สำเนาบทสนทนา 🔗
cmo to cmo episode 3 v3 (1080p): Hi everyone, welcome to CMO to CMO, ROI or DI, where we talk openly and unfiltered about the business of marketing. My name is Marie Bahl and I'm the CMO of Uptempo. I've led marketing for the last 30 years, and I'm here to tell you we are in an existential fight for credibility. AI didn't just change the tools, it put the whole function on notice. If we can't explain our value in numbers and language the rest of the business can understand, they're going to find somebody who will. That's why I'm talking to CMOs around the world and asking how they're proving that value, how they plan to own the seat we've collectively failed to interpret effectively. Today I'm joined by Dariusz Paczuski of CLG Partners. Dariusz has spent nearly three decades navigating major technology shifts. He's held several marketing leadership positions with companies that include Franklin Covey, Yahoo, and Microsoft. Dariusz, welcome. Thanks for joining me. thank you. Great intro. Appreciate it. Happy Friday. And I have to say, we have what we have a company in common. I was part of the team that launched Yahoo, I think I shared before, back in the mid 90s. I came in quite a bit later than that, and we did not have the luxury of being on the rocket ship that you were probably on creating. It was actually it was pretty funny. I was working for Niehaus Ryan Haller, hi Ed. and they were taking Yahoo for all stock. it was somewhere in the mid nineties. I'm gonna say like ninety six, ninety seven. and maybe it was early in that, maybe it's ninety five. I'm starting to lose track of time, Dariusz. and it was a lot of fun, I'll have to say. It was a it was a lot of fun. Well, I can tell you that was one of my first transformative internet experiences. I was in GE's financial management program, a rotational program for people who wanted to be in finance and business. And I was a marketing guy, so I was definitely getting my butt kicked in the program. But it was a rotational program, and one of the rotations I had was at NBC. and some someone in the IT leadership development program, he's like, Come over here, let me show you something. And he walks me over to this massive IBM, you know, beijy terminal and loads a browser, first of all, Netscape, which is where I ended up going shortly after that, and types in these letters, www.yahoo.com, and up shows up this director, and he starts clicking around, and my mind is just blown about, you know, what we're actually seeing here. Yeah. And I just I knew in that moment that that was something that I wanted to be a part of. And So, fast forward almost twenty years later, here I am at at Yahoo working on Yahoo Mail and Yahoo Search and the home page and all the products and all the brands. It was it was awesome. That's that's amazing. I know it's hard to believe, Dariusz, because we are both still so young and good looking that we were that we were there. Like, you know, as I was dialing the river side. Exactly. that we were there at the beginning. I mean, I remember those years so fondly, and here we are at another inflection point. and as as disruptive and cool as the internet was, this one, this change and this shift has the potential to be even more disruptive. and and more exciting in many ways, but daunting. it's one thing to unleash AI, I think, on the plup the public sector. To allow LLMs to crawl and interpret public information is quite another as enterprises start to think about how to do that internally and as marketers start to think about how to augment their effort. Yeah, no doubt about it. I I like I said in the in the beginning, it was quite, quite different. And that browser that we were talking about, it came with preloaded folders. and one of those folders was called It had different channels in it and I managed that program. And what was interesting was, you know, this is the early internet, and that was a cost per click program. We basically had partners that pre-filled the sports and news and finance folders, but each of them paid on a per click pay-per-click basis. So this is very early on internet where business models and then tracking as a marketer, you kind of see like, well, is this thing working or not? with very direct impact on revenue. So I think that for me that was one of the formative experiences that I had was how do you make sure that whatever product or program you're putting out there as a marketer, how do you keep track of things and what's working and what's not working? And so that's always been, you know, true quest of mine. I mean, so I've been in technology most of my career, worked in lots of different verticals and obviously quantitative metrics are high and the tech category. And I think marketing as a whole, as I alluded to in the intro, has struggled to effectively communicate the value of the effort that we put out there. I always I always laugh when you hear, I think I told you this story. I was working at CU and there was a sales leader who came in and spoke to my 51 students about the value of sales and she alluded to the fact she said the the greatest thing about being in sales is that we're the only revenue driver of the business. and I almost passed out. And then I said, I'm pretty sure we get the ball at the one and hand it off to you at the twenty and you guys fumble about forty percent of the end zone. time at percent of the time in the end zone. And and the problem though is that we're very ineffective at laying claim to that first 80 yards. we've got certain things that are easy to measure and demand r in kind of demand marketing, demand generation. media has been measured for quite a long time, but we have models like media mixed modeling and attribution models and the like, but they're all just one step short of fully expressing the value of marketing. And all of them have to be collated together to in order to really interpret or understand the value of the entire effort. And what
Load full transcript
happens is we end up having raids on our dollars mid-year. I'm sure you've experienced this where in July the CFO comes along and says, How much of this is uncommitted? And you say, But wait a minute, it has value. And he says, I don't care, it's uncommitted, I'm gonna take it, because we don't have a way for them, for him to sign up for an impediment to the forecast. and I was talking to you a little bit earlier about the notion that if we are truly running marketing as a business. Then should we be treating the marketing programs that we that we build as individual products that serve the function of a business that takes on an investment every year, you know, whether that's a hundred million dollars or five hundred million dollars. And that makes the raid a little bit more expensive when the CFO comes in the middle of July and he has to sign up for a forecast shift. Yeah, and these businesses are so different too, right? If you're talking about product led growth. Then those businesses, everything is pretty damn trackable. you know, you put a dollar in and you kind of, you know, really can analyze where all the clicks are coming from, what's happening on the website, all the way down to the purchase and the revenue, and then you start calculating LTVs and whatnot. It's the sales-led growth motions where things get a little fuzzier, where you're driving leads and you're counting the leads, but you know. What role did marketing play in generating that lead to begin with? What role did sales play in generating that lead to begin with? And of course it's never linear, ever. Right. Ever. So that's super challenging. And then if you take a further step back and you start looking at the brand spend, I think what you know the the way I I just generally approach it is you start with the end in mind, you start with the revenue that's being generated, and then you try to work your way backwards. And you do it in a way where you're not trying to necessarily defend your marketing investment. You're trying to explain it. and you're trying to do it in partnership with sales. and so that's been one of the hallmarks of my career, and what I've always tried to do is are is the CRO and I are we aligned on the goals that matter, the metrics that matter, and then you loop in the CFO of course. And now together the three of you have a can have a logical, slightly less emotional conversation about like who did what and what who contributed what to each program and or to the revenue rather. And in marketing, like you said, I do like your the approach that you're evangelizing here, which is take a program that you're running and and that is part of the total cost. And so what is the total RI of marketing? but I think it starts with you have to understand revenue. what's where the revenue is coming from and what's influencing the revenue. And then you can start peeling back the onion on which of the marketing programs contribute to revenue. And then you're probably gonna l be left with, you know, an 80-20 where 80% you can kind of figure out this is working, this is not working, but that 20% often is the elusive part. And and so that's why you never want to just talk about the 20%. You want to talk about the whole program and is your marketing investment actually performing And is the performance improving, right? Like you can have a flat investment in marketing and revenue can grow, that is a good sign. That could mean sales is doing way better, or it could mean marketing is performing much better. But at least you start with a common baseline of revenue pipeline and then working your way back to the investments that you're making. There are so many complexities involved in that. And if one if if AI could do one thing for us, it's actually tackle. That complexity, right? So when you talk about all of the data that's required to be able to interpret whether or not each of the functions inside of marketing is producing enough of whatever it's supposed to produce in order to drive the outcome that is revenue, that becomes an almost impossible question to answer unless you've got a unified data stack and unless you've got analytics that's robustly running across that backbone and then ultimately, I think in in this day and age, AI agents that could surface insights for you. Almost no marketing department that I know of today has that. We've got disjointed, disconnected systems, siloed insights, proxy indicators that when we try to marry them together, we're, you know, we're doing that with spreadsheets or we're handing it off to PhDs over in MMM land to run science projects that they come back to us a quarter later. this is part of kind of what we're tackling at Uptempo. And we're seeing that this inflection point with AI actually could be the first time that that marketing gets the opportunity to bring that big data set, those disparate sets together, reconcile those taxonomies, and begin to start to report out insights week by week. We never even knew we could report out because we haven't had that at our fingertips. What are you thinking about from a tech stack perspective right now when you're making recommendations to your customers? Yeah, no, I think first of all, just the acknowledgement that everyone's on on a different stage of this journey. and some companies are further along than others. at at Yahoo we were fortunate enough where we were we had such a sophisticated tracking and measurement system in place, even though we had some silos, the data was there. And we could actually track performance of individual creatives down to the pixel, down to the specific app that they were driving in. So we knew exactly this dollar pushed this creative in this mobile app store, Android iOS, to generate this download and this user and with this revenue profile. Super spoiled. Right. You're you're build you're talking about a t you know, a proven technology company where data and is part of the DNA, search and
algorithms is a part of the DNA, and so we have really sophisticated data scientists and BI people to help us out. And then, you know, I've been at fifteen-year-old companies, B2B companies, that you would think for fifteen years they've been around, they would have had the latest and greatest and all the tracking capabilities and data sets and all on a standard platform. And they're still spending 80% of their marketing dollars on events and field marketing with zero trackability. so I think you know, one of the first steps that you know I always start with is is A one. in my seven A's that you know we can get into another time, but it's basically audience. Who's your customer today? Who should it be tomorrow? How do we know our customer by heart? And what Tools do we need to put in place to make sure we capture that? That can be the qualitative side, but more and more it's gonna be the quantitative side. To make sure that we can c capture every single step of the customer journey and be able to associate it to either a marketing action that took place or a sales action that took place. So let me take a step back, right? Because I've been an enterprise software my entire career. And so I kind of see this as a personal failing in a certain way. But I'm gonna challenge a little bit what you just said, not not challenge it, but get you to think beyond it. So you're you're not working for Yahoo now. You're working for XYZ gigantic brand that invests, you know. 500 million in marketing a year and is served by Meta and Yahoo and Google and you know, retailers and Walmart and you know, all of these people who purport to give you excellent data that's all designed to do what? To get to get you to buy more of kind of what you're purchasing from there. How do you, as a brand whose personal responsibility is actually to drive market acceptance and build a market for your product? how did we get to the point where we're asking those brand leaders in technology as a as a technologist? How do we get to the point where we're asking them to be systems integrators? Where we're putting the responsibility on knitting and marrying those disparate data sets together and making sense of it. We have not done a good job of that. I think that enterprise software has failed marketing in a lot of ways. And if you tried today, for example, to dump all those data set data sets into a snowflake instance and run ChatGPT on top of it, you're gonna get as much garbage out as you put in because we have not reconciled each of those outputs in a way that is marketing ready, marketing understandable, that it that it, you know, reconciles the different time zones and time horizons and you know. Geographies, global, MDAs, extra all the all of the the metrics that we track. Yeah, I mean well to be fair, you know, I I'm working with smaller companies these days. you know, Yahoo Verizon Media was a seven point five billion dollar revenue company. the companies I tend to sp specialize in
these days is in the you know fifty to five hundred million and a couple of hundred million. And and so just to set expectations, you know, I find that those those companies they're they have meaningful problems, but they're a little easier to solve because they don't have while they have meaningful marketing budgets. I like focus. I don't try every single channel that's available to me. It's not Google and Bing and Amazon and LinkedIn and and and it it tends to be, you know, very, very narrow and focused. And if once we have fully maximized the spend right on one platform that is working for us, then we think about the next one. and so and and so the I don't end up with those crazy complex problems that you're talking about, you know, y if you're if you're the CMO of Verizon, you're basically spending money on everything because your budget is so big. You're talking about a half a billion dollar marketing budget or more, right? And you're putting those dollars in the highest ROI, but it's like y the you know, how do you fill a jar? you know, you start with the big rocks and then you put in the pebbles and then you put in the sand and then you put in the water. And at some point it spills over. And that spillover is basically where you shouldn't be spending money anymore. but you know, I tend to focus on the big rocks and then the pebbles and then the sand and then the water and all the marketing that I do so that I'm able to control the spend, control the strategy, know what I'm doing with a little bit more confidence. But yeah, the larger you get, the more complex all that gets. And you know you're talking about stitching together the data from all the platforms that you're talking about. I am surprised that some of these large platform providers aren't making that stuff easier for us. but but AI does is the is the promised land, right? It is incredible how much complexity you can upload into Claude or or chat or dots or pick your favorite and and how much with the right prompt. So you have to kind of understand what are you actually trying to get out of it. So even even if the data in might be, you know, a lot of garbage I think with right questions and prompts and knowing what you're looking for, you can probably get at at le worst case scenario, directional data. Sure. So let's so let's talk about that a little bit, right? So if we're talking about you have eight, you know, one to eight billion dollars in marketing spend. And by the way, that that percentage of revenue hasn't changed for the last 50 years. Why has it not changed for the last 50 years? Because if it's not broken, don't fix it and we can't figure out how to report on it. It's a big data problem. I sa I sat around with a bunch of CMOs who have this problem and they just kind of looked at me exasperated and were like, yeah, that, you know, ROI for us is just a bridge too far. And I thought, you know, I'm pretty sure we've decoded the human genome. I'm not, I know, I don't understand how this is a bridge too far. It's just an enterprise big data problem that we've solved. for many, many other industries. So yes, AI is the holy grail, but think about it this way. AI chatbots and LLMs have been trained to Please you basically, right? And and and hallucinate in the public sphere. That's not something that we can carry over. And so there's two things. One is information architecture has to be robustly governed and reconciled, right? So we've spent the last 30 years, our entire careers, talking about how enterprise software is enterprise quality because of the governance that we have on it, right? It's not just because it's really available, reliable, secure, you know, scalable, all the abilities. It's because you're able to show certain information to your intern and certain information to your CEO and ensure that none of that gets cross-pollinated. There's also a lot of, you know, kind of healthy taxonomies and large understanding. So there's tons of ERP systems and other that that you know deal with this heavy, laborious enterprise class requirement. Now we're saying, okay, marketing, do more with less. So that for the first time, we're gonna ask you to reduce that number. So we've you know you got between eight and fifteen percent on a regular annual basis. We now want you to reduce that. And you're gonna do it with AI. So let's throw bots at systems that you don't yet have reconciled that aren't fully governed and and bots that aren't trained by the way to understand how to act functionally inside a marketing department. You know The the reaction from marketers so far today has been like, hold on, we'll just point it at our content because we know it can't do a lot of damage there, right? But we know it can do damage to data. To me, th data is the holy grail. Getting clear on that governance and allowing AI to handle the the boundless complexity that we deal with, that's gonna be the holy grail, but we've gotta do it in a reasonable way. Yeah, I couldn't agree more. And you know Again, where is that data? How is that data structured? And do you even have access to it? And you know, when you're trying to stitch together a bunch of things. I'm not gonna name any names, but there are plenty of companies recently that I've worked with that had no customer data platform. It's crazy. The data was not accessible in a single place. So you start with what you got and you take one step at a time and you try to figure out, well, we can isolate the effects of this one channel. Great, now let's add the next one and the next one. But it, you know, there's a lot of plumbing that you need to spend time on as a CMO. And it's challenging because you know the the tenure of a CMO is challenging. I'm not gonna say it's short necessarily, but you know, you don't have a lot of time to prove your value as a CMO. And so, you know, if the first conversation you're starting to have with the CFO is like, where's my data? That's a tough
conversation because they're looking for like, well, where's my ROI to your point? Well, and my other favorite subject is that, you know, you say, Where's my data to the CFO? He points at the CTO. And the CTO says, Yeah, yeah, I can build that. You know, we are gonna go through one more cycle of IT departments thinking they're gonna build the next class of enterprise systems, right? I feel like every 10 years or so they raise their hands and they're like, we could build that. You know, we've been there, we've crossed that that chasm before. That it's never successful. And and eventually, you know, it's not just about building it, it's about maintaining it. It's about ensuring it. It's about making sure that what that data that what's being reported back hasn't been hallucinated. So it it it scares me a little bit that IT departments are raising their hands and saying, Hey, just you know, snowflake instance in a chat bot and you're good to go. I was I was born in Poland, I grew up in Norway, and I now live in America. And what keeps me sane as a person and as a CMO is my Polish pessimism, my Norwegian realism, and my American optimism. And and I think that's the only perspective that you can have on data these days, because it will never be perfect. Yep. It'll never be perfect, but we've gotta get to at least par, right? Nobody would say that your ERP system is, you know, knock knock. perfect or your CRM system is perfect, but they at least have a place to point to and gather data and and and you you can work on it, right? You can insist that, you know, the latest 900 fields that you added to Salesforce. I'm pretty sure you can take that thing. I think I'm pretty sure you can use it to go to the moon. I'm not sure that salespeople can use it yet though, but there's ways at least to to go after the problem. But for marketing, we don't have that, right? We've been given you know marketing automation tools and email tools and work management tools. Those are all ancillary to the ROI steel thread. The steel thread for ROI is what is my budget? Where am I planning to spend those dollars? How did I spend them? How did they perform? And back. And for sure. And by the way, all that stuff lives in Excel. Yeah, exactly. Totally does. It lives in Excel. You look at it a quarter. you know, say somebody says, Hey, how did those campaigns run? You're like, hold please, let me get back to you in three months when I have, you know, my marketing ops and MMM teams crunch some data. It's 2026 and AI is knocking on the door. And I just think, you know, I've seen a lot of articles out there. I know if you've seen them, Dariusz, where it says the CFO is gonna become the Chief Performance Officer, right? basically because we can't seem to report out, it's our fault, right? We can't report. It's it's because we're not financially savvy. I think we should, by the way, allow this because I would love to see CFOs running my I was gonna say I say I say we should do it, right? I see no downside of that. 100%. A year. Let's see how far that train gets driven off the the track. But but in all honesty, like we should be standing up and laying claim and saying, Listen, we we can't function. with you know 18 point solutions and an MMM and and an attribution model that live in left field. We've got to get this consolidated view and and it's not going to be built by the IT department who's running your you know Wi-Fi through the ceiling. that those aren't the guys who are gonna stand up your enterprise solution. Yeah I mean it's w what what is the what is enough enough when it comes to the CMO job and responsibilities, right? I stand fully by our mandate to drive to create and keep customers, be held accountable for the ROI on that, the investment that we are given, being good stewards of that investment, and thinking about the entire customer journey from brand to demand and the product that delivers on it, and playing a role in that. But yeah. To to tack on now we have to become experts in in you know enterprise infrastructure and data platforms, I think that kind of stretches any human a little bit too far. But we have to have the clarity about we don't have to own the production of it, but we have to own the outcome of it and that it gets And you know, and I do find myself evangelizing a lot. This is something that we need to do not just for marketing, but for the entire company and especially for revenue and marketing. And so let's work on it. Yeah. Systems integrator is not the thing that I think we all signed up for when we decided to become marketers. We weren't like, and I wanna talk about tech stacks and you know, integrations and you know ta common taxonomies, right? Somehow we've gotten off the path. And I and I honestly think we've come by this honestly. We're great stewards of our of our investments. In fact, most marketers come in under budget on an annual basis. I wish we would spend it up to the line because that would be working on our behalf. But we're great stewards of our money and we're also unbelievable apologists, right? And I think it comes from the fact that marketers came from a service or service organization mentality. And we've carried that over into these CMO roles inside the business. And I think we have to do a little less serving and a little more standing firm and demanding the things that that are gonna allow us to do our jobs and allow and you know make third parties deliver that. so you know ask your vendors, you know, how Do will will you be able to reconcile all of my data? Can you align it with common taxonomies? Are you marching towards an ROI? ask them. And if they haven't delivered it, by the way, for the last 30 years, they aren't gonna all of a sudden show up tomorrow and offer it. If they haven't been delivering it, they're not going to try to drive those solutions for you. Dariusz, thank you so much for joining me today. it was a pleasure to talk to you. I love your life philosophy. I might adopt some of that, although I have none of that heritage. thanks so much for joining me and I hope to talk to you again soon. Love it, love the energy. Thanks for all the questions. Great conversation.