The Truth About Marketing Analytics Costing You Millions

[00:00:00] The main thing that I've always challenged my upper funnel marketers with is you have to show me one down funnel metric that is moving from this, right? Like you don't have to show me the direct lift in revenue, but show me that more people are searching. intro: Welcome to Contrary to Popular Opinion, a podcast hosted by Bo Jee Digital. In each episode, we bring you provocative topics and novel perspectives around leadership, business, and marketing that defy the norms, because exponential growth demands unconventional thinking and visionary action. So open your mind and let's get into this week's episode.

Host Hey everyone. Kel McGuire with OU Day Digital. Excited to talk to today's guest.

Host 2 Hello, happy Monday. It's Todd Kelly's partner here. and we've got a guest today, so you don't have to just listen to Kelly and Todd talk. We've got Jeff Settleman from Funnel Analytics and Jeff, I'll let you do a little bit of an intro, but we like to have [00:01:00] people on that have a lot to say about something.

Host 2 and Jeff has got some cool stuff to say about measurement. In analytics in his time at funnel and more than a decade of experience from companies like Venmo and, a whole bunch of other companies where you've been, heavily involved in measurement. We are a funnel customer for the record, we have been using the funnel platform product for many, many years to ingest and structure all of our data for all of our data and analytics, outputs.

Host 2 and so, we thought who better to have come on and talk about, The myths and the beliefs about data and analytics. Then you, Jeff. so welcome.

Guest Yeah, thanks for having me on. funnel.io is the company, we're kind of the backbone of a lot of marketing analytics stack. So excited to be here, excited to talk to you guys in podcast form rather than in a, in a work call.

Guest So,

Host 2 yeah. so do me a favor, just to start out, it, sort of explain what is. Funnel, what's the core product? How's it [00:02:00] used? just give a brief explanation of it.

Guest Yeah, so Funnel is what we'd call, marketing Intelligence Suite or a Data Hub Suite. And it's essentially about taking all the various places that you do marketing.

Guest And combining that into one easy to use table so that everyone in your company, whether it be an agency or a brand, can point to a single source of truth of what your marketing has done and what it's doing. past life, I was, when I was on the brand side, we were looking at vendors for this and. my manager at the time, he said, we don't know where we were spending yesterday, then we'll never get to the level of visibility and performance we want to get to.

Guest So funnel helps brands kind of answer that question. Brands and agencies, where did I spend yesterday? By campaign and by creative. And it always boggles my mind how tough that question can be for a lot of people to answer. So we're trying to [00:03:00] make it easier for them.

Host 2 it is a tough question.

Host 2 So as an agency, our job is constantly on the line to prove, what we did. And so, you know, the next dollar we spend or spent, we're constantly on the line to have to prove it. and we've built a pretty sophisticated tech stack around analytics funnel, again, is the backbone for that. but there's a lot of things that get done wrong relative to this particular.

Host 2 Space or there's perhaps a lot of beliefs that exist, out there. So I kind of wanna start with, you know, where are things a little sideways in your mind? And, and you made a comment earlier as we were prepping for this, that there are certain things that turning your hair gray prematurely, um, that are keeping you up at night that you, you get from clients.

Host 2 And so I kind of wanna start there. and we'll sort of unravel things, from that point, because I think. According to our podcast theme, contrary to popular opinion, I I do wanna focus on where things aren't right and how can we help people think about getting things right? [00:04:00] So what are the top few things where, you know, the idea of advanced measurement is either being over hyped or under hyped or structured improperly so that incrementality or mm, MM isn't possible.

Host 2 what's sort of graying your hair?

Guest Yeah, well I think it comes from, there was probably a time in this universe where you could know everyone, whoever was exposed to your ad across a variety of platforms. And when I tell this story it was, it was before Harry and Meghan got engaged. Like even at that point, all these systems were broken.

Guest So there was a time when this sort of, you know, user level cross device tracking. Impression level tracking, that was all feasible, but that's gotten chopped away so much since GDPR, since you know when every app asks you to track or not. Each of those kind of advances have chopped away our ability to see that.

Guest [00:05:00] But the question that we get a lot is still like, Hey, I wanna see that complete user journey. I want to know every impression. I want to, I want to know what the optimal path is for someone. And Just having to explain to people that that's no longer feasible, right? Like you can never, you cannot see, you can no longer accurately track every single touchpoint someone has had.

Guest So for better or worse, you have to move to methods which are, which aren't as deterministic for your measurement, which has downsides, but it also has a lot of benefits

Host to it. Yeah, I mean, that, that's really, I think the, the area that I operate so much in, and what I struggle and, and some of the. I'll say arguments that I'll get in with Todd is, you know, I hear so much of the, the hype and the promise, you know, it's more and more with ai, but for as long as I can remember, I've, I've seen an endless number of the next miracle attribution tool that's come and gone over, time just because it's so hard to live up to it.

Host Yeah. But so much of it is based off [00:06:00] of what I call sort of this idealistic state, right? There's this theory of if everything is perfect. There's this amazing thing that we can do. Yep. And it's easy to sell that promise and that belief, and everyone wants to believe in this amazing outcome because everyone wants to solve the problem.

Host But almost universally, my experience is when you actually go to try and use one of these tools or. There's so many tripping points that come up so fast, and so much of it is data access and organization and history and all these things that quickly make it fall apart and really derail it from getting to that outcome so fast.

Host And I. This is something that I see coming so much stronger when you're talking about this big promise of AI and how much it needs, the visibility of what's happening and what are the outcomes in order to really draw these correct conclusions or it can go off the rails really quickly. So before I go too far down that rabbit hole in my own thinking, like what is your visibility and how important and how hard it [00:07:00] is to have an ideal state in.

Host Size of, of data availability, history, cleanliness, accuracy. And then the last thing I'll add in there is juxtapose the difference between a big company, but more importantly maybe a smaller, hungrier challenger company that may even have a harder time even possibly getting to a minimum viable data set up to do any of these things.

Guest Yeah, good question. you know, I think one of the, the problem. With marketing and or marketing measurement, I should say, is that idealistic world. 'cause you know, if I, I step outside right now with a thermometer, I'm in Boston, it'll tell me that it's 36 degrees on the dot. Right? With any marketing measurement system you get when you go and measure that, you're still gonna get a result with this wide band of uncertainty, right?

Guest Where. I'm gonna tell you, Hey, I've got a a great experiment result for you. Your [00:08:00] ROAS is between 50 cents and $3, right? And it's like, okay, great. If it's on the bottom end of that interval, I'm gonna cut all my spend. If it's at the top, I can, I can double what I'm doing. So the unfortunate thing is like with the way marketing data is structured today, where it's, it's aggregated, it's high level, we kind of have to deal with these.

Guest Wide ranges of uncertainty and just stack a bunch of estimates from different tools. And if you got six tools telling you the ROAS is closer to three, then you can feel better about that three than, you know, six tools telling you it's on the 50 cent side. So it's a more of the merrier thing, a good collection of uncertain guesses is better than just one.

Guest The challenger point. That's an interesting thing. I think we see actually a lot of challenges with big brands who are so established in what they're doing and kind of have these longstanding relationships with places where they're doing tv. It can get really tough for them to get all their [00:09:00] data together in a way that makes it actionable.

Guest We're working with, Someone today sent over, you know, we've got a great beat on what they're doing. Have their spend down to like the hour they dump us A PDF with a bunch of TV spend from the last year. Broken out by week, like somewhat normalized, somewhat not. So, you know, if they want to get a read on that, it's gonna be tough because it's not at the same data quality as everything else.

Guest So.

Host Yeah. So I mean, that, that's kind of what I see, right, is it's a lot of the challenge of understanding what is realistic, what can you get access to, how do you think ahead and proactively get creative in ways to fill in those gaps? Right? And, and you mentioned a bunch of the things that create these gaps, right?

Host We have the loss of data from Apple devices. We have the loss of, you know, UTM stripping. We have all of these, you know, gdpr and there's some places that opt in and opt out and all of these things. And so if we accept that. There can't be this perfect set of data reality. it really prompts the [00:10:00] creative nature of how do you fill in those gaps?

Host How do you look at correlation models? And then honestly, one of the biggest things that kind of you said that hits me is around the, what is my row as Right? And then there's, yeah. The context of that question, right? One of the biggest challenges that we run into is these big platforms that we use to access and spend dollars through.

Host They all like to grade their own homework, and they're very, very good at giving themselves credit and very good at, creating formulations that really inflate those numbers and say, spend more. But we need to look at things holistically, right? What are the touch points? And one of the things that Todd and I talk about all the time is.

Host Search last touch attribution is so critical in knowing really what the end of the road is, but you get sort of an overbalance in things like search, because of course if I didn't click on an ad and go there, I'm very rarely gonna remember the URL, right? I'm gonna go and search for it and I'm gonna get some credit there.

Host But then you have the question of, would I have ever conducted that search? Had I not seen the streaming CTV ad before, or [00:11:00] had I not scrolled past the Facebook ad, or had I not been to the site? And then you get into remarketing pools and all of these things. Yeah, so I think that you're trying to create that balance between the immediacy of the efficiency of spend on last touch versus the big picture contributions of how all these things work together.

Host That's where it highlights the importance of data structure and cleanliness and understanding and really the creative utilization. So I, yeah, I mean, it just sounds like you're sort of reinforcing a lot of these theories and thoughts that I have around all of these tools are great. AI is awesome, but if you don't spend, spend the time To get it ready to scrub it out, to evaluate it, to understand it. It's so easy to create this confirmation bias, right? If I go and look and say are my meta ads working for me? It's pretty easy for me to accidentally or inadvertently sort of craft the story that says yes, when if I take a step back and look at it more holistically and I have the right data set, it might be yes, but it could be better if you were to [00:12:00] create this mix of things and this journey paths and all of that goes down to access of data and all of that.

Host And so, I love the idea that Funnel iOS vision is. Sort of being agnostic in this aggregator of all this disparate information in one place so that you can ask the right questions and deploy the right things to have any chance of any of this stuff working.

Guest Yeah, I mean, I wonder who we have to blame for, uh, this last touch and ROAS quandary that we've gotten ourselves into.

Guest Maybe it's our, our finance partners across the globe. 'cause you think of that, that search example you just gave there. But, you know, I've been in a lot of meetings where. I'll have our, you know, our CFO kind of wondering, Hey, why, you know, why are you spending on video here? I see, you know, 10 last touch conversions, whereas branded search, you got 2000.

Guest Explain to me why your dollars look like that. And It's really tough for marketers to kind of explain some of that to people who aren't necessarily in the practice. So I think as marketers, we kind of need to [00:13:00] do a better job of explaining upfront what we're gonna use last touch attribution for and where it's not gonna be useful.

Guest there's kind of a soul to marketing, right? Like there's people who are gonna wanna buy the Jersey sponsorships. vibes based marketing, vibes based decision making isn't gonna go away, you know, despite AI and the proliferation of data that's available or not available. So kind of finding a way to balance that's gonna be tough for everybody.

Host 2 I, I will say in the last 90 days, most of the virtual conferences that I've been a part of, the general theme, and I think Kelly, you might have heard some of this at Digiday, was branding the idea of brand is coming back.

Host 2 Yeah. And that the idea of last click attribution and sort of this, you know, the finality of all that stuff is, I think a lot of the, the gaps in the. Things are starting to have matured enough where marketers are going, you know what? Brand still matters. Yeah. People want [00:14:00] to buy a brand that they fall in love with.

Host 2 They don't necessarily just wanna buy the latest thing on TikTok that says it solves a. Whatever problem they have or simply just introduces them to some new gadget that they didn't even know they needed or wanted and they just buy it and there's zero brand connection. But I did hear just for the sake of our own sanity and the soul of advertising that I've dedicated 30 years of my work career to brand is still a thing.

Host 2 maybe it's gonna make a comeback. Um,

Guest I saw a great quote today. Someone said, if you, you know, brand spending traditionally associated with these large, you know, large retailers, people have the money to do it. The quote was, you know, if you spend like a small brand optimizing for, bottom funnel, last click, you'll stay a small brand.

Guest But when you start. Thinking and spending like a big brand, then you can possibly make that rocket leap that everyone's really hoping to get. I wonder for you guys, you know, being on that side of often trying to find budget for brand, like how do you guys go into that [00:15:00] convo? Because I don't want this to be the my gray hair show, but you know.

Guest having that discussion with finance teams about, Hey, uh, remind me why you're spending on these tactics that aren't showing up in our scorecard. That's always a really challenging one.

Host 2 Kelly, you're the best answer us, but I'll just tell you point blank. It's a philosophy of ours that full funnel marketing is, is a core concept that we fundamentally and can statistically and factually back up we can show what happens with and without.

Host 2 To the bottom of the funnel and with the bottom of the funnel is always better.

Host Yeah. But I mean, it is, you know, we are held, feet to the fire and, and at least you know, building a case and showcasing kind of believable outcomes that show that journey and show that impact. Even if it's not one-to-one attribution seen ad by a product, we have to look at.

Host Contribution models of, you know, one of the things that we build so much in the programmatic space, because that's so viewed through, it's so, exposure based [00:16:00] that leads to the awareness and the outcome. But it can be, you know, a 30, 60, 90 day journey before it turns from awareness to conversion.

Host We do look at things like modeling contribution journeys, right? Where we'll measure the outcomes around search and, but look at when. A programmatic ad is in the pre journey versus not how much more likely is the conversion to happen or what's the contribution with meta? or we look at a lot of multi-touch attribution, but again, there's even gaps in that, right?

Host 'cause you can't see every device, you can't see every element in that. But as Todd said. We have to make that case. And I, and I love the scenario that you gave around thinking like a small brand and thinking like big brand is, there's generally a ceiling to that small brand mentality where you can ride the organic wave, you can ride the momentum, you can ride the the influencer, and at some point there's a marketing engine that has to be built.

Host And the fuel of that is generally not infinite. Put it dollar in, get more than a dollar out if you're just looking at the bottom funnel. So we have to create that case around. Why is someone believing? Why are [00:17:00] they journeying? What's their alternative purchasing power? What's the messaging? And then really paint that picture of.

Host If you spend here, you may not get your return right away, but here's how it leads to the next stage and fills this pool of data that we can now reengage and get some. So we at least have to paint the picture of creating belief in why those dollars can trickle down into outcomes. which can be difficult.

Host Right? Especially, I mean, we're getting further and further out of it, but I feel like marketers OR brands for the longest time, were trained around that meta. I almost call like the meta crack, right? You put in a dollar, you get $10 out, and the D two C experience was blown up out of that because it works so well for so long and it's just not the reality in the world that we live in anymore.

Host And you know, there's still a lot of legacy pain from that, that people still want those outcomes. Or I would say one of the hardest things we run to is when we talk to a brand. That has experienced explosive organic growth and hasn't had to fuel their own engine. It just worked. And they almost inevitably run into this cul-de-sac or [00:18:00] the ceiling where it's like, it's not working anymore.

Host Our, you know, our year over year doubling Yeah. Has stopped and now we need to look at it and trying to create the, here's what it's gonna cost to fuel your engine, to pay to create it. It's a very painful transition for these brands of like, I had to spend very little and now I have to spend big, big numbers for much smaller returns and incremental gains.

Host It's a really hard transition to teach people, but where we generally base that off of is if you're very smart about it and you're very controlled about it. Not every experiment will work. But it creates a controlled environment. It creates a predictability, it creates control over your own destiny. And so it is what turns you from sort of a, a trending brand to a lasting brand that can actually grow in maturity in those types of things.

Host But it's a hard conversation to have, but it's circling this background on the data side. It's so important to establish that. Again, marketing isn't magic. This is why I don't. Believe as much as what I'm hearing on AI is gonna take over [00:19:00] everything automated completely anytime really soon is because there's task management.

Host But there's so much of the nuance expertise and the human psychology around when do I do a sponsorship? When do I buy a billboard, when do I do a TV ad, when do I do meta? What is the mix of all these? And it's not zeros in ones. It's always informing it.

Host But the better we are about getting those zeros in ones, getting those, what did I do and what was the outcome that does exponentially grow those abilities.

Host So I'd love to hear something from you a little bit of, I've talked a lot about these gaps. We've talked about GDPR and those types of things. What are some of the biggest mistakes or shortcomings that you tend to see brands get into in terms of if it's not naming the structure of their data right, or it's not thinking about where and how to place their pixels, or we run into so many bad data setups where it's like double firing of conversions or hard coding versus tag manager.

Host Like, are there a lot of things that create what I will call data garbage that is problematic on [00:20:00] your side? That leads into that?

Guest that's a great question. The number one thing that I would tell brands to get right is the quality of signals that they're passing back to the platforms I've seen.

Guest That'll be the difference between a $50 CAC and a $10 cac. the amount of brands who struggle, who kind of push that, you know, that MarTech infrastructure to the side and say, oh, you know, I'm, I'm sure. Facebook will figure it out anyway, versus the ones who optimize and try to give meta as much information as possible.

Guest That's far and away the biggest game changer for, for making media more efficient. So, that's, that's the one that I would go for you, you know, look, campaign naming is a, tough one, but I think that's the analyst curse for, for, uh, you know, hating when those structures get messy.

Host Have you run into a lot of, like, one of the things that can lead into some of, again, that noise over time is one of the theories that I have when we do [00:21:00] these audits and we look into some of the structures is I feel like there's just a lot of like. Crossover and turnover in the personnel. That's sort of touching things, right?

Host Like every time you get a new personality, a new level of passion or desire, a new either account manager or an operator, you know, it just can start to get like some people, you know, change the UTM structures over time or forget to add in this thing, or again, don't update a pixel somewhere. And again, if you, if that happens over the course of years and every three to six months, you kinda have new people coming in, it just, it feels like it can be derailed so, so quick.

Host The problem with that is it's so hard to retroactively fix anything, right? Like if you miss something, it starts from that point forward, right? There's no uhoh I made this mistake, but I can go back and correct it for the last year. Like, if it's gone, it's gone, right?

Guest Yeah. It, takes a really sick and twisted mind to have the passion for data governance that is necessary to keep these systems up there.

Guest Right. So like [00:22:00] that, that's why so many brands struggle with that. you know, to find that person who can actually set that up for everybody else. It's, they're a needle in a haystack, for sure.

Host Yeah. I mean, that's why, again, when, Todd, when when we have these conversations, I get so hung up on the teams and the personnel, right?

Host Like the empowerment of these tools is incredible, but again, it, they have to be used and they have to be prepped for. And I think the big thing. I'm having this conversation is just the consistency that needs to exist over time. 'cause you can have a great engine today and it can be broken next week if you're not careful.

Guest Well, I mean, what's AI gonna do when it sees your UTM campaign is summer promo. Like, it's not gonna know anything about that compared to the.

Host 2 And it's why as a media agency, we've always invested our human capital sort of headcount in analytics is usually as big or bigger than our media team because of what you guys are talking about.

Host 2 Data structure, data hygiene, data, you know, all that stuff is where we've spent a lot of time and energy because Kelly can't [00:23:00] tell an accurate story if there's inaccurate structure to the information. speaking of. All of that in a, in a large sort of bowl of data. Jeff, one of the questions I wanted to ask.

Host 2 We've spent a great deal of time and energy this year as a company getting, iso, certified as an agency. going through GDPR certifications, HIPAA certifications, ISO 2 7 0 0 1, Certifications. So data, co compliancy, data security, all that kind of stuff, is becoming a bigger and bigger deal.

Host 2 where does that fall in your world at funnel? What are you guys seeing or you know, what's working, not working? Are you seeing big, risky things taking place or, how is security and compliance impacting, your space or the brands that are trying to get all their data set up properly?

Guest Yeah, I think it's, uh, difficult for a lot of the larger brands right now to kind of manage those processes with moving at the speed that they want to be. you know, I had some past [00:24:00] roles evaluating vendors and just needing all these compliances. That can be tough. Like, you know, I'm sure you guys know how tough it was to get that ISO or how long it takes, so, yep.

Guest I don't see too many people doing risky behaviors with it. Like most of our, a lot of our medical pharma clients, they make sure to keep stuff as far away, you know, the, they make sure to keep the HIPAA stuff as far away as possible. But that'll come up in a lot of pitches and proposals.

Guest Big brands. Do you think wo, was it, um, someone came in and, Our account exec needed to hit up the company Slack channel to say, Hey, are we, do we have like a human trafficking training to make sure that we're not partaking in that right? Because that was part of the requirement for the pitch. So there's a lot of, lot of hoops to jump through now, but luckily I think people are being pretty, pretty clean with their data.

Host 2 Well, yeah, and I think it's just gonna continue to get more and more scrutinized. But yeah, I mean it's, it's interesting. We, onboarding two [00:25:00] publicly traded or enterprise level companies this year. It's the first time that we were faced with it being in an MSA, that we actually, as an agency, were required.

Host 2 To possess those sort of data security and compliance certifications. So it's actually new business that drove us to go out and get that. But since we've got it, and now I'm seeing it come up more and more and more, and I feel so much better that we as a company have that level of certification and training that's going on on a regular basis in our company because we handle so much.

Host 2 Client data. we stay away from, you know, the PII and, and that kind of stuff. But, I was just curious how you, what you were seeing.

Guest I'll still get the eventual, uh, data file email where someone will keep the names and emails on it on the end, and I just have to shake my hand and, you know, put it in cabinet.

Guest We, we had a

Host 2 client many, many years [00:26:00] ago in the financial. industry that literally data dumped their entire customer data file onto a USB thumb drive. Yeah. And physically hand it to Kelly and say, here's our entire client database. On a USB drive for a financial institution. I was like, I'm not sure that that's probably a safe thing to be doing.

Host 2 Aggregate

Guest this up for me.

Host 2 Kelly, what do you, what do you got? You got another question? Yeah, I,

Host I got I my last question 'cause we touched on it and I'm really curious on your, on your perspective and I, I don't know where, so there's a, term that we kind of bounce around and can sometimes be interwoven.

Host So we talked about it. There's data analysts. Yeah. And then there's data scientist. And I'm curious as what your thoughts are on how you would define the roles of those two and the importance and, and sort of how you've seen those evolve. it's just a curiosity of mine on, on, yeah. No, that's

Guest a great question.

Guest I think the data analyst part is about. Being able [00:27:00] to take that result from the data scientist and tell people why they should care about it for some reason, like the way the world's been created. It's really tough for people who are good at the science, engineering and math part of data. It's really difficult for them to kind of, I wanna say like dumb that down for the everyday user.

Guest So there's gotta be that bridge somewhere in between it. And I think that's what makes an effective data analyst is being able to be, you know, tell you what it, what does it really mean when you're test tells you your ROAS is between 50 cents and $3, right? there's some gold in there within that wide range.

Guest So the analyst is kind of the one who helps you walk through that.

Host so it's really a translator. Instead of translating Chinese, you're translating zeros and ones or, or data and numbers and, and graphs into language. Yeah. I think that's a great description on it. That's something that we struggle with tremendously is, like you said, there's tends to be that personality that's really good at [00:28:00] modeling and using the data and, creating this great outputs.

Host But there they can be these very long. Hard to understand. Hard to the so what, so what does this mean or why did this go into it versus the storytelling that translates it into, oh, I understand what led into that. I understand why it happened, and most importantly, what I should do about it or where, where it's guiding me.

Host and bridging that gap has historically been something that we've navigated personalities with people that are great with numbers versus people that are great with language.

Guest that's one where I am hopeful on ai kind of being able to create that kind of super, superhuman, that like 10 x engineer where you can kind of use it to fill in the areas that you're weaker and, kind of help create a more, more complete person.

Guest one other thing when we talked about like the full funnel and the brand part, 'cause a lot of my career has kind of been on the finance side of that and kind of challenging that budget. But I think the main thing that I've always challenged my upper funnel marketers with is I. You [00:29:00] have to show me one down funnel metric that is moving from this, right?

Guest Like, You don't have to show me the direct lift in revenue, but show me that more people are searching. Show me that my, meta got more effective in that area. It's like I don't need the, I'm not, I don't need that rock solid finance evidence, but I need to see something that it's, that it's working in the short term because I can't see working anywhere in the short term.

Guest It's gonna be hard pressed to say it's just the long-term thing

Host we like to call the absence of that just vanity metrics, things that just look good, impressions, clicks, things that are like, okay, but what's the next thing that that resulted in?

Guest the number of impressions served?

Guest Should that be barred from all campaign wrap up deck?

Host 2 Yes, please. Yes, please. Kelly and I have been saying for more than a decade, none of that stuff matters.

Guest Yeah.

Host 2 What matters is you spent a dollar, did you get more than a dollar back out? How we got there, however many impressions it took or clicks, it took a lot of those [00:30:00] things.

Host 2 They're directional to media optimization, but they're not strategic in terms of how we decide to financially plan investments. Necessarily so I, I mean, well

Host without that down funnel impact, right? It's everything becomes highly manipulatable and that becomes very problematic, right? It's so easy to make things look good on paper that don't translate to business.

Host Again, we've had so many experiences where we've taken over accounts and clients where I can vividly experience these, where I'll go to the entity that handled it before us, and they'll have this glowing, incredible case study about the greatest work that was ever done on it. And then we're hearing from the client that's taken away from, oh, like our business didn't move.

Host Like, we're down. Like, we're not, the numbers aren't moving in the directions we wanna go. And it's just, it really highlights the how easy a story is to manipulate and change if, that's all you're focused on.

Guest To quote Zoolander, it makes me feel like I'm taking crazy pills when I see those things.

Host 2 Yes. All right. Speaking of crazy, I'm gonna put you on the hot seat, Jeff. I got, [00:31:00] three sort of questions left for you. One word or slightly more than one word answers in the hot seat. and then we can close. so the hot seat, first question is, what is the most, speaking of what we just talked about, what's the most overrated metric in marketing today, in your opinion?

Host 2 Overrated.

Guest Um, CPM. You get what you pay for. Oh, thank you. What's the most underrated metric? Let's go with, marginal, incremental roas. if that's low, then you can take the last dollars, that meta engine and, you know, maybe you get that jersey sponsorship. I like it.

Host 2 If you could ban one data myth from the industry that we all work in forever, what would it be?

Guest that ROAS is a single and knowable number. You're hired.

Host 2 You're hired. Alright. Jeff, thank you. Not often. We get to speak to, an in-depth, analytics mind, such [00:32:00] as yours. so just in final wrapping, where should people go if they want to take a look at funnel?

Guest Yeah. Come check us out.

Guest Funnel io. We've got, yeah, the data hub. Make sure you know where you're spending. advanced measurement on top of that. MMM incrementality testing, MTA, basically the only three ways to measure marketing in, 2025. And then the thing we talked about earlier, passing back quality data signals to platforms through our Activate product.

Guest easy, easy roas when they're, let, meta know when they're acquiring a lead that's gonna convert versus a, you know, someone who's just filling out the form for fun. So, funnel.io, check us out.

Host 2 Fantastic. Thank you. it was good seeing your team out in Boston last time. I was out there.

Host 2 Yeah. So, hopefully it's not too cold.

Guest Yeah. The Red Sox are at the Padres this year. Maybe we can all see you out there. Grab a fish taco.

Host 2 Yes, please. Yes, please. Thanks again, Jeff. It's good talking to you. have a great holiday coming up and, we'll look to get you on again. Sure. I loved it.[00:33:00]

Host 2 Foxing, intro: thanks for listening to contrary to Popular Opinion where we challenge you to think independently, push the limits, and take uncommon action so you can achieve game changing results in your business and your life. To stay up to date with our latest episodes, be sure to click the follow or subscribe button now. intro: Thanks again for listening. See you next time.