick from there like what's the right way to go and so allowing um, you to be maybe a little more strategic instead of doing a lot of manual kind of administrative type work and people can do more of the things that potentially are uh, enriching for their job instead of burdensome for sure.

Speaker B: As we kind of get to our final thoughts, I would love to leave the group here, Kevin, Bethann and Aaron with a uh, uh, posing thought starter if you will, on uh, just where we see things going in the next 12 to 24 months. I know we've talked briefly and lightly about maybe some things that could happen in the next couple of months, next couple of years, but just some final thoughts on where we think things are headed and then just any sort of advice that you may want to offer up free of charge, uh, as we don't charge to listen to this podcast. And uh, I think Aaron, uh, will start with you and then we'll go to Bethann and then Kevin will have you uh, wrap us up.

Speaker D: Oh man. Uh, I think final thoughts would just be, we've got to be better first of articulating uh, a more defined vision of, with more powerful tools, what's the more powerful impact we can make for the business, uh, and for our customers. I think that's step number one. I think step number two is it's then really looking at our end to end workflows and determining how do we really like rethink this and optimize this, knowing that we have a team of agents that can do things very quickly. And so how do we start with a more strategic data driven point of view. Uh, and then how do we have uh, human in the loop at the right moments, but also learn how to educate and just like we would upskill and equip someone in a human sense, how we can do that with agents and so that whole agent onboarding plan, the enrichment, the how you empower and guide decisions and how you cascade that into org structure. In my mind where we're going is a whole set of frameworks around that and how do you um, templatize that and really even leverage technology to continually refine and uh, make that more accessible for people where that might be a harder leap. So I think that's going to be a big change. Um, I think this whole single pane of glass in terms of catering to how someone works for their, you know, their, what their day looks like will become something that we continue to play in and then work through. As we talked about the ecosystem and how platforms uh, support that and then potentially come up with an even more robust solution. But exciting times ahead. Like we said, people process technology, data, platforms, similar concepts, bringing them together, but it's all about the outcomes and the impact, um, and continue uh, to just keep that North Star ahead.

Speaker A: So what I'm seeing, I think over the next 12 to 24 months is it was when I was looking at this AI kind um, of model of level one is some agents are doing some things when you prompt it. Level two is, you know, and agents are just running and doing things and then you review it and then level three is, you know, the agents are doing things and then you're just reviewing it like, so there's like this progression. So I think what we're going to be seeing over the next 12 to 24 months is I think people are going to want to jump to like the highest level of a high maturity. But that's not going to be to their, to their benefit because really if you start to skip a bunch of steps, you could get so far ahead of your skis that if you try to go back and fix, could take enormous amounts of hours and dollars to do that. And so I think really understanding the people and the process and getting things put together in the tools appropriately getting the agents to start to do some things and then as they are proven, being able to allow them to do more, uh, and then allowing the people to kind of continue to review and do the more strategic activities while the agents take on more of the manual administrative um, things, um, I think just like don't skip ahead because if you skip so far ahead and then you run into a problem, it's very, going to be very difficult to correct if we've not gone through kind of the phases to make sure that the agents have been proven that the data that they are working against is sound. Um, so just making sure to kind of go through the progression instead of trying to skip ahead. And once again don't discount um, change management, making sure processes are well defined and that your underlying data foundation is uh, very well established and works well.

Speaker B: Amazing. Kevin, you want to bring us home?

Speaker C: Yeah, uh, I think to me that my thoughts continue to go through just my head is okay, it's the power of this present time, meaning that we've got the companies, individuals understand processes, they know the domain expertise. So right now there's a place where the next bit of time it's going to require, you know, really that building and these tools are going to really help accelerate. You know, my mind goes to what is two generations down, five years down the road, 10 years down the road. You don't have uh, individuals that have that company background, that knowledge of all that. You don't have your resources that grew up doing things a certain way that now are being all done really in automatic or rebundled ways. And how is that going to work? I think about um, a birthday for one of my daughters today. I'm thinking about, man, what is life going to look like for her? 5 years, 10 years, 15 years down the road, how do we help instill all this knowledge that is just something that's going to come from. You don't have to think anymore. It's going to be so easy to have access. The question is going to be uh, how do you teach someone how to actually think versus or ask the right question versus what's going to be available. And so part of me gets really excited about that. I don't worry about really the short term impact of how a company or how society is going to continue to evolve. But I do wonder what happens when um, all that kind of institutional knowledge can't be deconstructed. Similar going back to Adobe. Adobe built a whole entire ecosystem around taking processes and functions and excuse me, having things that were done manly or those steps are now being done through. So we've seen this before, so that part's not new, but it's going to be a new frontier.

Speaker B: He gets really emotional talking about the future. I should have, I should have not brought it up. That's my future.

Speaker C: Yeah, the future. Adobe, kids, birthdays, all that stuff just makes me crack up. Uh, but uh, yeah, but I just think it's going to be this, I'm kind of laughing internally as our classic send off between how do you have to curious and be bold and be curious? And I think that really comes to that. And so I think right now as the leader of companies. You know, I'm thinking about how do you create this environment that's going to allow you to continue to evolve very rapidly and continue to go back, solve those business problems, the creating elements of that. But also how do you have the right safeguards in place? Not even so much from security and data perspective, but just even understanding what's being done, you know, when things, so many things are being done without truly understanding the how. Just think about the simplicity, simplicity of showing a simple chat CBT prompt to a young child and explaining to them how much information comes back at their fingertips. Now amplify that out at 100x of what the power of that is and the speed that occurs. I mean we're, gonna, you're not gonna have any construct of what's even happening. And uh, and I think that's going to be both exhilarating and scary and uh, um. So how do you do that in a way that uh, is not controllable? Because I don't think anything is controllable in the space, but is actually done in a way that is um, you can, you can really understand what's going on and then take advantage of that to really drive whatever outcome you're trying to hope. So it is a really fun time. Don't um, know what that future looks like, but it's kind of really cool to think about that. And I think it's going to be really interesting right now as we're seeing companies try to envision how they fit, you know, that, you know, uh, fit into that ecosystem in the future. And there's a lot of unknowns in that and a lot of money is going to be spent trying to get the answer.

Speaker B: So 100 and you know, for anybody wondering what the future is going to look like, even though we don't know, you can tune into the Pure Signal podcast to uh, to find that out as we have the latest and greatest and all of those things. Bethan, Aaron, thank you all both so much for putting up with the technical difficulties, putting up with Kevin and I, um, really appreciate yalls time today.

Speaker D: Yeah, thank you guys.

Speaker A: Thank you. Thanks for having us.

Speaker B: Of course, for myself, Ryan, and for Kevin, this has been another edition of the Pure Signal podcast reminding you. And I'm going to say it because you just said it. I don't make you do it twice. To be bold and please, please, please stay curious.