Making Sense of Martech

In this episode of Office Hours, Jacqueline Freedman and Juan Mendoza unpack the overwhelming success of their podcast launch, share details about the upcoming Martech World Forum London 2025, and dive into AI adoption, vendor dependency, and the evolving world of customer data platforms (CDPs).

Highlights

Podcast launch success & what's next: global reach, sponsorship buzz, and better audio quality coming January 2025

Martech World Forum London 2025: key speakers, VIP dinner, and how to network with enterprise leaders

  • Vendor partnerships: the dark side of over-reliance
  • Generative AI's harsh reality check (only 5% of deployments succeed)
  • The future of CDP pricing models
  • Hot takes: Pepsi's $57M AI ad vs Wendy's viral comeback
  • Lessons from the field: failed CDPs, misfired campaigns, and confessions

Key Takeaways

  • AI hype vs reality: start with small, measurable pilots.
  • Vendor partnerships should empower, not entrap.
  • Treat customer data like a valuable asset: maintain, enrich, and activate.
  • CDP pricing is shifting toward enriched profiles and behaviour triggers.
  • Fast, authentic engagement often outperforms big-budget AI campaigns

Timestamps & Highlights

00:00 - Podcast Launch Success

02:11 - MarTech World Forum London 2025

03:32 - Vendor partnerships: necessary evil or obstacle?

14:46 - Generative AI reality check

24:29 - CDP pricing evolution

32:03 - Hot take: Pepsi's $57M AI campaign vs Wendy's $9 clapback

35:17 - Pawan Verma on customer decisioning

38:03 - Question of the Week

40:51 - Confession Corner

Why You Should Listen

  • If you're an enterprise marketing leader or Martech practitioner, this episode will give you:
  • Insider knowledge on 2025 Martech strategies
  • Real-world AI lessons with measurable ROI
  • Advice on avoiding vendor lock-in
  • A sneak peek at the Martech World Forum 2025

Ways to Connect

Submit your confession or question for the next episode: Submit Here

Creators and Guests

Host
Jacqueline Freedman
Host
Juan Mendoza

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;04;27 - 00;00;08;01
Speaker 1
Welcome to the Making Sense of MarTech podcast. I'm Jacqueline Freedman.

00;00;08;02 - 00;00;09;17
Speaker 2
And I am one Mendoza.

00;00;09;17 - 00;00;14;23
Speaker 1
And this is Office Hours and we cut through the noise and discuss the latest and greatest in MarTech.

00;00;14;23 - 00;00;34;07
Speaker 2
And what an amazing first week, Jacqueline. We have launched and resurrected making sense of MarTech. We had email campaigns, we had UTMB, Lynx. We had everything you could think of around our launch. The feedback has been overwhelming. I mean, I've got so many people emailing me every day at the moment, congratulating us on the podcast, looking at also sponsoring the podcast.

00;00;34;07 - 00;00;36;01
Speaker 2
But yeah, it's been fantastic to see.

00;00;36;04 - 00;00;56;20
Speaker 1
Yeah, agree. It's been definitely an unexpected response. Obviously you could always hope for it, but you never actually know what's going to happen. And also the most constructive feedback was my microphone sucked. Any worker and so I have since purchased a very fancy mike. So hopefully the audio is better, but there's a caveat and not so great part of it.

00;00;56;20 - 00;01;13;17
Speaker 1
I've recorded all of the interviews through Beyond the end of this year already because I'm a bit of a perfectionist and overachiever, so I apologize. I will go in and out in terms of when and where the microphone will sound great, but just know moving forward it will be better. And I appreciate the constructive feedback.

00;01;13;21 - 00;01;31;25
Speaker 2
Oh yeah. So if you want to get on the hot seat with Jacqueline Friedman, you're going to have to wait until January, guys. But it is quite exciting because I think for us, you know, we've launched this podcast as a way to really engage the enterprise marketing technology decision maker and the people that are really driving the strategies around technology.

00;01;31;25 - 00;01;47;19
Speaker 2
And yeah, just seeing our initial stats blown out, people from all over the world are listening. And yes, some of the folks that reached out, both wanting to be interviewed, but also just curious to ask more and learn more has been a little bit overwhelming this week. So if you have sent us an email, please accept our apologies as we take a few days to get back to you.

00;01;47;19 - 00;01;50;20
Speaker 2
We're not used to this amount of attention, so thank you so much.

00;01;51;06 - 00;01;52;06
Speaker 1
I can't really like it.

00;01;52;15 - 00;02;11;14
Speaker 2
Yeah, thank you so much. Yeah, I love the attention to the work, but thank you so much for the love. We're really excited to continue this podcast site and other great news. Our next event is coming up. MarTech World Forum in London are really excited for it. It's on the ninth and 10th of September. I'm right in the middle of London, right near Marble Arch in Hyde Park.

00;02;11;14 - 00;02;20;27
Speaker 2
They're really excited. Love to see you there. If you're looking for a conference to come and hang out with bunch of enterprise Motley execs, but we're actually doing something a little bit special. And what are we doing?

00;02;20;28 - 00;02;43;17
Speaker 1
Wow. There's a couple of things. One of which there is going to be a VIP dinner for a very select few of brand side folks. So if you're a marketer at an enterprise, we want you to come because we want to have the conversations that we're all having is just rarely do you get to commiserate among your colleagues, your peers, and realize everyone is dealing with the same things and nuances and learn tips and tricks.

00;02;43;18 - 00;02;47;17
Speaker 1
And it's really exciting. But what else are you excited for? One.

00;02;48;03 - 00;03;08;20
Speaker 2
Obviously excited for the brands that are coming down and sharing what's working in their marketing technology programs. So we've got Sharon Broad. She's a global CRM leader at Virgin Active. She's talking about creating hybrid teams and succeeding in the feature marketing world. Now, James Taylor at HSBC, he's going on about their journey towards taking full control of their data.

00;03;08;22 - 00;03;31;29
Speaker 2
Massive global blank, really interesting. And then we have Diana Kozma. She's a global director of customer data products at Condé Nast. So she's talking about the AI reckoning, the new MarTech enablement framework for business growth. So we've got just really juicy, meaty topics around how enterprise leaders are actually driving results with MarTech and all different ways teams, you know, data infrastructure, data technology and then AI and sort of new technology as well.

00;03;32;04 - 00;03;47;29
Speaker 2
So really excited for it. Go check it out. It's in the ninth and 10th of September. Let's dove in to our first topic for today, which is another I got in trouble again last week. Jacqueline, I keep getting trouble on LinkedIn. I think. I think at some point you're going to have to take my logins away. Someone in the company, I.

00;03;47;29 - 00;03;56;11
Speaker 1
Think I like the sound of that. But at the same time you've got good thoughts and opinions that deserve to be out into the ether, even if they're a bit of a hot take.

00;03;56;22 - 00;04;13;22
Speaker 2
That's nice. Well, let me read you my last hot take. Okay. I'm just going to read this straight off the page already. So overheard from an enterprise marketing technology exec today. So last week, at some point, quote, My goal is to spend as little time as possible thinking about my vendor partnerships and as much time as possible thinking about our business.

00;04;13;22 - 00;04;29;27
Speaker 2
I don't think a lot of vendors get this when they deal with the enterprise. They don't want to be at your product launch. They don't want acute account health catch up. Your keynote thought leader has no chance of getting a meeting. They certainly don't want to be haggling over price increases. A VIP luncheon isn't the perk you think it is?

00;04;30;02 - 00;04;47;10
Speaker 2
Sometimes it's just about the commercial impact of managing a relationship that gets in the way of the work. Let the tech speak for itself. If you need a bajillion customer success managers to get an email out, then you have a big, big problem. Now, I posted that and of course, a bit of controversy. You had a lot of technology vendors saying you don't understand.

00;04;47;10 - 00;05;04;05
Speaker 2
We're an extension of the teams that we work with. They need us. They need guidance, they need strategy. We deliver it. You know, you're missing the point here. A vendor that is basically saying that they want to minimize those partnerships, must have everything sorted and have a perfect world and they don't need any help at all. But Jacob, what's your take on this?

00;05;04;05 - 00;05;15;00
Speaker 2
Because I think it's a very mixed response. Some folks were saying this was exactly it. Right. Like brands want to focus on their business. They're not there to build and set up technology. You know, like so what's the what's your take on this?

00;05;15;02 - 00;05;42;21
Speaker 1
Yeah, it's tricky because there is value in having your vendor as a partner however you want them for their expertize not to do the job for you ultimately. So yes, upon implementation, it's definitely helpful if they're the ones with the expertize to help you implement. However, you also want to know how to use these tools. That lack of institutional knowledge is what I've seen time, time and again, both when I've been in-house, but with also with clients.

00;05;42;21 - 00;06;06;16
Speaker 1
That is the Achilles heel. They lose one person who knows all the details. There was no documentation. If you don't know what you're doing and you don't know what we've done and what you've failed at, how are you going to make those programs even better? And so I think there's an overreliance on vendors where as it relates to quarterly business reviews, all of those things.

00;06;06;16 - 00;06;28;21
Speaker 1
That's not to say they're not valuable, it's just the way in which it's being in practice and ways folks are using it right now. It is surface level, it is unhelpful. And I want a vendor who actually is a partner, meaning they're an extension of my team and they are bringing real value and giving pointers and showing expertize.

00;06;28;22 - 00;06;52;05
Speaker 1
Oftentimes when I've worked with vendors, I know more than they do about their product. And that is a problem because as a practitioner, you can go from company, the company as you leave and leapfrog around. But you're still using the same tool. However, on the customer success time side, customer support side, you'll only be there for a few years maybe, and you'll move on to the next and you have to become an expert in that.

00;06;52;05 - 00;07;01;23
Speaker 1
But you can't really become true expert. If you only have a year or two years in a tool, you have to really be hands on using it. So that's my take. What are your thoughts?

00;07;01;25 - 00;07;28;12
Speaker 2
I would see that there are positives and negatives with the vendor extension of your team type concept. You know, on one side, I think the quote stands up by itself. The business is there to serve their customer, not to set up technology, number one. You know, so for technology vendors selling in, you know, I would say that if it's all this obfuscation and all these different partner things that need to happen and all this sort of hand-holding, there's a fundamental problem with the technology.

00;07;28;13 - 00;07;47;20
Speaker 2
Yes. Enterprise organizations are complex. That's the problem often, isn't it? The vendor, they're right in terms of that complexity. But in itself, you know, we have our own partnerships and our relationships with our technology vendors and they often just get in the way. You know, we hire a whip crack, awesome, smart person that can actually just come in and build things with us without technologies.

00;07;47;26 - 00;08;07;18
Speaker 2
And then that system just often just gets in the way. So like, you know, often that we just said, Oh, yeah, I mean, look, they'll catch up, you know, we took a call. It was hilarious three weeks ago where they said, Oh, okay, you know, you guys are not using these features by using these features. Instead they said, duh, because we don't need the 20 features that we purchase, but we can't customize it, you know?

00;08;08;00 - 00;08;25;03
Speaker 2
So it's like scratching the head going, Well, okay, do you want to use them more? We're like, No, we don't need to. So, you know, often those things happen where, you know, it's just a bit of cluelessness around what is actually needed. But on the other side, I mean, a lot of enterprise marketing technology came at a time when there were no digital skills.

00;08;25;09 - 00;08;43;07
Speaker 2
There was no person on the other side of that call where there was. They could just call out the nonsense and go, This is what we need. This is what we need to build, you know, and get going. I'll give you one example. So we have this problem with a customer recently. We're sitting down with them. There are, you know, $15 Billion company and large organization, very complex.

00;08;43;14 - 00;09;04;16
Speaker 2
But Salesforce came at a time when they had no digital skills or real maturity in the business. And so Salesforce themselves built their entire API infrastructure from Salesforce marketing cloud down to all their downstream tools. They want to shift to next generation technology, particularly around marketing, cloud and marketing automation. But they can't because it's a black box. They don't know actually how all of those APIs work.

00;09;04;16 - 00;09;23;04
Speaker 2
They have no idea. And so that's it is pressing. And it's it's a real sticking point where companies now what now they have the maturity and the skills. And now we've been through sort of a good 15, 20 years now of this digital transformation. They're stuck because the vendor has ingratiated themselves so much, enmesh themselves so much into the business.

00;09;23;04 - 00;09;30;24
Speaker 2
Don't get me wrong, that's a great way to minimize churn. If your customer literally cannot churn, that is a fantastic business strategy.

00;09;30;24 - 00;09;35;20
Speaker 1
Don't get more than a bit more like a leaked situation. My personal opinion.

00;09;36;14 - 00;09;59;18
Speaker 2
Exactly, yeah. A late last week we talked about avocados. Now we're talking about leeches, but. And the same, same. But the reality is that, you know, 70% of digital transformations fail. You know, McKinsey said that 67% of MarTech goes on use. That's from a recent 2024 CMO survey said 50% half of CMO's are disappointed with the payoff of MarTech investments.

00;09;59;25 - 00;10;30;16
Speaker 2
Why? Because it's because a lot of the technology that is coming into enterprise still needs a lot of that specialist skill set that really only the vendor can provide. So it's a bit of both. I would say that, you know, there is a dichotomy there between, well, is this really the vendor trying to leach themselves into the business or is this really genuinely the enterprise brand wanting help, needing help, and then the specialists that understand this technology understand how it works with several other companies and their other clients, you could see there there's value in those partnerships.

00;10;30;16 - 00;10;43;02
Speaker 2
But at the same time, you know, marketers are not here to set up technology. You know, in fact, I would go so far to say if your job is really just setting up technology and you're in the marketing department, you're missing something because it's a bet.

00;10;43;02 - 00;11;01;00
Speaker 1
That's my favorite thing. And that's what I would do, that there's a the wing of the marketing technologist ma ops that I always see it as a kind of a church instinct. You have your market hours, but then you have the technologists who are integrating the systems and working with engineers and product. And it is a distinctly different skill set.

00;11;01;00 - 00;11;37;06
Speaker 1
Doesn't mean you can't go between the two, but to your point, the tools should be self-serve enough to get a certain point done. And where you have that implementation support is really the things that very few have knowledge of, whether it's deliverability, actual best practices on data structuring and working with the data engineer for that. And just like specialized components, I mean, there is such value of, hey, we can give you a demo of your actual data in 5 minutes versus let me show you this splashy figma or demo of things.

00;11;37;06 - 00;11;52;26
Speaker 1
And you know, there's smoke and mirrors. It's deeply disappointing. But also I hope it's the kick in the butt that enterprise companies and really all MarTech vendors realize they need to have a better solution and better product in order to sustain growth and happy customers, to say the least.

00;11;52;28 - 00;12;16;23
Speaker 2
We recently in our business was looking at some data Richmond tools. We came across this really great product that used A.I. to sort of spot all of the missing data in our CRM. And that is an interesting example. There of you're really just writing prompts and then the tool does the work for you. And I think increasingly, you know, we're coming out of this phase of clunky, you know, software that you have to have coding skills to use.

00;12;16;23 - 00;12;38;25
Speaker 2
You know, like AI is the ultimate no code. There's another way here which is worth considering, which is I think as we the trend will be less than partnerships and less incentives to be trained just because A.I., the promise of A.I. is that the technology just works. You know, the reason you need all these people in your business, because the technology isn't working in some way, and often that has a skills gap.

00;12;38;25 - 00;12;57;21
Speaker 2
There's a talent gap to make that happen. But the A.I. flattens that talent gap, I think. And so that's like I would say anything that quote, that original quote, that we want to minimize our time on vendor partnerships and spend more time in our business. I think with that quote will become more and more true with the age of A.I. tooling and the ease and accessibility of tools and technologies.

00;12;57;21 - 00;13;00;10
Speaker 2
You don't need to write SKU anymore. You can write props as an example.

00;13;00;11 - 00;13;10;02
Speaker 1
Okay, but who's going to be paying for certifications to ensure they're considered experts? What goes wrong there? I digress.

00;13;11;11 - 00;13;30;27
Speaker 2
I look vendor of certifications. I'm a big believer. I love them. I give you rounds. You're not a very interesting. I got my my best education. I came from technology vendors. So in the sense that because it's not really like if you wanted to learn a CDP, you're better off going to a CDP that's actually built actually as a course than something that's a little bit more agnostic.

00;13;30;29 - 00;13;49;03
Speaker 2
You're better off going deep into one tool in the beginning instead of, you know, getting a high level view. I've had a comparison. They like a CDP course from a vendor and then from a like a third party. The third party was too superficial for me with the other, the vendor. You go deep into the tech, you're building a use case with the technology in a sandbox also in that regard.

00;13;49;03 - 00;14;10;16
Speaker 2
So in a future, why do we need certifications in the first place? You know, like, you know, if the promise is, is real and that is the promise that the technology will you be able to use it with basic skill sets and not having have this really hyper specialized knowledge or skills, you know, then it really does flatten what the role of the vendor the vendor is just often provide the technology.

00;14;10;20 - 00;14;30;27
Speaker 1
I will say I've made it this far in my career with zero certifications, and that's been both by happenstance and by purpose. Didn't work for companies who knew to even do a professional development budget and or got to the point where I knew the tool better than every customer success support you name it. And then I just would carry on to the next one and do the exact same thing.

00;14;30;27 - 00;14;39;14
Speaker 1
And it's just trial by fire and practice. And so I'm a staunch believer and you can learn far more than, of course, by just. Oh.

00;14;40;03 - 00;14;46;17
Speaker 2
Yeah, oh, yes. Very true. Speaking of trial by fire, let's jump to our next topic, which is RAM.

00;14;46;17 - 00;14;47;23
Speaker 1
Yeah, it's a big one.

00;14;48;06 - 00;15;09;25
Speaker 2
The air bubble and disillusionment. Now, MIT came out with a big report. It's caused a bunch of controversy in the marketplace. They're saying that only 5% of generative A.I. deployments based on their study are successful. So that's companies using generative AI in some capacity and it's not that bad. Let me read you a quote. Benedict Evans, he's a prominent analyst in the tech world.

00;15;09;28 - 00;15;24;16
Speaker 2
He said, this, I thought is hilarious in response to all this controversy. He said that I'm hearing disturbing rumors that quick pilots of a brand new and very immature technology do not always produce medium slam dunk successes.

00;15;24;16 - 00;15;25;03
Speaker 1
No kidding.

00;15;25;26 - 00;15;28;26
Speaker 2
No kidding. I am not surprised by 5% at all.

00;15;28;28 - 00;15;51;06
Speaker 1
Yeah, I mean, I would have hoped for maybe 15, but every and I said it in our previous office, our episodes, it's a bubble and everyone's in their one year posse and you're lucky if you get two years. And if not, you're not going to be in business for very long. And ultimately what really needs to happen is the economics of leveraging AI.

00;15;51;07 - 00;16;09;27
Speaker 1
Right now it is just very expensive per call, per API call, it adds up and if you're looking for incremental increases in revenue that's eats away at that versus if you're doing really big swings that could go horribly wrong or horribly right in the best way. That's how you're really going to be able to have a successful deployment.

00;16;09;27 - 00;16;18;28
Speaker 1
But I think for the most part, it's not working, but it will eventually. But costs have to come down and also probably macroeconomic conditions have to change a bit, too.

00;16;18;29 - 00;16;24;23
Speaker 2
Yeah, there's a very, very ancient saying when you chase the wind, you reap the whirlwind. And what that.

00;16;24;23 - 00;16;27;04
Speaker 1
Means, I don't know that what I like it.

00;16;27;17 - 00;16;42;23
Speaker 2
What it means is that basically you're grasping at nothing, right? Try to chase the wind. You know, you try and capture it in a bottle you can't hop is almost exactly the same when you have companies, when you have all these A.I. experts with, you know, all of a sudden there's like a million LinkedIn A.I. experts. You could always keep.

00;16;43;00 - 00;16;55;01
Speaker 1
Like get rid of expert and guru from our vocabularies, because pretty much if someone immediately says that I'm a guru and X, I'm an expert in this, I just kind of shrug and laugh and move on to the next person.

00;16;55;08 - 00;17;14;12
Speaker 2
I get. I guess I don't shrug. I get angry and try and criticize them. But, you know, horses for courses. But I guess, you know, the thing here is that the Silicon Valley industrial complex around hype, the integration between media, the progenitors of social media, the distributors of hype and then the technology companies and the VCs behind them.

00;17;14;21 - 00;17;33;09
Speaker 2
It's such a well-oiled machine now where one use case, one thing that generates a significant outcome or even just an idea gets all this hype and draws all this internal pressure in the enterprise. But I would say that for us, even in what we see day to day, we're not seeing a lot of successful case studies in the magic world around.

00;17;33;25 - 00;17;57;15
Speaker 2
You know, I'm not seeing a lot of explosive AI projects that are driving huge amounts of revenue. In fact, a lot of technology companies come to us and ask, what do you think about our go to market for for A.I.? And I said, hey, draw a dot. Put a dot on this page. It's strategy. This is right. You have to put one dot on the area that is actually driving revenue and they can't because it's still new, it's still working it out.

00;17;57;20 - 00;18;15;03
Speaker 2
And all of that is quite exciting and fine. But I would say that the like going back to this study, there was a common thread between homegrown and integrating directly with an alum and there was a big difference there, you know, so the folks that have an internal team that have already been working on this for quite some time, they're more successful.

00;18;15;08 - 00;18;31;05
Speaker 2
The companies are just picking up an enterprise subscription and then trying to make it work. Is seeing that, no, it's not going to work like that. You know, it's not just about buying software and making it work. It's actually about the transformation, the mindset within the business to make it work. It's interesting. I'll give you one really good example of this.

00;18;31;05 - 00;18;55;24
Speaker 2
We recently did a case study with Rabbit, so they are a version of like Uber Eats or DoorDash for Latin America and the head of MarTech. As such, he basically set up this strategy to complement their existing A.I. decisioning. So they have a very popular app. They've got millions of customers on the app every day. Basically, they had all these different offers and deep links and advertising campaigns going out every single day, thousands of thousands of them.

00;18;55;24 - 00;19;11;08
Speaker 2
And what was happening was that there was all these breakages in the experience. So someone would see an ad on Facebook or on Instagram or something that click through, they click through, and then that the deep link into the app would break and this would be happening. They couldn't really measure or figure out what was going on. And so.

00;19;11;26 - 00;19;12;24
Speaker 1
That's breaking.

00;19;13;00 - 00;19;32;01
Speaker 2
Well, yeah, it's terrible and they'll lose. They knew that because they're losing like like hundreds of thousands of dollars a day in revenue just by these broken journeys, you know? So what they did, which is fascinate, they employed a company called Mobile, which is they basically have a massive factory filled with robotic fingers and cameras. This is going to sound weird, but basically what they did.

00;19;32;01 - 00;19;34;23
Speaker 1
Sounds a little futuristic and terrifying.

00;19;35;08 - 00;20;03;26
Speaker 2
And dystopian. It's a little bit outside the Lynwood, but stick with me. So it's an alien system that uses robotic fingers and then computer vision and an AI model to test every single version of the app as it's being released. As new campaigns go out and they've got thousands of these robots and basically that computer vision would test it on an actual device, and then a robotic finger would be pointing and clicking at the elements of the website page and over that initial program of work.

00;20;03;26 - 00;20;32;02
Speaker 2
So they just did a pilot like just a month, I think it was, and it only cost $5,000 to run the pilot, but they got $150,000 in return, purely just by targeting some key journeys and using computer vision to fix those. And then it actually uncovered where all the broken journeys were. So the AI system there machinery, computer vision and then a robotic finger actually drove incremental returns from the ad spend so much so that the budget went out of product, actually went to the marketing team for that technology.

00;20;32;11 - 00;20;54;27
Speaker 2
Now to me that's a great example of a company thinking really clearly about how they can solve problems with AI. You know, and I would say for large consumer brands, you see that a lot, you know, like the amount of QA that needs to happen over an experience that changes a thousand times a day is unfathomable. So you do need AI to actually enable that and to manage it so you can get what I see, right?

00;20;54;27 - 00;20;56;20
Speaker 2
Like there's this things there that can really.

00;20;56;20 - 00;20;57;10
Speaker 1
Opportunity.

00;20;57;10 - 00;21;05;00
Speaker 2
But there's transformational change and the only transformation that can actually happen there is through your team thinking differently about using technology. And that's a great example.

00;21;05;00 - 00;21;33;01
Speaker 1
Yeah, anytime someone is asking How do I implement AI into my tech stack, my immediate question is Know what needs improvement, what gaps, what opportunities you have ahead of you. So then as a result, we can figure out maybe I can be plugged in, but the crappy example so helpful because there are a lot of large companies, not to mention sort of regular, you know, mid-market to snobs that don't do uot they don't do okay in the way they should.

00;21;33;13 - 00;21;55;16
Speaker 1
And it's not only just as important as your customer support, but you literally don't have a technology that works if you do not make it work for every customer. And so it's anything it's an accessibility requirement to an extent. If your customers are not able to access it, why did you build it? What was the point of this journey and where was the rigor?

00;21;55;16 - 00;22;06;01
Speaker 1
And so I think it's really great and helpful and achievable. Example five K is a drop in the bucket for so many companies and the results speak for themselves.

00;22;06;12 - 00;22;30;10
Speaker 2
It's an amazing outcome. And if you extrapolate that over like that, was this a very small segment running for just a month? Right. They extrapolate that out. You're talking millions of dollars in incremental revenue from ad spend just by fixing broken journeys. You know, it's boring. It's not the exciting, sexy, not the sexiest. Yeah, yeah. It's not the crazy stuff you see from Microsoft and Openai and all that gurus on LinkedIn, but it's stuff that works.

00;22;30;10 - 00;22;51;19
Speaker 2
I think for marketing. I think that this is what we're starting to see is that at the back end of business, you know, all logistics operations, customer support is stuff that doesn't make the headlines is where all the revenue is coming from in terms of improvements and where all the productivity improvements are coming from. But in marketing, it's art, you know, it's the flashy stuff that doesn't seem to be as revenue generating as you think it is.

00;22;51;19 - 00;23;14;19
Speaker 2
So, again, very interesting time. I think the controversy has definitely damaged the stock prices of some companies, especially my T study, you know, even sent Altman last week according to Ed on Sam Altman. By the way, Ed has got a fantastic newsletter called Where's Your Ad at a very pessimistic on tech, however, very good on reporting. And he's saying that even sell mountains recognizes a bubble here and there are going to be companies that get hurt.

00;23;14;21 - 00;23;35;16
Speaker 2
There is a very Silicon Valley esque thing going on here with the bubble and the hype drive, all this investment in momentum. And then out of that, you know, it's kind of like a forest fire, right? The collapse of the bubble decimates a bunch of companies, but only the best ones kind of remain. You know, it's very Darwin, you know, like Darwinism where it's like, you know, the ubermensch or the, you know, the strongest only survive.

00;23;35;16 - 00;23;36;05
Speaker 2
I think that's what.

00;23;36;05 - 00;23;59;15
Speaker 1
Survival of the fittest for sure. So speaking of this, this disillusionment that really does tie into Gartner's hype cycle, and they even recently said that we are entering the trough of disillusionment and that 5% number makes so much more sense. And from what they say, the low point is really expected next year, and then it'll eventually just become a productivity tool, which I feel like that's what we're seeing in the space.

00;23;59;19 - 00;24;21;04
Speaker 1
There's a few outstanding examples like the rapid one. There's an upcoming interview episode on the hot seat where we've got a really great fashion brand that is walking us through how they've implemented it successfully. But those are only really two substantial examples that are not just an incremental productivity and efficiency gains, which can be great for certain manual processes.

00;24;21;04 - 00;24;29;13
Speaker 1
But to your point, marketing is art and they're science, but it's a lot of art and it's not as easy to make things cooperate as a result.

00;24;29;25 - 00;24;58;15
Speaker 2
Very, very well said. Moving on to the matrix pricing proxy problem, the matrix fraud pricing proxy problem, it's a bit of a tongue twister. We're talking about composable CPUs and the shift around how CDP infrastructure is being priced. And we're seeing this really interesting feedback from from our members and our customers around this, around saying, well, now we're starting to see that the way CDP is used to process themselves has changed a lot.

00;24;58;15 - 00;24;59;21
Speaker 2
But what's actually happening here.

00;25;00;04 - 00;25;26;18
Speaker 1
I think folks are finally realizing I shouldn't double pay for my data and also I shouldn't silo my marketing data from the greater business. And so for me, I find that actually the growing pain that's been necessary for so long, the amount of times I've been on, I would say, a traditional CDP demo, and I leave the demo saying I have no idea what I just saw and this is my stuff that I, I know.

00;25;27;07 - 00;25;47;13
Speaker 1
And as a result, all of that looks like murkiness and lack of clarity means not only are people confused why they're paying a significant portion of money, but also they don't actually know what it's doing. And it's intriguing because I know one is really difficult to price all these types of things. I mean, there's entire pricing teams sometimes at different companies.

00;25;47;13 - 00;25;49;10
Speaker 1
And so treasure data has a new solution.

00;25;49;19 - 00;26;11;02
Speaker 2
That solution is is actually quite interesting. So what they're saying is effectively cost the compute. So data storage data set between one app to another, they're actually killing that. So they've got a no compute pricing model, which is pretty different for most. CDP pays most of it pays out pricing things around compute around storage. You know, that's actually how they make a lot of money.

00;26;11;02 - 00;26;29;02
Speaker 2
The traditional CDP is just really the data storage component in their business model. But Treasure Data's come out with this idea to go, Nope, we're going to only charge you for two things the amount of customer profiles you have in your CDP and the amount of behavioral events that are triggered when a customer actually does something on your website.

00;26;29;02 - 00;26;58;28
Speaker 2
So the rationale goes something like this. You as a marketer want to see the most value coming out of your data. So personalization, segmentation, you know, analytics, all that stuff leads to revenue and outcomes. However, if you're paying just for really infrastructure costs, that doesn't ladder up into an actual tangible ROI for marketing and so treasures bet is that if you price it based on the amount of customer profiles in your account and also the amount of events that are firing off in that profile that's enriching, the story goes that it's actually building up that data asset.

00;26;58;28 - 00;27;17;25
Speaker 2
You're actually the money that you're spending with treasure data would be investing into enriching that customer profiles. So that data set over time and then theoretically you'll be getting more value out of that data set through everything I just said. The way to activate it, the campaigns that go out, the personalization strategies and then the analytics and insights as well.

00;27;18;02 - 00;27;31;21
Speaker 2
So that's sort of that the core of what they're doing. I have a smoker take on this one thing I would say that still about say what? Yeah, yeah, well, yeah, news flash. But I would say that the.

00;27;31;22 - 00;27;32;22
Speaker 1
Big problem.

00;27;33;00 - 00;27;49;04
Speaker 2
You may yeah the main problem with this pricing model is that it does take a bit of a mental shift in the enterprise to see that data as an asset. I believe it or not, most companies don't think about their customer data as an asset. They think about it as just numbers, as churn, as fodder to generate revenue.

00;27;49;14 - 00;28;03;15
Speaker 2
But the smart companies think about it as an asset like you would as a portfolio manager or. AVC or a, you know, an investor to go, okay, I've got all these different assets, and they're accumulating in value for a company. Your data is your asset. You know.

00;28;03;18 - 00;28;29;22
Speaker 1
You're giving me terrible premonitions of like Palantir owning everyone's data, and that is their asset and only business model. So you're giving me some doom and gloom, but I've never really had that official like a through line of seeing it as your company's primary asset. I It's always been important, but I've never phrased it that way in my head and I really like that concept because it's understandable, especially among all the folks we're working with and we talk to.

00;28;29;22 - 00;28;49;17
Speaker 1
We're all in tech in some way, whether it is our tech or you work at a tech company, it depends. And if you're able to shift your thinking, your mindset around what your data is and realize the significance and importance of zero and first party data, screw the third party data it only has from a graphic help. And even then it's dicey.

00;28;50;03 - 00;28;58;04
Speaker 1
But I really appreciate that phrase and it's changed my perspective. So thank you for changing my perspective on something I didn't know existed.

00;28;58;19 - 00;29;06;11
Speaker 2
It's very interesting, isn't it? Because every asset you buy a house, that's an asset you have to maintain it. All the walls will start to cave in your mind.

00;29;06;26 - 00;29;09;04
Speaker 1
It's so expensive. Yeah.

00;29;09;17 - 00;29;24;02
Speaker 2
Exactly. You know, you know, when we bought our house, we put up a new fence, we painted the house, we bought a new couch. You know, we set up a projector. We did all these things in our house. And it's exactly the same with the house, with the customer data. You want to maintain it, it costs money to maintain it.

00;29;24;06 - 00;29;42;27
Speaker 2
But guess what? As that asset grows, the value grows because you have more marketable customers, less churn, you can get more value out of those customers. It really is such a cool concept and I think that's what Treasure Dave is banking on, is that brands increasingly see that customer data as an asset. But my critique is that most companies don't see that today.

00;29;42;27 - 00;30;02;22
Speaker 2
The CDP alignment to our ROI is always been tricky because they don't send emails, they don't put a website personalization on the website, you know, they don't do things that they can actually see the outcomes straight away. And the other categories of MarTech that do those things, they can go and say, We launched this campaign. It was personalized to drive this uplift.

00;30;02;22 - 00;30;18;08
Speaker 2
And then they go back to the executive say, hey, look, we're generating significant returns out of our investment where the CDP is just upstream of that. And so, you know, I would actually suggest that companies think about more about the data activation pricing around that. So the data that comes from is CDP and is activated downstream into the tools.

00;30;18;18 - 00;30;35;27
Speaker 2
Every time that's activated, you know, then you're clipping a ticket in some form or another. But, you know, there are different mental models and different ways in which they pay a price. I just think that this one is particularly interesting because it's focusing more on the customer profile and that data asset concept. Again, fascinating. And there's got to be a lot more change in the space to guarantee it.

00;30;35;27 - 00;30;53;20
Speaker 1
Oh, yeah. I'm still thinking about the analogy of thinking of your entire textbook and just your data as the house that you purchased, because there's three things you can do and didn't know it's going to go all Martha Stewart here. But there's three things you can do to increase the value of your house is upgrade the bathroom, upgrade the kitchen and landscaping.

00;30;53;20 - 00;31;15;25
Speaker 1
I always consider MarTech the plumbing of every company, whether they realize it or not, is the most complex. Because you have to bring the product analytics, you have to bring in the orders, the revenue, all of this into one place versus just one focus, one primary. Did this increase lift on X? That's so without the context of the greater business.

00;31;15;25 - 00;31;36;16
Speaker 1
And so if you only upgrade the landscaping, that's great. You're going to get an r y there. However, people care about the bathroom, they care about the kitchen, and we're plumbers. Ultimately, we're the back of the house. We are just making sure the integrations sing. And there's no leaks, there's no issues. They call us when something goes awry.

00;31;36;16 - 00;31;47;09
Speaker 1
There's very few thank you's usually as well, especially if something all of a sudden just bursts and leaks and they blame it on you. Versus sometimes, you know, it's like, yes, it's certain things happen.

00;31;47;13 - 00;31;53;22
Speaker 2
Very hard to sell a house with bad plumbing, especially if someone looking coming to look at the house, turn the tap and nothing comes out. So or.

00;31;54;00 - 00;31;54;06
Speaker 1
Oh.

00;31;55;01 - 00;31;56;07
Speaker 2
Yeah, it comes up.

00;31;56;08 - 00;32;02;03
Speaker 1
Real talk about we'll bleep it out but fitness right there.

00;32;03;09 - 00;32;16;19
Speaker 2
Yeah, yeah. Open it. Open up your CD app and turn the tap on your clip. Only grab water comes out you know you don't want that, right. And I think that's what a treasure date is trying to head off. But moving on to something a little bit more fun here, a hot take of the week. I love this.

00;32;16;19 - 00;32;34;02
Speaker 2
I'll take we call that Pepsi versus Wendy's. So there's this group called Higgs Field Products. Basically what they do is video. So they do lamp videos. You know, that's something that people are experimenting with. So the tweet went like this. So Pepsi took a $57 million campaign and made it for nine bucks using the Higgs field product. Right.

00;32;34;02 - 00;32;52;08
Speaker 2
And then it's got the demo of the video, the videos. It's the most swap stuff you would ever say. Right now. This is terrible. But the entire Detroit team gone the ad world forever, right? That's what the Post said. And then went, oh, that's under way. Yeah, exactly. It's like, oh, yeah, doom and gloom when these comments very simply, that was a waste of nine bucks.

00;32;52;15 - 00;32;52;28
Speaker 2
You know.

00;32;53;24 - 00;33;15;14
Speaker 1
Wendy's always wins here and it's touted often, but social media teams, they deserve a raise because that is really quick and really smart and very true. And this is where a great example of internal hurdles and bureaucracy gets in the way of actually going to market and doing what you need to do when you need to do it.

00;33;15;14 - 00;33;28;17
Speaker 1
And just think about if they had to wait 24 hours for approval to say waste of nine bucks versus as if if you're just on your phone, I got a thought and you posted it. So whoever is at Wendy's who did that, you deserve that race.

00;33;28;26 - 00;33;33;20
Speaker 2
We love the people at Wendy's. They're very, very funny. And they've been like that for years. Right. Like just said, the joke is set.

00;33;33;20 - 00;33;34;07
Speaker 1
The standard.

00;33;34;07 - 00;33;54;02
Speaker 2
You know. Yeah, the absolute minimums over there. It's very funny what they do on Twitter. But I guess it goes back to this point is that we're seeing these hot takes. We're seeing this like, oh yeah, you can do a $57 million ad campaign for nine bucks, you know, and you know what? I actually think there is a coordinated campaign between a lot of tech companies and other media folks.

00;33;54;09 - 00;34;13;10
Speaker 2
They're basically boosting it's boosting those kinds of posts. Right. Like social media manipulation is not a new thing, you know? And around the Alibaba, we're seeing this right where there's a lot of targeted, coordinated campaigns, basically folks saying, you know, you can do all this stuff for nine bucks, creating all this controversy. Basically all that's doing is raising the profile of A.I. tools.

00;34;13;10 - 00;34;32;11
Speaker 2
That's all it's doing, regardless of the controversy. It reminds me a lot, actually, of of a group that I criticized earlier this year, which basically went out and said that they had billboards all over San Francisco. They basically was saying, you know, stop hiring humans. You don't need to hire humans anymore. Just you can hire an SDR and that the ACR can go do your sales for you.

00;34;32;12 - 00;34;45;01
Speaker 2
You know, but what they're deliberately doing there is, is just injecting controversy into the media conversation. And and, you know, I think that Wendy's is just hilarious and that this is just stupid. Like basically once. That's what Wendy saying there. But but it's this is clickbait.

00;34;45;09 - 00;35;17;06
Speaker 1
It's clickbait version of this in between physical and digital component. Now, it's not just headlines and YouTube videos. All right. We had to add in an inspiration segment this week, mainly because I couldn't get Juan to shut up this whole week. He wouldn't stop talking for the past 10 to 14 days about this article about Porn Verma at BUPA wrote on customer decisioning and who gets to define customer decisioning is it the vendors?

00;35;17;13 - 00;35;22;24
Speaker 1
Is it the customers? And I know, Juan, you've got a lot of comments here and I'll hand it over to you.

00;35;22;27 - 00;35;41;24
Speaker 2
So this essay basically slapped me in the face. I thought it was one of those brilliant enterprise perspectives on air decisioning, what palindrome calls customer decisioning. In his newsletter MarTech script, this whole idea of A.I. decisioning has got it's a lot of hype around it right now. And he just gets to the core of like, what is actually going on in this space.

00;35;42;04 - 00;36;07;13
Speaker 2
So I'm going to read you four quotes about this essay that. Blew my mind. Okay. And then I'll let it speak for itself. The first one is, for most practitioners, the real questions about AI decisioning, what constitutes good decisioning, and how it should be architected and how to govern it responsibly remain unanswered. Completely agree with that. What good looks like is an unanswered question in this space, when maturity is defined by product adoption, the complexity of true decisioning capability is flatten.

00;36;07;13 - 00;36;28;27
Speaker 2
And this is what we're seeing is that technology vendors want us to believe that adopting that technology means maturity actually does. There's a lot there's a big iceberg underneath that. The third quote, the history of CDP show how vendor led definitions can stall innovation for years before the market corrects itself. And this is again, just speaking to this idea that A.I. decisioning should be led by the brands, not by the vendors.

00;36;29;04 - 00;36;43;12
Speaker 2
When the brands start to get together and think about the definition of decisioning, I think better things happen because it comes from hardware to your experience. And then my lucky last my most favorite quote is quadrants. Compare vendors. They do not define disciplines.

00;36;43;12 - 00;36;45;09
Speaker 1
I And why you like that one.

00;36;45;21 - 00;37;06;02
Speaker 2
I like that one because the quadrant is kind of the best we have around the tech category. You know, it really is. You know, it's like, all right, I order now for now. Data analytics. Yep, CRM, all these things. That is a definition of, say, an entire discipline. It's not it's actually just a marking out of the vendors in that space.

00;37;06;10 - 00;37;11;08
Speaker 2
And so very interesting comments from Paul on them. As I say, definitely go check it out. It'll be in the show notes.

00;37;11;08 - 00;37;42;01
Speaker 1
It's a good read and a good reminder. And for me, the like my brain immediately is like, let's think of solutions. And this is where I think there's both a benefit of Cavs customer advisory boards within companies, but also like in third place is outside of the specific vendor outside of your specific company. It's talking among your peers and really requesting but demanding more than anything, this new definition, the actual definition and definitions can be really hard.

00;37;42;01 - 00;38;03;19
Speaker 1
And if we continue to let vendors choose the definitions, we will continue to have confusion. We do not actually know what they mean. I mean, I can ask ten people what a CDP is and I get ten different answers. That's a problem. And speaking of CDP, we have a question of the week from someone anonymously about CBP's. So I feel this pain.

00;38;03;19 - 00;38;16;07
Speaker 1
I'm on my third CDP in seven years at my company. I failed all for very different reasons and I'm starting to doubt the value of the tech overall. What's your best argument for the CDP tech and keeping it? That's tough.

00;38;16;07 - 00;38;35;00
Speaker 2
It's a great question. We're starting to see that where there's like a second and third wave of CDP coming in. Those acquisitions happen in different ways. It could be the new exec comes in and Basil or the tech they used to make can mean that the technology just doesn't work. But it can be very frustrating being on your third customer data architecture and, you know, seven or ten years.

00;38;35;09 - 00;38;55;22
Speaker 2
So first of all, I have to say that's frustrating. I feel your pain, not fun. But what I would say is that CDP play a role in enabling marketing to understand their customers better, I would say, and be able to activate on those understandings. So I would say stick with it. It's probably more of a reflection of the company that you work at or they're on the third then perhaps the tech category itself.

00;38;55;22 - 00;39;20;12
Speaker 2
But we've seen I've worked with companies that had a CDP and have seen incredible outcomes. I've been working them some CDP. I've seen they've been working with them for 15 years. So my only advice is to go and actually ask as people come to one of our events or even have a chat with us about how other are actually be able to play the long game with various CDP because it is possible and we've literally seen companies being with one CDP before it even was listed for more than 15 years.

00;39;20;12 - 00;39;35;15
Speaker 2
And so it is possible to have a very long term relationship. I mean, actually bookends our conversation at the start of this episode about those partnerships as well. But you know, I would say that have a think about is that the reflection of the internal dynamics of the business or is it the category itself that's not working for you?

00;39;35;15 - 00;39;58;23
Speaker 2
If it's a category, then trying to pick off if that's the right business model to have the CDP, sometimes CDP agents don't work for certain business models. If it's within your internal situation. Understand the political dynamics at play that's actually causing these technology changes and what's causing that. I think spend more time analyzing your own business, understand where you're at, and then go and learn from others about how they actually are more successful with their acquisitions.

00;39;58;23 - 00;40;21;09
Speaker 1
Yeah, yeah. Everything you said also applies to other vendors. I mean, the amount of times I've seen folks go from ESP to ESP, DSP, not realizing none of them are going to work because your data is not wrinkled, is not hygienic, you just keep lifting and shifting to each new platform. And that's why every single time it's going to be a failure because there was a flaw in the original logic.

00;40;21;20 - 00;40;42;12
Speaker 1
It's like base code. If there's a flaw in the very original piece, it's going to disseminate everywhere. You have to go to the root of the matter. And to your point, is it politics? Is it outdated infrastructure? Is it actually a philosophical thing of just we like to try new things every couple of years just because we get bored and need to have a reason to exist.

00;40;42;12 - 00;40;51;11
Speaker 1
And so there's a lot of different ways to think about it, but I wouldn't limit the concept in the thought to just CDP is because it's applicable everywhere. MM.

00;40;51;29 - 00;40;56;25
Speaker 2
And Lucky last is our confession corner. Jacqueline, what is the confession going to tell us about?

00;40;56;29 - 00;41;27;21
Speaker 1
It's the place to just get it off your chest. I mean, the amount of times I have conversations with folks and they just need to tell someone because if they tell someone at their company, there's maybe consequences or they might get caught or something. And a lot of these mistakes, whether it is you sent the wrong link or you forgot to send it at the correct time, like things happen and like you can mess up an import and you have to completely set up a workflow to revert some of the changes you've made on accident.

00;41;27;21 - 00;42;03;25
Speaker 1
And not everyone needs to know that unless it because of further downstream problem. Of course you need to alert your stakeholders and fix an address, but sometimes you just got to tell someone like I massively messed up. And so, for example, I messed up a couple of weeks back. I was trying to be helpful in. Zapier and little did I know what I was updating was in a channel I was not in, so I couldn't actually kua I just didn't realize that it went to a different channel and instead of a once a day update that everyone was getting because it was broken, for what it's worth, it became a every hour update and so

00;42;03;25 - 00;42;11;04
Speaker 1
I spammed the team on accident and did not know. And so for that, that's my hopefully only confession for a little while. That's really bad.

00;42;12;04 - 00;42;33;11
Speaker 2
Yes. And I love this confession from Stew Pokey recently forgot to send an email at the right time because I forgot about it until 7 p.m. sent it then. Luckily the metrics didn't change and I think he was probably sweating because often, you know, when you're running high volume campaigns, if you missed the deadline, you know, the whole thing can fold out, the whole schedule could fold out because then customers get two emails at the same time, you know?

00;42;33;16 - 00;42;53;27
Speaker 2
And for folks that are like right on that edge where they like every campaign's going out every day and they're on that, it can be quite challenging. So thanks to Pokey for that. And that wraps us up for another episode of Making Sense of MarTech. Check out our latest episodes and our Hotseat interviews and please subscribe at the Magic Weekly dot com Fulwood Slash podcast.

00;42;54;01 - 00;43;02;14
Speaker 2
We love to see you there. We'd love to engage with you and keep your questions and your confessions coming thick and fast. Folks, we'd love to see you at the industry and steak areas.

00;43;02;17 - 00;43;40;27
Speaker 1
All right. Thanks so much.