Subscribe
Share
Share
Embed
"We built an entire industry on the promise of precise measurement, and most of us don't actually trust the numbers." — Juan
Marketing technology promised precision, but the reality is far messier. In practice, only 3% of CMOs can reliably demonstrate ROI on more than half their total marketing spend. In this Office Hours episode, Jacqueline and Juan unpack the measurement and attribution crisis plaguing enterprise marketing teams. From data fragmentation and siloed dashboards to leadership instability and organizational misalignment, the conversation covers why the ROI gap persists, and what to actually do about it.
They go deeper than dashboards, introducing the concept of an "epistemological crisis" in analytics: when teams can't agree on what's true, the entire measurement stack collapses.
But it’s not all doom and gloom. Jacqueline and Juan break down practical ways forward from fixing taxonomy and defining a measurement tree to embracing uncertainty and speaking the CFO’s language. Better decisions, not perfect data, should be the goal.
Timestamps
00:20 — Martech World Forum recap, data releases, and renewal horror stories
04:34 — The uncomfortable truth about ROI and attribution gaps
07:08 — The “epistemological crisis” and why marketers don’t trust data
13:32 — Leadership churn, siloed teams, and broken measurement
24:27 — Fixing the foundation: taxonomy, definitions, and data hygiene
28:11 — Measurement trees, North Star metrics, and the case for measuring less
33:57 — Speaking the CFO’s language: customer margin and commercial impact
Sponsor
Brought to you by Hightouch, the leading composable CDP and decisioning platform trusted by brands like Domino's, Chime, and Aritzia. 90% of customers have a real use case live within their first week, delivering world-class personalization at scale. Learn more at hightouch.com/msom.
Connect & Subscribe
Creators and Guests
What is Making Sense of Martech?
Unfiltered takes on the biggest shifts in marketing technology. We spotlight what matters, who's leading (or lagging), and what's next. In Martech, clarity is power — and we're here to deliver it.
00;00;05;01 - 00;00;08;08
Speaker 1
Welcome to the Making Sense of MarTech podcast. I'm Jacqueline Friedman.
00;00;08;08 - 00;00;09;07
Speaker 2
And I'm one Mendoza.
00;00;09;07 - 00;00;22;29
Speaker 1
And this is Office Hours, and we cut through the noise and discuss the latest and greatest in the martech landscape. So to get us started, it's been a little bit now, but I got to visit the land Down Under for the first time.
00;00;23;01 - 00;00;40;00
Speaker 2
Oh, yeah. Australia. How'd you find it? I am curious, Jacqueline, about your first impressions of Australia never being there before coming for our MarTech World Forum conference in Melbourne. The first one of the year. It was great to have you. It's great to have you on stage as well. But how do you find Australia? What's your first impression?
00;00;40;03 - 00;00;57;07
Speaker 1
I'm a big fan. Not that I had that much time to do any exploring. So this means I have to go back and actually get to do all the fun stuff. But I'm a big fan of multicultural, walkable, clean, good food. Didn't have a bad meal. Can't really complain. Well, that's.
00;00;57;07 - 00;01;00;20
Speaker 2
A good that's good advertisment. Walkable. Good food.
00;01;00;22 - 00;01;15;26
Speaker 1
You know, I do have a complaint. My flight, which I was lucky it was direct, had no wifi both ways, and it ruined my work planning. That was a new one to me. I had never experienced an international flight without wifi. And, yeah.
00;01;15;28 - 00;01;33;26
Speaker 2
That's, that's Qantas, our national airline. It's notorious for not offering WiFi. It's still stuck in 1990 for some reason in that area. It makes no sense. But I guess the Australians don't complain enough to make a change. So, you know, I'll take that for what it's worth, but it's great to have you. Great to have you at Material Forum.
00;01;33;26 - 00;01;53;10
Speaker 2
We did a great a whole bunch of stuff, which is really great. At the conference for the first time, we released our first ever big data release on, the state of marketing technology for enterprise leaders. It was a fantastic bit of insight there. We boiled down what is 15,000 plus tools into about 202 marketing technology platforms that enterprises actually use.
00;01;53;10 - 00;02;08;12
Speaker 2
Yep. We had to also, you were on stage as well, which was awesome. We did a really cool, like, kind of ask the analyst session, had a bunch of questions from the audience, from all kinds of stuff. And we will solve like the data fragmentation problems, you know, a whole bunch of questions around how to renew with a vendor.
00;02;08;14 - 00;02;17;17
Speaker 2
You know, what's that sort of best advice for renewal? And also some great discussion tables as well. Tell me about yours. You did one on renewing martech. That was that I did.
00;02;17;19 - 00;02;40;15
Speaker 1
It was excellent. We had to choose, use and renew as the three different topics and different tables. And so I was renew and we talked about different ways to leverage negotiation tactics to reduce your bill, ways to circumvent also different levers you can pull to, you know, get the most successful outcome. But how about yourself? You had quite the audience and crowd.
00;02;40;17 - 00;03;00;11
Speaker 2
Yeah. It was it was really fun. We I put everyone to work. So we had three tables of roughly ten people each. And I gave them this big, worksheet, which was basically a self-reflection on how you utilizing marketing technology. One of the case that was, that was circling around the conference was about a third on average of a marketing technology stack is actually being used to its full potential.
00;03;00;14 - 00;03;22;02
Speaker 2
So my session was all about getting people to actually sit down like you're in, like you're in school and reflect on why, what other platforms that are causing this disconnect between using it and actually buying the technology, and what platforms were they and then what was the actual causes for that as well? And then we all went around as a group and shed some ideas on what we can take back to work.
00;03;22;02 - 00;03;37;11
Speaker 2
So it's a really fun exercise. I think it's like a good sort of, good little exercise to get people something they can go back and talk to their boss about, hey, why don't we use that CDP? You know? Hey, you know, we have this engagement platform and we use literally just that for email. Why do we use it for more things?
00;03;37;11 - 00;03;40;01
Speaker 2
So, yeah, it was a really, really great time. I mean, it was fun.
00;03;40;08 - 00;04;03;26
Speaker 1
Yeah. You went more academic, I went more let's go with more tech horror stories. Tell me your renewal horror stories and or, you know, endings. And so we got a lot of people to open up. Everyone was willing because everyone has one. So it was good just sharing what has happened in the space, because you don't normally talk about it outside of with your procurement team, if you're lucky.
00;04;03;29 - 00;04;19;05
Speaker 2
Yes, indeed. And it's is a great time. Yeah. And, and now we're back. You're back home in the US. I'm here in Australia, but I'm actually heading to New York in a few weeks. And so in between all the travel, we're back for another office hours. And, Jacqueline, what is our topic for today?
00;04;19;05 - 00;04;34;13
Speaker 2
We've been going through the six pain points in marketing technology. You can catch the a few episodes back. Back early this year, we did office hours on what those six pain points are, but today we're focusing on one that's a a really a showstopper issue. What is that, jacks?
00;04;34;13 - 00;04;54;19
Speaker 1
Yeah, it's a really uncomfortable truth and tension that everyone is dealing with, but it has to do with measurement and attribution, ROI and how it's really hard to prove and I'm glad everyone else feels it. And it's nice words and named and validated. And it's so interesting because we've built this entire industry on the promise of precise measurement.
00;04;54;19 - 00;05;00;18
Speaker 1
And most of us don't actually trust the numbers we're working with. It's just a tricky, tricky thing. And what are your thoughts on this?
00;05;00;24 - 00;05;24;12
Speaker 2
Yeah, I think that the ROI problem is it's one of those situations where it really is a show stopper. If you can't show the demonstrable impact and, the commercial outcomes from your marketing activities, what you did martech in the first place, you don't know exactly. You know, and I think that's what a lot of executives say from the CFO to the CEO, sometimes even the CTO like what is a demonstrable value?
00;05;24;12 - 00;05;47;08
Speaker 2
We're getting from our marketing investments? Both on the technology side and martech, but also on the advertising side and on also all of the campaigns and channels and everything that a brand will push out. Not getting this right, one of the problems that it's the problem that supersedes every other problem. And I think a lot of other, the other sort of pain point areas, the data fragmentation, the executive pressure on I not getting personalization right.
00;05;47;10 - 00;05;54;07
Speaker 2
All that ladder up to this at all really ladder up to this problem, which is what's the return on investment with all this stuff.
00;05;54;10 - 00;06;07;11
Speaker 1
And what's disappointing is this is the norm. It's not like this is the corner case. This is the atypical situation. This is what everyone is dealing with. And I know we had some interesting stats from Melbourne. Do you want to share what those are?
00;06;07;14 - 00;06;31;26
Speaker 2
Yeah. So we actually polled the entire martech well forum Melbourne audience. So all enterprise, all senior leaders in marketing technology, about 150 of them, about 20% of them said that they are struggling big time with ROI proof gaps, difficulty measuring, incrementality brand impact, cross-channel attribution. So this bucket of attribution challenges, and really those are the folks that would admit it on a survey.
00;06;31;26 - 00;06;51;23
Speaker 2
Right? So then the number of folks that would actually struggle this and not admit it is probably much higher as well, but, you know, again, that's there's just a pretty large share of folks that struggle with this issue. They continue to struggle with this issue. You know, there's actually a great quote to potentially frame this whole problem, which is and we built an entire industry on the promise of precise measurement.
00;06;51;25 - 00;07;08;11
Speaker 2
And most of us don't actually trust the numbers, you know, and that's a really cool quote. And a way to think about this is the problem with measurement and attribution in the martech world is not about the tech. It's all about data. It's actually about trust. Can you trust the numbers?
00;07;08;14 - 00;07;34;13
Speaker 1
Yeah. And interestingly, it's it's not just you know, of course you're the Melbourne audience, but all of the major analyst firms agree that this is a huge problem. So at Gartner, only 52% of marketing leaders prove or are able to prove their value and their team's value. Meanwhile, Forrester is saying 76% of teams are moving past single touch attribution and only 11% use advanced attribution.
00;07;34;13 - 00;07;50;19
Speaker 1
And so if you're only able to measure a small portion of your cross campaign, cross-Channel integrated impact, what are you supposed to do? Is this, just evergreen problem? Everyone is is trying to navigate, and it just adds more pressure to the CMO plate.
00;07;50;23 - 00;08;14;01
Speaker 2
It's a common, theme with the analyst firms and the large consultancies as well that, you know, a very small amount of CMO can really show that demonstrable ROI from marketing. It is such a murky, such a dark, starkly lit hallway, you know, in terms of what is actually driving the outcomes. McKinsey has this great report that came out last year called Fast Forward the modern, the Modern Rethinking of Marketing Core.
00;08;14;02 - 00;08;33;26
Speaker 2
And they said that it was a 3% reality. Only 3% of CMO can reliably show a marketing ROI on more than 50 or more than 50% of their total marketing spend. So bear in mind only 3% on 50% is the threshold of their marketing spend. So it's not the full 100% marketing spend. It's just 50%. Only 3% can actually show an ROI.
00;08;33;29 - 00;08;54;03
Speaker 2
That is a big problem. It's a big problem. And here's why. The biggest companies in the world, the Google meta, Amazon, even some of the biggest companies in the world, the majority of their revenue and their shareholder value lies in this promise that in the digital world, marketers can get precise targeting and precise ROI of their ad spend.
00;08;54;04 - 00;09;13;04
Speaker 2
That's literally the majority of Google's business, no matter all the different business units, including, I don't think that the big thing in the middle that draws the most revenue is still search meta. Majority of their revenue is still advertising on social media. So these two platforms and many others as well, are predicated on this idea that you can get an accurate ROI.
00;09;13;07 - 00;09;33;13
Speaker 2
Why would you want to advertise on a Google or a meta when you can advertise on a billboard or a TV? The reason is you can target people specifically, and you can get a clear, direct digital ROI from those ad dollars. All the analyst firms agree, even our own data from, the TMW intelligence Hub agree that that has not been happening.
00;09;33;16 - 00;10;04;09
Speaker 2
And so we have this very interesting catch 22 of like, okay, these companies are so highly valued. Clearly the world values precise targeting and precise measurement. But the reality is that very few signals, as McKinsey reported, that only 3% of CMO can actually get a demonstrable ROI from their marketing and not getting this right across companies large enterprise, tens of hundreds of millions of dollars, mostly in wasted ad spend and wasted wasted time and building campaigns and launching things without actually having a clarity about what it's going to achieve.
00;10;04;12 - 00;10;17;02
Speaker 2
So that's kind of the the big sort of elephant in the room is that this whole massive, literally multi-trillion dollar industry is predicated on this idea that in reality, most CMO can't actually see the outcomes.
00;10;17;02 - 00;10;43;05
Speaker 1
Yeah. And it just makes me think of more and more and more. The more tools, the more data, the more dashboards, the less clarity. And it's just it all adds up into like, how is this happening? And I think one of the top line reasons for that is actually leadership and stability. And I mean, everyone kind of knows the general term of just CMO tenure is roughly 18 months.
00;10;43;05 - 00;11;03;11
Speaker 1
And, I mean, I know of a number of large companies at the moment that have rid of their CMO and decided, you know what? We don't need most of our marketing team. I can do it. Let's see. Let's see how that plays out. But because of this consistent revolving door of thrashing the strategy and it changes, it resets.
00;11;03;11 - 00;11;15;19
Speaker 1
What is measurement and how you can actually mature it. And so you never actually get to implement what you want to because you need more time in the space for your vision to come to life.
00;11;15;23 - 00;11;42;13
Speaker 2
Yeah, it's interesting, isn't it? There's so many causes to this. And that's why the ROI challenge in marketing technology, it's it's such a hard one to pin down because it's not just one thing at any organization. It really isn't. It's a concerto of many different things happening at the same time. So as you cite leadership instability, but the average CMO tenure, only 18 months, every new CMO comes in, has to have a new strategy, new, well, a regardless, new person leading the helm of marketing.
00;11;42;18 - 00;12;01;29
Speaker 2
There's got to be new strategy, new measurement priorities. And every team is resetting, on average, every 18 months to two years with a new imperative from the. For the marketing leadership, you can never finish what you started. That's one of the biggest issues, I think, in this space. But it's not the only one. Jacqueline, you have some direct impact on this actually with a few customers of ours at the moment.
00;12;02;01 - 00;12;07;08
Speaker 2
But also in your own work in previous companies on organizational misalignment. Tell me about that one.
00;12;07;14 - 00;12;33;06
Speaker 1
Oh, yeah. I mean, not only are the senior leaders not able to implement their actual strategy, their vision, but when what is created is siloed measurement because it's not unified. And as a result, that means you have different teams not only relying on different data, but also are they even using the same data or definitions. And you would be surprised how decentralized those definitions are.
00;12;33;12 - 00;12;56;18
Speaker 1
What a conversion means to one team means very different things to another, and you have to be in constant communication. And it just it causes a lot of mine, mine, mine, everyone claiming credit over the same thing. And so it's just it's tricky. See there's no shared definitions. It's basically everyone's just wanting their own dashboard for their one singular use case, not thinking how it ladders up.
00;12;56;18 - 00;13;12;19
Speaker 1
And it's kind of like Oprah does. Like here's a dashboard, here's a dashboard, here's a dashboard that'll solve everything right? Because that's usually the first ask of a new senior leader. I need this dashboard. And it's like, what are you going to get out of this dashboard that you can't figure out until then? What are your thoughts?
00;13;12;21 - 00;13;32;02
Speaker 2
Yeah, every company is trying to build their own measurement, their own incrementality, their own models. Data analytics in one part of a business, maybe doing something here and then analytics team in the marketing organ might be doing something over here. Often the left hand doesn't know what the right hand is doing. There's been some very interesting examples of this where I can't name the company.
00;13;32;02 - 00;13;53;06
Speaker 2
But, recently I spoke with a company where they had a overall revenue downturn over a calendar year. However, there were three teams in different business units claiming that they've had record revenue because of their marketing activities. And so you can see from the executive point of the big disconnect that some teams are doing victory laps because they smashed their targets, but overall revenue revenues declined.
00;13;53;07 - 00;14;13;01
Speaker 2
How do you reconcile that? Well, when you silo all your metrics, all your data into different pockets, what ends up happening is that people get laser focused on those specific metrics without the situational awareness of other, measurement, other things that need to be in concert with that. So they chase that specific metric and then they achieve it, often to the detriment of other folks.
00;14;13;01 - 00;14;35;03
Speaker 2
So that misalignment, that sort of jigsaw puzzle of connecting all the different, analytics together and also getting teams to be on that same songbook effectively of like, what's this concerto of harmonizing measurement? That is not easy. That is extremely difficult because human nature wants to focus on we want to focus. We don't want to have all this broad understanding of everything else.
00;14;35;03 - 00;14;44;06
Speaker 2
It's very hard for us cognitively to manage all of that. So a lot of people just focus, you know, what's going to be successful for their team and also more importantly, for their role.
00;14;44;08 - 00;15;07;01
Speaker 1
Yeah, everyone wants a piece of the same pie and it's a matter of from the get go. From a leadership perspective, having company wide goals, metrics, you know, whether it is monthly active users for some or really you name any different one depending on the business. But you have to have a singular 1 or 2 metrics that everyone is being stacked up against.
00;15;07;07 - 00;15;30;13
Speaker 1
Otherwise it's going to be just everyone's fighting until there's you're confused and you're in that same situation where a company is, you know, some teams are saying claiming success, when in actuality it's not brought to you by our sponsors. Looking for a smarter way to activate your customer data, meet high Touch, the leading composable CDP, an AI decisioning platform trusted by brands like Domino's, Chime or Radio and PetSmart.
00;15;30;13 - 00;16;08;07
Speaker 1
With high touch, you get the flexibility of composable architecture, powerful self-serve tools for marketing teams, and industry leading time to value. That's why companies choose high touch as the clear choice to personalize every interaction. See what high touch can do for you at high Jet.com forms. And now back to office hours. But I know you have an interesting take on not only we were talking about the disconnect and things like that, but where the trust is going and the crisis it's in as a result, which then directly impacts our industry and the veracity of its need and how impactful and what a catalyst for exponential growth it can be.
00;16;08;07 - 00;16;45;23
Speaker 2
Yeah. It's interesting. We often talk about marketing analytics from a level of a quite, I would say, quite superficial level of data analytics platforms, often in the realm of spreadsheets and dashboards and all of these things. That's actually quite superficial. There is a world below that, believe it or not, I call it the epistemological world. And that's just the academic word for what do you believe is true, where the beliefs of an organization, particularly of the leaders in marketing and also an analytics, they are formed by what is true, what is real, what is not real and what is real in the sense of can we actually trust this information that we're receiving?
00;16;45;23 - 00;16;56;20
Speaker 2
Data is just another word for information. And all of us are exposed to information all day long, and we form beliefs about that. So, for example, Jacqueline, is it hot today at your house where you live or is it cold today?
00;16;56;24 - 00;16;57;16
Speaker 1
It is hot.
00;16;57;17 - 00;16;59;11
Speaker 2
Is it hot? How do you know it's hot?
00;16;59;14 - 00;17;01;13
Speaker 1
You can feel it. Yeah.
00;17;01;15 - 00;17;25;12
Speaker 2
ACS on the air conditioning is on. You walk outside. It's warm. That guy. That's nice to hear that. It's warm because it's been quite cold recently. And you're part of the world. But how do you know that? Because you have empirical information that you've experienced. You felt it, you know, and that is an epistemological reality that your opinion of the weather right now is based on your own experience of that weather.
00;17;25;12 - 00;17;58;12
Speaker 2
In a same way, in a philosophical sense, how do you know that you're going to hit your quarterly target for conversions? I'm going to go into maybe a dashboard. I'm going to go in and maybe, look at Google Analytics or Adobe Analytics, maybe amplitude. I'll go into these platforms and then I will learn, okay, this is what the metrics that these platforms have collected is telling me, and this is how I'm tracking against those outcomes, the same, same sort of framework that or that you believe that these platforms are collecting data that is from real people.
00;17;58;12 - 00;18;19;02
Speaker 2
You believe that they accurately reported stored, collected and filed, that the numbers that you see on the screen reflect the commercial reality of the business. Now, if you ask a marketing leader, can you say that hand on heart? That is true for every single metric you look at in your marketing organization. I would say most marketing we say, no, we don't trust the data, okay?
00;18;19;06 - 00;18;40;18
Speaker 2
The data is inaccurate. The data is messed up. We differ. We argue about the differences and what this data means. No, this metric means this thing, not that thing. And when that happens, what ends up, what goes on is that there's a what we call a epistemological crisis. And analytics. If you can't agree to what's true and what's real.
00;18;40;18 - 00;19;12;08
Speaker 2
And my definition for this is what accurately represents a person doing something on a property, whether it be a website or an ad or an email, etc. if you're not absolutely certain that these numbers reflect real people doing real things, then you're in a physiological crisis. And that's where we see things fall out often, is that we talk at this layer of superficiality, of metrics and dashboards and platforms, but not actually what do we perceive as to be real, real information that have a real, tangible and real impact on our business.
00;19;12;10 - 00;19;23;13
Speaker 2
And so I'm going to throw this back to you because this is a little bit heady, a little bit philosophical, but I think it's worth calling out that a lot of teams don't talk in this in this way, but you can actually be quite helpful to realigning where analytics go wrong in a company.
00;19;23;14 - 00;19;45;26
Speaker 1
Yeah, for sure. And I think a lot of that is multifold, where if you only have a partial visibility in certain things, you're only able to be confident or in non confident in those particular areas. And same goes you, you know of course you're mentioning the data from different platforms. Often times those vendors grade their own homework in terms of what is attributable.
00;19;45;27 - 00;20;19;19
Speaker 1
And that is not how you get an accurate representation of anything. And this is not a just at any platform. It's just that's just the truth. And that's why most teams end up exporting their data or having it sent to their data warehouse so they can do their own analysis. But what's always interesting to me in terms of the trust crisis is the often forgotten component of you realize bots exist, they're only going to get more and more, and we've got a great episode on that called The Battle of the bots, where we talk about there's good bots and there's bad bots, but also there's just a generally inflated metrics.
00;20;19;21 - 00;20;46;13
Speaker 1
And further can be said like, I know I'm a marketer's worst nightmare because I remove Utmb, I clear my cache, I don't want to be tracked. I have all of the ad blockers, everything you can imagine, and that means you're only getting a slice of the real story if you're trying to do a full customer journey. And because there's this quote unquote bad data, you don't have all the data you need to make the best decisions, and you're never going to get all of the data.
00;20;46;13 - 00;21;04;02
Speaker 1
So I'm a big fan of recognizing where data is a signal and a directional perspective, as opposed to tried and true. And I know, Juan, you've got a kind of a horror story of someone getting fired for going off the data as if it were the Bible.
00;21;04;04 - 00;21;21;25
Speaker 2
Yeah. So this is a story about a travel company. Obviously can't name who they are, a large company. They used to hire and fire people based on what they saw in Google Analytics. So they would do a weekly report on, okay, which channels are driving the most traffic, where the conversions coming from as a revenue coming from.
00;21;21;27 - 00;21;44;09
Speaker 2
And every week almost, they would determine whether to up or down their teams. So should they hire another email marketer? Should they hire more teams to do more of this? Should they put their agency on notice that often people would be fired? And the reason literally how HR would say is what we're seeing in analytics aka Google Analytics, is that what you're doing is not making a conversion impact.
00;21;44;09 - 00;22;00;09
Speaker 2
It's very short sighted, initially very short sighted to make a lot of hiring and firing decisions based on conversion metrics. But this kind of business are in their digital business. They're very performance minded. Some companies like that. However, when we went in and had a look at their analytics set up, there was a lot of data that wasn't correct.
00;22;00;09 - 00;22;18;10
Speaker 2
And so when we broke the news to them, it was a bit of epistemological reckoning to go, oh man, we're making really serious decisions, but we don't have Google Analytics instance set up in the way that we needed to. And so that that this like ROI what what are the teams driving. That's a company that's kind of over indexing in one area.
00;22;18;12 - 00;22;43;18
Speaker 2
They've picked a metric which is literally conversions, conversions on their platform, on their website. They've picked that metric, and then everyone else has to be pulled into that metric and then they're assessed on that metric. Whereas in marketing that is not the case, is that, you know, if you believe everything about conversions, what you'll end up doing often is you'll end up doing a lot of short sighted and short term tactics and drive conversions and awareness.
00;22;43;25 - 00;23;01;24
Speaker 2
There's a lot of unknowable data. And then this comes back to this logical concept, which is there are things in marketing that you will never know. I tell my team all the time in our editorial and at our campaigns and our events. I tell them it's like you'd be lucky to know the 5% of the total impact you're making on people.
00;23;01;27 - 00;23;20;01
Speaker 2
As I read your content, consume your videos, even this podcast, and even as I attend the conference, we will never, ever know. And you have to be comfortable with this. And I think we said this in a previous episode many moons ago, which is you have to be comfortable with the mystery. You have to make margin for the mystery, which is you will never, ever know the full story of your marketing impact.
00;23;20;03 - 00;23;40;18
Speaker 2
And that's the epistemological conflict that we're in, which is the technology world has promised that we can do that. And that travel company I mentioned believe that wholeheartedly using Google Analytics data, saying we can know exactly the impact of every single team member, every single campaign, and then we can hire and fire based on that. But actually, it's not true.
00;23;40;20 - 00;24;03;08
Speaker 2
You know, there's things they will never know. There are things that people will react to that you will never see. And so, you know, again, this is a massive, area, an issue that we have to work through in marketing analytics is that often you're not making decisions on bad data. You're making decisions on unknowable data. The great unknown, you know, and you have to be comfortable with figuring out that probabilistic reasoning.
00;24;03;08 - 00;24;04;03
Speaker 2
So, yeah, it.
00;24;04;03 - 00;24;27;04
Speaker 1
Just reminds me of the famous movie quote, except for this time, it's true. Like, we truly we can't handle the truth because we don't have access to it. So I think now it's time to really outline some solutions. It's impossible task. But there are ways to, you know, get towards the asymptote of truth. And so you've kind of alluded to it.
00;24;27;04 - 00;24;50;19
Speaker 1
But for me it's always back to basics, taxonomy definitions, documentation. If you can't define it you can't measure it. It's not sexy. It's not, you know, going to make the waves of we want all of these new flashy things and, and shiny tools and systems. But you have to have your house in order. If you don't have your house in order, you can't do the rest.
00;24;50;19 - 00;25;15;20
Speaker 1
Is the primary building block for anything, and it bothers me. So when people forget that without that, you've got nothing. So get rid of those dupes, clean up the hygiene, get everyone on the same page, have a data dictionary, have the same definitions, or, you know, place in your data warehouse. You name it where you're using the same information and data for everything.
00;25;15;20 - 00;25;28;25
Speaker 1
Otherwise you're not talking the same language. And that's a problem if you're a senior leader. If you can't trust your own team's data because of the other team's data, and that's not their fault. It's a philosophical one, to your point.
00;25;29;02 - 00;25;50;15
Speaker 2
Yeah, but but also intensely practical. Like if you know what Eva, 39, is and Adobe Analytics and only, you know, and no one else, what about 39 names? Then there's a problem. If your team are not aware, they don't have a source of truth to to to go back inside. This is exactly what this metric means or what this variable means, what this dimension means, what this customer profile, set of schemas.
00;25;50;15 - 00;26;06;03
Speaker 2
What? You know, the problem we have is that the complexity and the overwhelm of that make it really hard to go, hey, like, hey, I want to personalize this email or I want to run this campaign and I want to figure out the specific metric. Where do I find it? What does it mean even answering that question? Like what?
00;26;06;03 - 00;26;09;00
Speaker 2
How do we find it? Where do we find it?
00;26;09;02 - 00;26;10;21
Speaker 1
That's sometimes the hardest question.
00;26;10;24 - 00;26;14;05
Speaker 2
It's really not like, where is this data? Initially. And then you have to.
00;26;14;05 - 00;26;20;19
Speaker 1
Think about how many teams do I have to talk to? How many people to have to talk to you before I find out where that data is? Yeah, yeah.
00;26;20;21 - 00;26;53;19
Speaker 2
Yeah. And and it's a sign of leadership. Like, it's a really good sign of leadership where you make the time and effort to really take stock of that, create that taxonomy, as you say, Jacqueline, and actually document it so that you, the team is are empowered or enabled to do that. In fact, one great example of that where we're seeing, we just did a research project in collaboration with Treasure data on AI agents and the value of AI agents, and one of the most valuable use cases that among 13 brands that told us, one of the most valuable ones was data taxonomy, it was actually a data agent to you can actually ask
00;26;53;19 - 00;26;53;25
Speaker 2
it.
00;26;53;26 - 00;26;54;27
Speaker 1
Not surprised.
00;26;55;02 - 00;27;20;06
Speaker 2
Yeah. Like like anyone in the business can go, hey, what is this metric mean? Where is it stored? Where is it collected? And I will tell you now that is a really cool breakthrough that like, that's an example where it's a little bit boring when it comes to AI in comparison. Other stuff. However, it's very, very cool because you can, all of a sudden it doesn't live in a dead empty Google doc or a spreadsheet somewhere, you know, or some other platform where people make it hard to access.
00;27;20;13 - 00;27;44;21
Speaker 2
It can be an agent. It can be a tool that any brand can really just pick up any person and a brand can pick up, use and then get the information they need, which is clear, clarified for them. So yeah, taxonomy is practical. It's kind of boring. It's a little bit time consuming. However, it's set your team up for success if you all are on the same songbook, if you all have the same definitions and the conversations get quicker, I guarantee you that it's not a waste of time.
00;27;44;21 - 00;28;11;18
Speaker 2
It's definitely, a valuable use of money and time to get that right. But more importantly, let's talk about the North Star. So I mentioned that example from that travel company that measured everything against the conversion metric. Getting a North Star is so important. It is so important, not only to define what is the overall marketing objectives and the metrics that stand with those objectives, but we often call those a measurement tree.
00;28;11;20 - 00;28;27;19
Speaker 2
And we as users with our customers a lot. Which would we we would say what are the leading versus lagging metrics. And this whole idea of your lagging metrics, other metrics that will happen in three months. So that would be conversions and revenue. Sometimes a lot of lifetime value. Those are the overall objectives you're trying to reach.
00;28;27;19 - 00;28;43;28
Speaker 2
But we know that in marketing they don't happen to more. They happen in a while. And then what do you lead metrics. So you lead metrics lead into your lag. So the metrics that you see today yeah open rates you click through rates. You conversion rates your boot rates all of those types of metrics that show you the digital performance for today.
00;28;44;00 - 00;29;13;00
Speaker 2
Now those metrics, those laid metrics are not the ones where you hire and fire from, you know, it's overall lagging metrics. But in seeing it in a measurement tree where you have those leading smaller metrics leading up to those overall outcomes, those lagging metrics is a great way to define exactly what you're trying to measure. And what it does, I think was the most valuable thing it does is it creates awareness across it's not just this end metric that we're after.
00;29;13;01 - 00;29;33;29
Speaker 2
We recognize that teams and ad budget and all these things, are feeding into these leading metrics that will eventually have an outcome at the end. And so it gives people a bit of a view to go, okay, well, we're not just measuring these 1 or 2, three things. We're actually measuring multiple things, but they have a purpose in those 1 or 2 things that we really care about.
00;29;34;01 - 00;29;51;21
Speaker 2
And so, you know, most teams don't have a measurement strategy. They kind of have a dashboard, you know, like it's still true today. Like they still have a dashboard. They have some key metrics they want to track. But without that awareness of what are those leading measures that contribute to those overall outcomes, you can't you still shooting in the dark in a lot of ways.
00;29;51;23 - 00;30;17;14
Speaker 1
Yeah. And I love the concept of the tree because yes, you can have the very specific things like Roas, click through rate, those types of metrics. It's always like, okay, but what I always think about that from a leadership perspective, your leaders only care about, okay, about what? Like what does that mean? Why do I care? And so it helps, you know, okay, this is within this Wim Lane.
00;30;17;14 - 00;30;42;29
Speaker 1
And so that's where that measurement strategy of just knowing what your system is and how to upwardly translate it. But to that point, I also think we should measure a little less. And as a result, you can go deeper and be more focused. When you have 600 different things you're measuring, it gets kind of hard. It's one thing if you're doing experimentation, that's a different type of tracking and that's more specific.
00;30;42;29 - 00;30;56;18
Speaker 1
But in general, in terms of your marketing metrics like keep it simple, otherwise you're going to have a dashboard upon dashboards, upon dashboards, as we've already mentioned. And no one wants that version of Oprah. Yeah.
00;30;56;21 - 00;31;14;04
Speaker 2
And this is true that like there's simplicity and then there's there's depth and breadth. Right. So you can go really go really deep to truly understand the 5 to 10 metrics that really matter, the ones that really, really, really change the business impact the business, the things that you want to stand behind even in your own role and say, hey, I delivered this.
00;31;14;04 - 00;31;42;15
Speaker 2
We achieve this. We, you know, etc.. What are those 5 to 10 things are really deep on those. But then, you know, also consider the broader metrics. And it's kind of like gradients, right. Like the deepest you go in is those 5 to 10 metrics that really matter. But you also have to have awareness around the other ones, you know, and I think we're often where it goes wrong is, yeah, a lot of folks try to understand the whole world of all the metrics that literally thousands of metrics that you could be measuring in some cases, but in reality, there's only 5 to 10 on average that companies really need to measure.
00;31;42;21 - 00;31;57;11
Speaker 2
But you need to have enough awareness of the other ones, as well, that especially the ones that have a direct impact on those 5 to 10. But so let me tell you a story about my data bestie. So my data bestie, I used to work with her. She knew Adobe Analytics more than anyone I've ever met in my life.
00;31;57;11 - 00;32;21;12
Speaker 2
She set up all of the tracking. She set up the tag management. She knew every single Ava. She knew every single data point down to its roots. And her job was to make sure that every single analysis project, every single business case had correct and verified data that spoke to the truth, of what was actually happening in that business.
00;32;21;15 - 00;32;43;05
Speaker 2
And oh my love, my data bestie. Because, I would come to her with some of the squirrely, most craziest questions around personalization. Hey, do we have a segment for this? Hey, what if we did that and she would literally come back with a dashboard, a spreadsheet, a report with the data and why? What data? That is why it's collected, and then what it means to some of my questions.
00;32;43;07 - 00;32;47;25
Speaker 2
And I think we all need a bit of a bestie like that. Jax. Like there is in marketing, we can't say.
00;32;47;25 - 00;32;48;21
Speaker 1
The best is.
00;32;48;27 - 00;32;49;06
Speaker 2
Yet.
00;32;49;13 - 00;32;54;09
Speaker 1
There and measurable there. That important irreplaceable.
00;32;54;11 - 00;33;11;28
Speaker 2
And it's a question for you, right? It's a question for the audience. Who is your data bestie? Because they are people in a lot of organizations that know the data more than anyone else. Rely on them. Lean on them often. You need that high powered thinker. You need the business knowledge, and you also need the technical data knowledge and the rare, extremely high demand people.
00;33;11;28 - 00;33;34;23
Speaker 2
But when they come in on a martech team or a marketing team, they have a huge impact because then they become your source of truth. When they leave, you have a big problem. However, when you have the bestie, they create the foundations, the documentation, the taxonomy them, the measurement trees, all of those things they can really support and enable because they have the time, the capacity and the skills to really understand your data.
00;33;34;26 - 00;33;57;13
Speaker 2
But in the last one, this is our last little bit of advice is speak CFO, not martech. Now, we recently had Eloise Gillespie. She's associate director of martech at Optus, a large telecommunications company here in Australia. She had some great insights on this topic, but, what did you learn from that conversation and how does it apply back to measurement and ROI and what companies could be doing about it?
00;33;57;16 - 00;34;19;12
Speaker 1
Yeah. Where do I begin? First of all, I always treasure. Highly recommend listening to all of her wisdom. It was so nice to actually meet her and witness her on stage speaking very similarly in this respect, but you really have to get on the same level as your CFO, and so you need that data bestie to help you do what you need to do.
00;34;19;15 - 00;34;53;00
Speaker 1
And that's where you come in and be that storyteller. So you have to transmute just pure numbers and data into the language the CFO cares about, which is numbers and data, but they don't care about your marketing words, terms, acronyms. They care about customer margin, Incrementality, commercial impact. And Louise's examples are particular to the betting industry, where you know your customer and the personalization is much easier to navigate.
00;34;53;02 - 00;35;17;05
Speaker 1
Doesn't mean it's easy, it's infinitely difficult, but you really have to think, how am I going to get my cfo's attention and thinking of that? Treemap what are the top couple of branches? You know what? What are they? I don't know, every company is slightly different, but highly recommend if you want a full deep dive. Her episode is immeasurably important.
00;35;17;08 - 00;35;26;23
Speaker 2
Can you explain what this idea of customer margin is? Because again, this is a very interesting metric that a lot of companies I don't think use today. What is that customer margin.
00;35;26;23 - 00;36;02;16
Speaker 1
So customer margin, it depends on your industry. So we'll start there. But you have to think how much money you as a company are investing into your customers. And this is actually related to retention marketing lifecycle marketing. It's not explicitly acquisition because there is no real lifetime value. Just fight me. It's not real. I've said customer margin is how many resources, how much budget are you spending expanding of your team to real this customer and in some way, shape or form.
00;36;02;18 - 00;36;34;06
Speaker 1
And you have to determine are you doing that at an efficient rate? Are you making money off of this customer? Are you neutral? Are you not making money? Where do you need to redirect? And that is the the ways in which you can better the company's revenue. Based on that, your CFO will be your best friend. At the same time, if you can't explain these types of metrics to your CFO, you just don't understand them both well enough or enough to be in that conversation.
00;36;34;08 - 00;36;44;08
Speaker 1
And to be fair, I think there's very few people who are in that place. So it just it's a great thing to aspire to and become, and you have to kind of be in charge of it.
00;36;44;08 - 00;37;03;21
Speaker 2
Yeah. I like how she she took the concept of a marketing persona, you know. So here's Bob, he's 35 years old. He does this and this and this. But she took the concept of that persona of that segment, and then she layered on to that. Okay, this is on average what this type of customer spends. This is how frequently they purchase.
00;37;03;23 - 00;37;23;29
Speaker 2
You know, here's the, the history of this customer. What's her total lifetime value and here's how much it cost to service a customer as well, you know, which is the CFOs language, like. All right. To serve this customer actually costs us a lot more when you can bring those two worlds together and you can actually create a segmentation plan or a persona plan where it's, hey, we focus on these customers because it costs us less to service them.
00;37;24;05 - 00;37;41;09
Speaker 2
We make more margin. And also they may spend actually less than other customer segments, but the total, that's why this idea of margin comes into it. Think of it less as in marketing is so prone to this. Often is we think about customers as how many how much revenue can we generate per customer. And that's a metric.
00;37;41;12 - 00;38;01;29
Speaker 2
Okay. So like, we can generate a lot more revenue for this customer. However, they complain ten times more and they cost the business a lot more in customer service requests saying complaining that picture now that is true ROI. When you can get down to the margin, the probable margin of a customer customer group, that's when you're really singing, singing the executive songbook.
00;38;01;29 - 00;38;20;03
Speaker 2
I think, and really shining through that. We're seeing this not just as what revenue can we get from which customer groups, but what margin, which you like easiest to service, the quickest to service the fastest service, the cheapest to service, but then also that highest revenue opportunity as well. So very cool. Very, very cool. There's a great quote.
00;38;20;09 - 00;38;38;23
Speaker 2
I think it came out of the podcast. If you can't explain your metrics to a CFO, you don't understand them well enough. You know, and I think that is such a great through line here as we have our closing thoughts. But, Jacqueline, what is your closing thoughts as we wrap up this episode about the whole world of measurement and attribution and ROI, and how difficult it is for folks.
00;38;38;26 - 00;38;42;09
Speaker 2
What is your closing thoughts?
00;38;42;12 - 00;39;07;25
Speaker 1
I mean, like a sigh, relief, sadness, all of the above. I mean, truly, the goal is not perfect measurement. There is no answer. I mean, there's no solution to that. But you do want to be able to create an atmosphere and an environment for credible decision making. So you can be smart about not just the marketing budget and spend the platform, the team members.
00;39;07;27 - 00;39;16;18
Speaker 1
It's all about the decision making, top down. And so that's where I would focus on where can you build that trust? What about you?
00;39;16;20 - 00;39;42;00
Speaker 2
I think for me it's we really don't need more data platforms or analytics. We need to get very clear on what is true, what information reflects the truth of your marketing efforts. Agreement of what actually matters is way more important than the data you collect. I think, you know, sort that if you're having troubles with ROI, get out of the weight of the everyday data analytics and arguing over metrics and attribution gaps and all those things and tracking problems.
00;39;42;03 - 00;39;55;19
Speaker 2
If you get yourself out of the weights for a minute and think through, what do you actually need, talk to your team about that, and I think you'll find that you'll have a lot more clarity of what you want to measure and the places you need to go really deep, and you have a lot more clarity of what that means for the business.
00;39;55;19 - 00;39;59;16
Speaker 2
The more you do that, I think the more successful, folks can become in this area.
00;39;59;18 - 00;40;10;25
Speaker 1
Last thought, what is this is for everyone to think through. What is one metric your entire company actually agrees on? And if the answer is silence, that's where to.
00;40;10;25 - 00;40;29;12
Speaker 2
Start. That's a great question to end on. And, Jacqueline now concludes our third major pain point, out of our survey last year of more than 200 enterprise brands on what is keeping them up at night. What are the challenges that they're facing? Measurement ROI and proof gaps is definitely one of those big ones. So I hope you found that valuable.
00;40;29;15 - 00;40;43;29
Speaker 2
If you want to keep on following on with us, you can follow us on LinkedIn. We have a LinkedIn page where we release episodes every single week. You can also check us out on TikTok. Woo! We're actually now on TikTok, which is really cool. And so we're we're posting videos and clips there all week, every week, which is really cool.
00;40;44;02 - 00;41;00;03
Speaker 2
Check us out on YouTube, like and subscribe. Hit that button. So you can get this in your feed every single week. I mean, other than that, you can subscribe on the Ma Tech weekly.com and you can get, the newsletter with all the updates from every single episode every week. But for now, I'll say stay curious.