AI won't save a broken organization

Speaker A: Your marketing organization and every other marketing organization had an opportunity to reimagine its organizational model when AI burst on the scene. Did you do it? On this episode of Conversations with Martek, we're talking to Ryan Warren, chief CRM officer at Razorfish, about the state of the Martech stack in the AI age. Using AI to improve the customer experience and more. Welcome to Conversations with Martech. Hey.

Speaker B: Yeah, I'm uh, Ryan Warren, chief, uh, CRM officer here at Razorfish. I've been with the agency for about five months now. Um, before that I had a 20 year career, uh, starting at exact Target and um, later Salesforce after that acquisition running all of their pre sales advisory and go to market strategy functions. So advisory practices, et cetera. And so at Razorfish, I think it's an interesting time where we're seeing a lot of flux in the market. We saw it a lot at Salesforce and felt like it was just an awesome opportunity to join Danny's leadership team and start shaping what the future of CRM is. And um, you know, it's, it's really, I think when you break it down, CRM is more of a way of doing things than a thing. And we um, have uh, the opportunity at Razorfish to essentially kind of design that future. And uh, so yeah, it's um, I think it's a, it's an interesting time. And um, and we, we really are excited about what we're uh, what we have in store and what the type of work that we're doing, the clients to be able to get them over that hump.

Speaker A: Yeah. Okay. So Salesforce, you mentioned, uh, maybe the best example of how CRM is where data and AI come together, um, because they made so much noise in that space. You wrote earlier this year in media posts that marketers blew an opportunity for a fresh start with AI. Can you talk about what you mean by that?

Speaker B: Yeah, I mean it's, it's a lot to unpack, but I think to just boil it down into the simplest concept I believe, and I saw this in my 20 years in software where there's so much innovation happening on the tech front and now we've got this uh, agentic future that's coming online. Personalization was a big topic before that, automation before that. Um, very easy for marketers to look at capabilities and reimagine what their organization would look like with those in their hands. But in reality what we felt, what I felt during my time in software was the organizational model really precedes the technology. And um, as a Result, when you're bringing all this tech on board, you know, this isn't a salesforce problem. This is a broad Martech, um, issue where technology is brought into an organization, into an existing flow of work and a way of doing things. And as marketers we've kind of missed out on the opportunity to reimagine how we want to do and work, uh, how we want to do the work in the future with those uh, capabilities in hand. And so as a result, right, there's a lot of disconnected pieces and marketers are struggling to figure out how to rationalize it, but more importantly how to turn that into the outcomes that they really saw in the beginning as they started to invest in the technology.

Speaker A: This episode of Conversations with Martech is brought to you by Semrush. Meet your unfair advantage. A new Semrush uniting search and AI discovery into one powerful marketing growth engine. Delivering richer data, more keywords, deeper AI insights, and prioritize next steps. With Semrush, you can turn brand visibility into coordinated action. Get a 14 day free trial and a discount code. Visit martech.org SEMrush Coupon or scan the QR code. So let's stick with this idea of the Martech stack and like you said, it's been around for more than two decades now, uh, going back to Exact Target and a lot of the early players. What does a Martech stack look like in an AI world? We've got all of these tools that have AI built into them, in some cases sort of bolted on in a haphazard fashion. You've got people who are vibe coding tools and saying, oh, I can build that, I don't need to go and buy that. And then you got, you know, dozens of legacy apps inside marketing organizations that have been there for years. I, uh, feel like the average Martech stack is kind of a mess.

Speaker B: Yeah, I, you know, I wouldn't disagree. I mean, I, I don't think it needs to be as complicated as we sometimes make it. Right. I mean, I think you look at some of the emerging technology that's happening around the periphery of, let's say the traditional, um, ESPs, right? Just to take out that old, it's a bit of a deprecated term, but, um, everything around that has gotten infinitely more complex because there's point solutions trying to help solve and create efficiencies and in uh, the gaps that exist in the current landscape. And so it's really hard to navigate. I mean for us we're doing a lot of technology rationalization with our clients. I think an agency like Razorfish has a very unique perspective of the world because we're actually in that flow of work and doing the work. And so with a high, ah, amount of technology and architecture acumen, it allows for a firm like us to be able to make sense of it all. Um, to me, there's really only about 8 different technology domains that you really need to understand. You've got your big data clouds and your hyperscalers, and then you've got this emergence of TDP or customer data platforms and how that connects to your marketing system. Would you start looking at the system as a whole and thinking of it as an assembly line of data to the end experience and then figure out, okay, what are the bottlenecks in there in our working flow? The problem that I see marketing teams is there's a flow of work that goes across all those domains of tech. And the reality is there could be 13 different teams that are measured differently, that roll into different parts of the organization that touch that. And sometimes you just need to take a step back and look at the business architecture, the flow of work and the technology architecture together, be able to find the opportunities of harmony. And when you do, you start realizing maybe you don't need everything, you know, that you've invested in. And maybe you can start cutting costs or you can start creating efficiencies by just relying on a core set of technologies to do what you need to do. Um, so I think that we have a lot of opportunity there.

Speaker A: Is ownership still a big hurdle for Martech. Stacy, you mentioned like the data warehouse, the data lake, whatever IT is you're using. That was always an area where you know, who owns the data? Is it it, Is IT marketing? So who owns the system that holds and manages the data? And that's where I feel like a lot of people got really complex because it wasn't just all within marketing.

Speaker B: Well, no, it's not. And you know, there's a broader, I mean, you look at, let's say, uh, an organization like Databricks or Snowflake and the hyperscalers like AWS and Azure, I mean they're, they're being built for use cases that go way beyond marketing. And obviously that lives inside of an IT organization. But how that turns into a curated operational profile of an individual like marketing needs to have a vested stake and have the skill sets to be able to navigate and partner with their IT counterparts more effectively versus just be downstream from them. And so we're seeing kind of new roles emerge. And this whole new paradigm where data competency and data foundations is core to the language of marketing. And I think that that one word, the language, the marketers and data engineers haven't always spoken the same language. And so to land on that common lexicon between those two different parties is really important. And sometimes frankly, marketers need uh, an organization to help them with that translation layer and to help be their advocate. Right. And I think that's where the market starts to lean in on, um, you know, companies like ours to be able to help them.

Speaker A: So sticking with the topic of data, there's a lot of opportunities for AI to help a marketing organization manage their data, organize their data, clean their data. That's another area where I feel like people haven't really capitalized on, on that. And in some cases, you know, AI has worsened the data situation because of its, the way it amplifies things. Would you agree with that?

Speaker B: I, yeah, I think I would. I mean it's a very nuanced situation. Right. But you know, every organization has to wrangle, structure and model their data. And then a marketing organization has to take that in a curated state and turn it into audience action and goal. Right. It's that simple. The end of the day. What's interesting though is you think about all those different layers of technology and there's different states of data that live across that whole stream that it can get really, really unwieldy. And so how you take the signals that you're getting from the customer experience layer. The big reason why CRM, um, at Razorfish is actually all own channels, data technology, strategy and creative, ah, in one group because we believe that the future is all about connectivity versus you operating in old ways of working. And so I would agree with it. But in reality it's a job that's never done and you have to keep evolving over time because there's new data sets that come online all the time. We're implementing new use cases, things like that. So this is a run state for every business moving forward. And sometimes I think we're too, too quick to put the flag in the ground and claim victory as if we completed it.

Speaker A: Yeah, there's been so much that's changed over the last couple of years, obviously the continued emergence of AI being one of those things. But if we were to have this conversation again, say a year from now or two years from now, what do you think we'd be discussing?

Speaker B: First off, a year might be too long given the current state, don't you think? Yeah, I mean that's another recorder. You know, I'll tell you what I hope. I'll tell you what I hope, because it. We talk about this a lot inside of Razorfish in terms of like what is the future really going to be? And I think that the optics of the future probably are on a 12 month horizon and then after that they start getting a little bit more opaque. But I would like to look back at least with our clients and in the market that we actually were able to understand that AI is bigger than just say an agent inside of a platform. And we start to recognize that if we really want to maximize the potential of AI, we need to embed it into the flow of work that we do on a daily basis. But to do that effectively you have to understand how a process perspective what those pieces are and what you're trying to solve for. And that we've created in a year's time the foundation and early indicators of success in a way that can shape the market for the next five years. I think that we're in a very interesting time where we're kind of getting away from these ad hoc experiments of AI and we're moving into bigger architectural, organizational level strategies. And um, you know, in the next 12 months I would like to see some of those strategies start to actually show a yield and to be able to create inspiration for the laggards, to be able to catch up because um, the marbles are rolling down the hill and it's not possible to put it back in the bag at this point. So um, you know, we just have to keep on better. Better never. Best keep progressing every single day and driving towards the mission of um, tapping into this amazing technology.

Speaker A: I mentioned before that there's a lot of AI uh, that seems sort of bolted on. Right here's the application you've always used. Now here's the AI button or whatever you want to call it. And I've talked to vendors in spaces like uh, in digital advertising who were using machine learning like years ago and they're like the AI was transparent, like it was just part of the tool. You didn't know. Now everyone needs to, I feel they need to advertise that they're using AI. So they put the icons and the buttons and stuff and it, but it does, it takes away from the, it should just be part of the workflow.

Speaker B: Yeah, yeah, but I mean that's a really interesting thing you just said though. Like the buttons, I, I think buttons are going to start to disappear and you know, and um, uh, you know there's been some software announcements that some organizations are getting rid of login buttons. Right. I mean like we're going to start seeing that, but people have to get comfortable. You know, we just, for decades we were just used to the tactile feedback of pressing something and interfacing machine to M machine interfaces or human to machine interfaces that you're using audio input versus, you know, clicks and things like that. I think that there's a behavioral change part of that that we have to recognize that, you know, it's, it's changing the paradigm and not everyone's ready for that entirely. And, and so it's going to take us a little while as just like a human race to start feeling more comfortable in those worlds. Like I can't imagine setting up a journey automation through my Alexa device. Right. But I think that we're starting to see some of that level of sophistication possibility come online and um, we have to rethink the way that we work and rethink the way we skill our people.

Speaker A: As a result, it may finally be the end of the open office layout when everyone is speaking to their apps because you'll just have a room full of people talking and no one will understand anything.

Speaker B: Yeah, yeah, yeah. Well, I mean it's a really interesting time and you have to kind of think about the scenarios and possibilities of what could be. But you have to ground yourself in now. You have to be realistic of uh, what your workforce is capable of doing. What I think me, and I'll just kind of go ad hoc here, but I feel like AI is like not only the generative AI and the agentic portion, but just the machine learning and the predictive element, like all of it wrapped into one is giving us just a really unique opportunity in the space to be able to really crack through this cognitive overload that we see in customers. Because I think that what we don't talk about enough is the fact that the volume of emails and direct messaging as a result are going up 70% year over year. The average number of emails that a consumer opens in a day is like between 130 and 140. And so there's saturation happening and we're seeing performance decline as a result. So as, as brands and organizations are looking at AI, I would urge them to look at growth as a primary objective, to be able to implement that and to be able to move faster around getting towards personalization. Full funnel connection. These things we've been talking about in the martech space for 20 years, but haven't really seen happen. We're finally starting to see those scenarios happen in the marketplace, at least with our clients. Right. And um, so what it does is it allows you to be more efficient so you can work on the complex stuff. Because humans and AI. AI is not replacing human beings. It's about making human beings more efficient to work on, um, ah, higher complexity work that we haven't been able to do because we've been stuck in the whirlwind just trying to get a campaign out.

Speaker A: Right.

Speaker B: And to me that's what's making this era so exciting.

Speaker A: Yeah. I think a lot of the early accomplishments, for lack of a better word, that people talked about had to do with scale. And like you mentioned, we almost can't receive more emails and messages. Scale is not, Scale's not the answer. Right.

Speaker B: Yeah, yeah, my assistant would attest to that. I mean, I, I don't know, happen. Oh, don't open half of anything. I get, I just, it's just too much. I can't. I just, my brain starts to shut down. Right. And it's real, it's really real.

Speaker A: Yeah. One more thing I want to get your opinion on because it came up episode, um, we did with Stephen Quash, who's the CMO of pipedrive, um, about how we train our teams with AI. He thought there's so many tools available and so many tools sort of freely available. You got a Google Workspace account, you've got Gemini, you can go get a ChatGPT account. Um, we sort of tell these people this stuff is out there, but we don't do the handholding or the training that we would with like any other application we put in front of them. What are your thoughts on that?

Speaker B: Yeah, listen, I, I think that it's a leadership thing. Honestly. I, it's um, you know, for me, if we roll out a new offering or we roll out some new part of our go to market or capability that we're answering a market demand. It always has prompts along with it to make people better at doing it. And so I believe, to me it's not tools. You know, everyone has tools. It's. It's just making sure that your people um, have access to like the prompts to be able to teach them and to pull them up into that world. And that's just a leadership thing. Right? It's just, it's making sure it's part of your day to day activity and, and then there's a certain competency that comes with that over time in which you can start bringing more, more nuanced tool sets into what you're doing. And, um, you're approaching it from a position of authority and understanding versus letting the tool try to, um, teach you about what that is. Because every tool is going to be unique. And I think in the future, going back to that, what does it look like in a year? I think we're moving away from trying to look at tool by tool and look at the orchestration of all of them. And that requires a higher level of understanding and how this technology works.

Speaker A: All right. Ryan Warren is the Chief CRM Officer at Razorfish. Thanks for joining us on Conversations with Martech.

Speaker B: Thank you.

Speaker A: Mhm. Ra.