Garbage In, Garbage Out: How Bad UTMs Wreck MAP, CRM & CDP Attribution with Dan McGaw

Hi, I'm Michael Hartman.

I'm Naomi Liu.

And I'm Mike Rizzo. And this is OpsCast, a podcast for marketing ops pros and rev ops pros created by the MoPros, the number one community for marketing operations professionals.

Tune into each episode as we chat with real professionals to help elevate you and your marketing operations career.

Hello everyone. Welcome to another episode of OpsCast brought to you by marketingops.com, powered by the MoPros. I'm your host, Michael Hartman, joined today by Mike Rizzo.

Hey everybody, welcome back. I'm here.

New and improved setup there.

We're trying something new.

At least new. Anyway, so we've got a guest with us today. We have, and we're looking forward to this, talking about this, is Dan McGaw. So Dan is an award-winning entrepreneur and speaker. He is the founder and CEO of eutm.io. He has been coined as one of the original growth hackers and is the former head of marketing at Kissmetrics. In 2015, Dan was selected be a United States Ambassador of Entrepreneurship by the United States government, a thing that I didn't even know existed, where he had the privilege to advise universities, governments, and private corporations on how to build entrepreneur ecosystems. So Dan, thanks for joining us today, late in the day for you.

Yeah, thanks for having me. So I appreciate being here.

Yeah. So I'm excited. So we're, let me just see, we'll hit the ground running here. So one of the topics And you and I talked about this, Dan, that we've covered regularly on this podcast and continues to be a hot topic. It feels like every week or two there's something on LinkedIn or in the Slack group for our marketing ops folks out there is reporting analytics, including attribution. So, and one of the things I think we've talked about that's really important is data governance, especially around UTM parameters, something you know a little bit about. And At least I often say discipline, the discipline of using them consistently is key to getting quality reporting. So what's your take on just the importance of that and why, especially as we get into this age of AI where I think we've all seen AI produce sort of fictional stuff, I guess is the best, nicest way I could put it.

Yeah.

So, your thoughts.

I think, well, I mean, I think the first thing we have to understand, like, just even with AI and AI agents and AI reporting and all that stuff, right? If you feed it bad data, it's not going to really be all that helpful. So, I mean, I think data governance is becoming even more and more important because AI is simply automation, right? And if you automate a shitty process, you're just going to have a bigger pile of trash than you would've had in the first place. So, you definitely want to make sure that you have a process set up to do that. But When we think about like naturally, why do you want to have a good data taxonomy and a good way to enforce governance is at the end of the day, you won't have good attribution. And most organizations, especially a B2B organization, right? Like you need to make sure that you can actually understand where pipeline is coming from so that way you can either spend more on what's working or spend less on what's not working. And without clean attribution, you just don't have that ability.

Yeah, Mike, you're about to say something.

Yeah, I just think, you know, attribution's the like 4-letter word, it feels like when it comes to marketing ops. Man, it's such a hot, it's such a hot button item. And I think sometimes the, the, I don't know, like on the subject of automation and AI and just like proprietary sort of walled gardens, giving you what they think is working versus, or what they say is working. I think the value of having a really strong foundation for your UTM products and your own process for analyzing your data is that I'll bet you if you dig in sometimes those proprietary walled gardens are actually lying to you. And we're starting to see a little bit of that.

Right.

Like we're, we're, we're noticing like the down funnel sort of like attribution metrics, quote unquote, uh, are not necessarily as cut and dry as the systems are saying. Um, because if you actually go unpack some of the, the traffic that you're getting, um, if you have a strong system in place to do so, you're, you're able to, um, tie back some of the transactions or at least the conversion points and the traffic sources that they're saying maybe wasn't really there before. And I know that I know I'm kind of like speaking aroundabout terms because I've only seen a handful of these little moments pop up on LinkedIn and it's not my experience. So I don't want to like steal somebody's thunder on that. But you know, Dan, I don't know. I would love to hear your thoughts on you know this idea of. of these, again, like LinkedIn, Facebook, these are all walled gardens, right? Like they have their own ad networks and ways of tracking things. Do you agree, disagree? Like, yeah, what are your thoughts on that?

Yeah, no, I mean, I think at the end of the day, you should never be using Facebook, LinkedIn, or Google Ads conversion tracking at the end of the day to be what's driving the ship. I mean, they're going to take as much credit as they possibly can, no matter what, because it's in their best interest, right? If you don't have access to that data, you can't actually understand if it's true. And you can't tell the difference between a bot on Google Ads like you can on your own stack. So like, if you want proper attribution, you really do need to have a first-party tracking world. And that's where UTMs are going to be one of the most powerful things. And I can't tell you the number of companies that I work with that they're like, hey, listen, we're spending $2 million a year on Facebook. We're spending $4 million on Google Ads. And we're just using the out-of-the-box tracking from Google Ads. We've got the GCLID on the backend and we trust all of our reporting to them. And once they move to a first-party world where they're actually storing, they're using one, you can still use Google Analytics as a first-party tracking tool because it's just going to report what you give it. Don't get me wrong, you don't have access to all the data, but if you start tracking UTMs in your Marketo forms and of course you're processing all the way through, you're leveraging products like a CDP that's consuming your data, sending it to a data warehouse, you are gonna have a much, much more realistic picture of what's actually happening compared to applying too much credit to any one of those channels. And in a lot of cases, right, we understand like with multi-touch attribution, which again is a, I would say is also another 4-letter word out there at this point.

Mm-hmm.

Right? Like luckily with multi-touch attribution though, at least you understand, hey, this one conversion that Google's taking credit for, LinkedIn's taking credit for, and Facebook's taking credit for, That was only one conversion, right? So you can have a better ability to actually improve your return on ad spend. But I think like that's just the initial few stages of that, of making sure that you have your own tracking methodology, you have your own first-party world where you're capturing your own data. I think what's interesting is like when you start tying that information to, okay, so we understand somebody came in from a Facebook ad, we can actually identify that because of course it has appropriate UTMs. But I think the second part of that is how do you make sure that you design a UTM taxonomy to ultimately create a convention, which is gonna give you additional reporting? And like going to like what we see a lot in like the B2B space, especially with marketing ops people, is like you've gotta take in campaign type, you've gotta take in the awareness channel, you gotta take in, was it a lead gen thing? You've gotta make sure you know what geo it's coming from. You've gotta understand what was the key asset type and creating that convention in the name field of a campaign. Is where you really start getting the granularity and you really start getting the actual understanding of like what type of things are happening, because you can now report on it in a Looker Studio or a Power BI report or inside of a Salesforce report, which you wouldn't be able to do in any of those walled gardens at all. They wouldn't give you that information.

Yeah, no, I mean, I've always, I've said for a long time, like the UTM, getting your UTM parameters or any other things you're using when you're putting stuff out in the market and having discipline around it. That is the glue, right, to everything else, right? It doesn't matter really. Like if you really want to understand what's working, what's not working, and you don't want to do what I inherited at one place, which is every campaign got a unique landing page, right? Even if it was exactly the same as the one that was used for the previous campaign, because that was the only way we could track it. And I was like, I put a stop to that nonsense as quickly as I could. Well, I mean, it's just, well, it's inherently—

No, it's a good—

I laugh because it's so true.

But that was, but that was the way it was.

Yeah.

Yeah. So it was a matter of like getting the right scripting on the pages and capturing it and getting, you know, but to me, it's like there's 2 pieces of this that are really important, right? One is like having the right taxonomy, having the right standards you want for the codes or campaign names, things like that. But also then those do no good if people don't actually, who are putting this stuff together to go out in the market, are not actually using it. Which I have seen.

Yeah.

Right, and it's easy to miss along the way if you're saying, I'm going to just, I'm reusing the same campaign, I'm doing something across multiple channels. So I've got your URL and I make a mistake in copying and update, I forget to update one of the parameters, right? I'm going from Facebook to Google Ads to whatever and I forget to do that. So do you, like, how do you, maybe how do you, how do you guide your customers and clients of UTM.io? Like, how do you guide them to try to address that problem? Is that even something?

The adoption piece?

Well, not just, it's not even adoption. I think there's part of it is adoption. Part of it is just people are moving so damn quick, right? That they sometimes miss that, right? So how do you, how do you kind of have the checks in place? That still allow you to move quickly, but not like still get that, the benefits downstream.

Yeah. Yeah. I mean, naturally, like, I think everybody's familiar with the UTM spreadsheet, right? So like, that is like the generic standard everybody's going to start with, but in the UTM spreadsheet, there's only so much you can actually do to kind of lock things in. And that's where we've seen, and we've seen some really cool things, right? Like we've seen people build Airtables. We've seen people do things in Coda, which like are pretty powerful, right? Like, and we've had a lot of B2B customers who have come to us after using Coda and like pushing Coda's databases and tables to the ultimate limit. And those products are great, right? And what I think in our product, which a lot of people really appreciate, is the fact that you can assign a rule ultimately to the way that somebody needs to build the UTM structure and then also be able to say, hey, if they select social as medium, they have to be able to, out of the sources, they're only gonna be allowed to select from these options. And if they select geo as North America Namer, Right? Okay, great. They're only going to have access to these campaigns, right? Or if they select EMEA, they're only going to have access to these campaigns. So being able to enforce those really, really complex things is really difficult, especially if you're like a global organization, right? Like one of our clients, Hexagon, right? Like there's 36 different brands based in 27 different countries trying to coordinate all of that and then do that in a spreadsheet or Coda. Coda Suite or something like that becomes just kind of unwieldy, right? And it just doesn't make it easy. So, and it makes it frustrating because at the end of the day, you've got a VP of marketing who wants accurate marketing attribution. You've got this poor marketing operations manager, a director of marketing operations getting yelled at by the VP, my reports suck. Then you have the marketing operations person trying to tell a social media manager, an email marketer, an advertiser, and a field marketer, hey, you keep messing up the UTMs and I'm getting in trouble. You gotta use this spreadsheet better. So it just, it really puts the MOPS person in a bad situation cuz it, it both sides, the VP and the social media manager are gonna complain, right? Which is, I think one of the things that is the un— with MOPS professionals, like they really suffer in this environment because they, they're in the middle of all the arguments and they don't get enough credit. They don't get enough empathy. They just gotta make it work. Right. And Yeah. so that's where I think there's cool tools that make it easier. Obviously, I'll shamelessly plug UTM.io in there. There's other ways to do it, like I mentioned, Coda and things like that, and spreadsheets. There's definitely— I mean, I've seen some shit. So I don't know, can I swear on this podcast?

You definitely can.

Okay, cool. Sorry.

I mean, look, we host the No Bullshit Demos for a reason. Like, it's all part of it.

It's a very good point. Yeah, no, it makes a ton of sense.

Yeah, I've seen some nonsense too. I have to agree. I think it's I just think it's a really clear case, you know, for our listeners, if you want to take a clip of this show right here, right now and say, hey, this is the clear case for why bespoke tools are, you know, better than a spreadsheet.

Yeah.

Because you think, oh, can't I just do that in a spreadsheet? And it's like, well, yeah, but at some point, It's actually going to break.

And I was going to say, I don't know, I can, I can tell you probably dozens of times where we've had a shared spreadsheet where we've had to go back and undo someone overriding those complicated formulas. Right. And yeah, because they weren't protected and someone was moving fast and screwed it up.

Yep.

All those concatenation rules that you have set up, the macros, and then you get some people who get like super advanced, they throw some Visual Basic on there and they just try to make it so super, super complex. Like, yeah, so it can get difficult. I mean, it's similar to like a product like Lean Data, right? Like you didn't know you needed it, but once you had it, you can't live without it because it just does so much that enables you to run so much better. So there's definitely tools that are going to make your world better. But I think like what we try to understand at UTM.io is that we have to remember at the end of the day, the person who's building the link, right? They do not care about your marketing attribution. They just wanna get their job done. They just wanna post on LinkedIn. They just wanna get the ad live. They just wanna set up their booth at the field marketing event. So like there, in some cases, some people make them like as if they're the enemy because they're the one messing up the data. But it's like they just wanna get their job done.

Yeah.

And that spreadsheet is not always the most easy way to go about doing it. But I will say I love spreadsheets. Like I use spreadsheets for everything, build a process. Once I've designed a process, okay, great. Let's figure out the tool. I just was doing a webinar maybe about 2 and a half months ago, and everybody was like, oh, we buy tools and it solves problems. And I'm like, no, no, no, that's not how this works. Like, you design a process to solve the problem, start with a spreadsheet, build the process, and then say, okay, we understand what the hell we're doing now. Now let's go find the tool that can replace the spreadsheet, because that spreadsheet's not gonna ultimately survive because there's so many error-prone ways. Like, I mean, even our pipeline at my other company, McGaw, Our deal pipeline with MedPic is currently being done in a spreadsheet. Now, do my sales reps like me for this? No, they want to go back to Salesforce. They want to make it so that everything's in Salesforce, but we're transforming our sales process. So we're using everything in a spreadsheet so we can figure out every single step, every single category. And then once that's done, I'm then going to go hire one of the freelancers in Marketing Ops Community and be like, can you build this in HubSpot for us so it works in Salesforce?

That's a good way to do it. We did the same. We started everything like our sales process when we're trying to engage with sponsors and things like that. We were like, all right, let's, let's just lay this all out in columns first and start tracking, hey, how many touches did you actually try today and all that kind of stuff. And then eventually bring that over. And I think that's true for, you know, running a business in general, let alone a department.

Right.

Um, I think it's important.

I think most people do that. They go get a tool.

Yeah. I think there's some value in living with those constraints, right? Because the tool is going to, if you're using a spreadsheet, it's going to limit you on what you can do.

Yeah. And it's, it's, it's a fascinating thing to think about too, just because, you know, you're, I don't know. I'm sure maybe this is true for you, Dan. Uh, certainly is for me, but I turn to AI to rapidly iterate on a lot of my thoughts right now, right? So, my second brain is this AI assistant. And when you're forced into constraints, you actually have to think. You have to work a little harder to figure out what your process is. And so, I think it's an important step to doing all this stuff. Anyway, it's a good subject to talk about.

Yeah, so it's interesting because you brought up this, to me, this is an often forgotten part for a lot of marketing ops folks. You said they think of the people who are actually deploying this stuff as the enemy, which I see a lot of that too, which I think is a wrong way to think about it, right? And I think you want it like, if you go into it with that, it's going to be hard to get them to listen to you when you are trying to get them to change their behavior. So maybe one of the argument— one of the ways to do that in my view would be like, help them understand what the downstream impacts are. So when you see these broken processes and inconsistent ways of using things like UTM parameters, like what do you see as like the downstream impacts that would impact those people who are the problem child? Yeah.

Well, I mean, the biggest thing that I always try to help people understand, especially like if you're in a MOPS role or even any type of role and you're having to sell something downstream to get them to take advantage of it, there's a great thing called the Hero's Journey framework by Joseph Campbell, and try to figure out how you make them in a story, right? You understand that you need to make this person the hero, which means they need to have a reward at the end of the story, and they're trying to overcome a challenge, and your job is simply just to be the guide that helps them get through that, right? So we always try to make sure that we understand what that reward is, and a lot of times that reward for a social media manager, hey, listen, you're gonna be able to show that the work you're doing is driving pipeline and you're either gonna get a raise, you're gonna get a bonus, you might get a promotion, but first you're most likely to get more budget, which is a lot of things what these teams really want is they want more budget so they can do more of what they're doing. So even like if you think about a field marketing manager, right? Like their job is to drive pipeline through events, but because they don't have effective means of tracking all of those folks at an event, They need tools that ultimately help them do that. So you need to be able to clearly communicate to them like, listen, I understand that like, hey, your boss is probably on your ass in regards to justifying how much we're spending on this event. But what you're now able to do is make sure that you have clean QR codes, which you're going to be able to use at the event. All of your links will be properly tagged. That way I can get Susan, the VP of marketing, to see that you actually are driving 20% more pipeline than what is not being captured, because no matter what you're going to do, you can't track everything, right? We still have dark social.

Yeah.

We still have word of mouth. Like, there are still those things, but there are ways to make it so that you can increase the amount of touchpoints that are trackable. And that's where you really can try to sell that story and make them the hero ultimately, where they have a reward of more budget, more accolades, more credit for the good work that they're doing. And I don't know about you, but like, I, as a CEO, I give my team good job, like when they did something cool, but when they actually drive pipeline and I can measure it to them, I give them money. And I think for a lot of people, money really matters.

Yeah, no, I think there's something— it's interesting because I love the idea that you're using— I forget, I already forgot what you said it's called, but that storytelling piece, right? And I think—

Yeah, Hero's Journey Framework.

It's interesting. What's interesting to me is there's a little bit of a side tangent here, but But this is something that I think is a misnomer. Like, everybody wants to be, quote, data-driven, right? And I think there's value. Like, I think being data-driven is important, but if that's all you do and you don't think about the storytelling part of it, you're missing the chance to really impact, like, really get somebody to listen to the story part of it. So they may not, like, the data may mean nothing to them until you can put it in context, give them the story. Yeah, that's how, that's how humans communicate in general, right? It's based on stories. So like, that makes total sense to me. Yeah, I love that. Um, um, go ahead, Mike.

I'm gonna ask this random, random question for, for you, Dan. The, the community, uh, sort of gamification person in me is, is— I've had this idea before, and I'm wondering if you've ever tried it or have a potential destination in your product for it. Like, it'd be great if we could just reward people. Like you talked about rewarding people, right? Like, hey, the outcome at the end of the hero's journey is potentially a raise or more budget or whatever. But along the way, it'd be really fun to gamify. Like, who did the most— who created the most trackable links?

Right.

Departmentally, as an individual, whatever it is. Have you ever heard of anybody doing that as a method to drive adoption?

Yeah. So this is actually something we're building into the product later this year as a leaderboard. We've had this request multiple times because they want to get more adoption and they feel by showing a leaderboard of, hey, we have a leaderboard who created the most links, which isn't always the fairest thing, right? Because your social media manager is going to create more links than your field marketing manager, but then also a Dialing in who has the most clicks and things like that, who has the most accurate links and different things like that. So yeah, we are looking at leaderboard and designing that for later this year to try to gamify it. We have tested some interesting things where you could, like, as a— in the link table, you can basically give somebody a thumbs up or thumbs down on their ultimate link because you can give them feedback.

Mm-hmm.

We tested that for a little while, which it was so-so. Obviously there wasn't a lot of thumbs downs, but you could give somebody kudos and credits That was kind of cool. But I think a leaderboard is going to be naturally what's going to be most helpful. What we want to try to do is attach it to pipeline in Salesforce or HubSpot. So whoever's links are driving the most pipeline, you can actually be able to see that in the product. So, but yeah, I think gamification really helps. I think the product, which is interesting, I don't know if anybody uses Fathom. I recently am testing Fathom Video for recording my meetings and stuff like that. You get points for everything in that product, right? So like recording your first meeting, you get like 25 points. I think they maybe are like skewed way too far because I don't know, maybe it's working for them. For me, I was like, I went to my profile page, I got 25 points. I signed out, I got 10 points. It was like, wait a second, I'm getting too many points here. So gamification does work though.

Yeah. Yeah. But you know, you could tell right out of the gate with that product, like they built it with the intention of like Like, you know, there's products that throw stuff in later and it sort of feels like an afterthought. And then you can, you could tell Fathom was like, hey, we're just going to start with gamification is like the way that we're going to try to get people to adopt this thing. And that's like a whole freemium model. It feels like they started there. I could be wrong about that, obviously. But I don't know.

I'm the late adopter to Fathom for sure.

Yeah, same. But that's cool. All right, cool. That's just where my brain goes. Like when you think about running You know, in a community, for example, we're often, you know, we're not using a proper community forum product when most of us are hanging out in Slack. I'd love for everybody to go adopt our forum product and we will see more of that in the coming months. However, you know, the benefits of those forum products are that you get to find who's sort of most topically relevant or answers the most questions around a particular thing or is most influential and all that other stuff. And so my brain often, you know, when I think about products like yours, utm.io, when you're trying to get people to find the right resources or perform the right action, you want to reward them for those things. And so it's interesting, you know, we'll have to come back around and look at it in real time once you guys get that out in the market. But thanks for sharing. I had no idea that was a— for my listeners, I had no idea if that was on his roadmap or not. I literally just Randomly thought.

No, I think, I think we need to add a confetti explosion when somebody creates a link because that we don't do that. So I think we should add those things. I think Trello recently rolled that out when you like hit a due date and you get confetti. That's a super easy add. That's not gamification, but at least it's some sort of celebratory thing.

I mean, the platform we're recording on right now does something like, or did, I think it's gone away.

They did.

They took it away.

I was so sad.

Yeah. Yeah. So I miss it.

It's like your recording gets uploaded, you got confetti. I was like, this is—

I love this.

I do remember that from my podcast.

Unsolicited feedback on that idea of gamification. I'd not— I would be cautious about doing a gamification based on just the number of URLs posted because previous life work, I did something with customer support and, you know, they're like closing number of cases was one of the metrics. People game that system pretty easily, right? They have their friends call or, you know, so they drive the metric up. But, um, I think the idea is interesting.

So it's funny because you bring that up in regards to like support tickets. I know Zappos in the book Delivering Happiness, they had like the typical metric of success in like a call support or call center is like, how short can you be on the call? And they rewarded people for how long they could be on the phone call. So you really have to be careful with those kind of those offers and incentivizing people, which I think Zappos, it was like a 19-hour phone call or something like that. Like, people would call them and be like, can you order me pizza? And just have phone calls. So like, no matter what, people will game any system you can create at some point. Hopefully your values prevent that from happening, but it can definitely be—

Right. I mean, yeah. I mean, yes, there's going to be people who take advantage of the system no matter what. Agreed. But I think that's Hopefully, they're outliers, and that's what you want. But I think the idea of somehow measuring it and sharing it and making it public, like, I think there's some value in that.

Yeah, yeah, for sure. I wanted to come back to something you mentioned, a topic that's somewhat near and dear to my heart, having not been able to really adopt much in this area, CDPs. So I don't know, I feel like, I feel like it's starting to become a lot more of the norm. I certainly at the enterprise level, we see a lot more of the data lake, data warehouse, CDP sort of discussion come up, particularly in the B2B side of things though. And that's where I want to sort of attack this question from. Is it unnecessary? Like, at what point does it become a necessity to start thinking about leveraging a database like that as you think about analytics and attribution and, you know, utilizing tools like UTM.io to pass the right data into the right places. You know, CDPs come up a lot in B2C, but B2B, it doesn't feel like it comes up as often.

Are you—

it's sort of a 2-part question. I already asked the first part. But the second part is like, are we seeing more of it come in earlier in B2B conversations? Rather than at the sort of enterprise level. What are you seeing in the market right now as it relates to that, Dan?

Yeah, I mean, I think at the end of the day, B2B in general, right? Like if you're B2B SaaS, like a CDP is primarily a requirement, right? Because you need to be able to have a single view of a customer and then also be able to have a view of an account and being able to capture that data quickly and then as well as distribute that data. Because we have to realize like there's not like the CDP landscape is so confusing now. Thank you, Hightouch, for coming out with composable versus non-composable and then Nobody knows what composable is. And now there's hybrid CDPs, there's customer experience CDPs. Like, at the end of the day, all a CDP is, is an ability to be able to accept data, relationate it with an identity, and then be able to distribute that data and then be able to build audiences. However, a CDP has taken on so many different lenses now at this point, it can be really, really confusing, whether it's warehouse first, which everybody's warehouse first for the most part now. So like, that's causing a whole bunch of confusion in the marketplace. But if you're like a company like a B2B SaaS company, it's nearly a requirement because you have to be able to track that user as they go through marketing, sales, product, and customer success. And that CDP is gonna be one of the things that really enables you to stitch that real-time clickstream activity to what's happening in Marketo to then be able to equate all of that back into a warehouse or into a product like Amplitude. Now, if you're a B2B company that of course is a little bit more traditional, we at my other company, which is a CDP consulting firm, We work with a very, very large company that rents construction equipment. So like their sales reps are writing on yellow legal pads, right? And then coming back to the facility and typing into a CRM. They have a CDP and they're able to really do a lot of cool stuff because they have all these fragmented identities because of the way these sales reps work. But in that type of industry, sometimes, you know, you can get away with just having a Marketo or a Salesforce or a HubSpot. But those tools at the end of the day are not really real-time clickstream products. Don't get me wrong, I understand that Marketo and as well as HubSpot track events. So I get it, they have that ability, but they are basically glorified databases which are meant to, I have a first name, put something in it. I have a company field, put something in it. You have to set something up for it to receive that, where a CDP is send me whatever you have and I'm going to automatically ingest it and then create a new row and a table for it. So I think the mental model is different. I will say, like, I think there's a lot of benefits But marketing operations and revenue operations professionals, because the mental model for them is really, really field name, field value, it does make it really, really hard to make the transition to the way that a CDP tracks because you're now not just dealing with a field, which is an identity value, you're also dealing with the actions at the same time. So typically you have an event, you also have an identify call. The way those 2 things work are really, really hard. And usually you have to have a little bit of a developer mindset to understand the way those things work. So I think it's harder in the B2B environment for a MOPS or RevOps person to understand that. I think it's just, it takes time to overcome kind of that paradigm shift that you have in your head. So do you need a CDP? I highly recommend it because it's gonna make it better for your ultimate customer data orchestration and then optimizing that customer experience. However, you also have to make sure that you're ready for it because they're expensive as hell. And if you don't take real advantage of it, it can be really, really painful.

Yeah.

Yeah, yeah. Go ahead.

I didn't say anything about UTMs on that. So I do apologize there. I think the one thing on UTMs—

I was just gonna ask you about that.

Yeah. The thing about UTMs though, is they work with almost every single marketing tool and almost every tool out there, right? So when you have a CDP, it's automatically going to ingest anything that you have as a query parameter for your UTMs. So that's great. It's gonna automatically track that. And then if you're sending it downstream, it's automatically gonna work with that, whether it's Google, whether it's Mixpanel, whatever, it's automatically gonna work with those tools. So it does give you a lot more flexibility on being able to have that attribution. Because where I see the biggest mistakes in regards to UTMs is people capture it in the Marketo form fields, they then pass it to Marketo, they then add it to a campaign, they then send it to Salesforce. That's great because that helps you understand your pipeline. But when you're doing true like experimentation, conversion rate optimization, analysis of your funnels on your websites, that data is not going to help you in Marketo or Salesforce. You really have to be leveraging an analytics product. And that's where the CDP makes it easy because it will send it to both of those places compared to having set up 2 separate configurations.

Yeah. So punchline, UTMs are foundational to, you know, once again, it's the glue to establishing more analytics.

Yeah, right.

Yeah, they're the oldest tracking we have. And it's interesting because you see providers, and don't get me wrong, like Adobe has their own tracking system, which is basically UTM, they just call it their own. Piwik, they have their own UTMs that they use for their systems. And it always is interesting to me why these providers come up with their own naming structures, their own query parameters that they add to a URL that only work with its tool in that nature, because everybody works with UTMs. It's the oldest tracking assignment that we've had. I mean, they've been around since the '90s. So it's like, why try to do something else?

Yeah. No, I mean, it's kind of limiting. There's— and these are all flawed, right? So in some ways, it's, you know, whether it's how it's deployed by some system outside of your control and everything else. But what— so we talked a little bit about like, how do you sell the idea of like getting consistent and disciplined about getting UTMs out there correctly. What are some, like, can you think of some examples of clients where, like, because they didn't have that, where, like, what kinds of problems did they run into when they were inconsistent or incomplete or just weren't doing that kind of stuff?

Yeah. I mean, the most, like, we have a lot of companies that come to us from field marketing teams that just don't have any ability to track almost anything, right? Those field marketing teams are in the wild trying to do stuff and they're not able to actually attribute their revenue ultimately to pipeline or their pipeline dollars. They're able to track the leads that get scanned, right? But they're not able to track the rest of the things. And that's where systems like using vanity URLs that are attached to UTMs, using QR codes that are going to be attached to UTMs, obviously the emails or whatever you're sending out there, being able to better attribute that. So we've seen organizations that are spending $500,000 to $1 million a year on field marketing. They're only able to attribute what the sales rep was actually able to collect. They were not able to attribute any of the sponsorship that they did on the side of the hotel building or on the sponsorship package or the lunches that they were sponsoring. Because the reps are never talking to them. So by being able to do that, they're able to attribute around 30% more of their marketing budget to actually driving pipeline and able to get more credit on that simply because they didn't have a means to be able to connect those dots. And I think that's where organizations need to get creative in regards to how they get people to ultimately click back to the website and then be able to get tracked. Vanity URLs are a great example. QR codes obviously are really popular. And those are some easy things. But I will say, like, even with those types of things, right? Like, I think companies need to definitely have a, how did you hear about us? Right? Because people hear from you on 6 different channels.

Yeah.

The thing that I see is the failure on the, how did you hear about us? Is people give you, you have to select from 1 of 9 options. Don't do that. Just make it so that they can type in an open text field and people will type. We found out at utm.io that over the last 12 months, our number 2 driver of users signing up is ChatGPT. Like, it blew me away that ChatGPT is now our second largest acquisition channel. But we wouldn't have known all the different AI tools that were coming in because we've seen Gemini, we've seen Notebook, we've seen like all these different tools that are now causing us, or AI tools that are causing us to get users. So with the UTMs, don't get me wrong, but still try to add a, how did you hear about us, open text field somewhere. Because people are going to mention, hey, I heard about you on the OpsCast, which I have no way to track anybody who's listening to this podcast if they actually signed up unless I had a vanity URL or if they actually wrote in that field. So the world's not perfect.

Okay, so you're here to hear first, folks, when you go to utm.io, say you heard about it from OpsCast.

That's right. It was interesting, Dan, I think, because there is this ongoing debate. Do you ask people where you heard from us or do you use UTM parameters? And like, to me, you don't need to choose. Like, both are valuable.

You do both.

Yeah.

You do both. Yeah, we do both.

And you should. Like, I think there's really— and I love the pushback I think people are going to give you on the don't use a picklist is twofold, right? Well, how, you know, how are we gonna know? Like, why do we give people all these options? Like, why don't we just give them limited options so we can do that? Second is, if we leave it open, it's gonna be a lot harder to analyze. So how do you, like, how do you address those kinds of comments?

Well, I, I'll tell a funny story. So we worked with an extremely large, um, uh, it's a company called Forks Over Knives. They were a very popular documentary years ago, and they are a vegan documentary company and they have absurd amount of traffic. And, we were asking a survey question in regards to— we had a 16-question survey people would fill out and then everybody would answer which state they were from. We had 75% of our users from Alabama, which is not where vegans live. I'm not trying to pick on Alabama, but it is not a vegan state. And, most of our users we could prove were in California. But, why did they select Alabama? Well, it's because it was the first option and they did not care about actually answering the question. And that's the problem you have with picklists, right? Or even radio buttons. They just go for the first one. So yeah, it makes it really, really difficult, but there's ways to solve that problem. One, you can build an AI bot agent that ultimately will classify those things for you in real time. So that's not crazy. Don't get me wrong, it's not gonna be like easy. I saw somebody building something like that with Trey.io recently. But there's also ways to be able to reformat that data. So when we've had problems with this in the past, we built— I don't know if it's still live on our McGaw site, but if you were to Google, I think it's McGaw keyword phrase parser, you can basically drop in 10,000 rows of open text data and it will reclassify that data and tell you what actually people did. So there's definitely ways to fix it, right? And there's definitely ways to reformat the data. But that extra effort that you're going to get by, or that extra effort you have to put in to reformat the data, it's going to wind up giving you much better odds of understanding the data compared to just thinking all of your customers are in Alabama because it was the first picklist.

Right.

And there's obviously tricks that you do with picklists where you can randomize your picklist on every single session. Still, you're just winding up with more random data, which isn't necessarily going to help you. So that's where it gets difficult. But I think that's where when you think about UTMs obviously give you one leg to stand on, right? You now have this, how did you hear about us? Helps out a little bit. But that's where leveraging smart analytics, going back to the CDP where it's automatically gonna choose where the session started from. Was it a referral? Was it dark social? Was it direct? You'll be able to do that. And when UTM launches its dark social tracking, right, you're gonna be able to like even tie some of those things closer together because I think dark social is the thing that we're all still trying to like really understand and try to get that going. How do you attribute all of these different things together, especially when Michael is sharing something with Mike? How do we connect those 2 influencers together to attribute where somebody came from? So it's not gonna be perfect, but you at least need to know what angle to point your ship, right?

Mm-hmm.

And I think that's really what people— why have I been successful at growing companies? It's because I don't need an exact answer. I just need to know whether I'm going north or south. And if I know I'm going in the right direction, at least I know how to grow the company.

Yeah, yeah, being directionally accurate is super, super important. And I think sort of leans towards that mentality of like, I don't know, fail fast, fail often kind of a thing. Like, if you're just learning quickly, you can kind of like get the right direction and keep going, and you'll just keep refining. What a lot of what you shared, Dan, and, um, just reminds me, uh, of, you know, of course It's probably just because I have a bias for it now, but I've been pushing really hard for people to think of their go-to-market tech stack as a product. And, you know, what you're talking about at the end of the day is a feature, right? Like you just talked about building a very unique and bespoke feature that comes with a little bit of experience when you have seen a couple of seen around a couple corners on some challenges, and you're aware of this. And now, you know, your HubSpot and your Marketo and your insert next, you know, marketing technology solution here, they're not built to do this stuff out of the box. And what you said is you need to go build a feature to be able to support a particular go-to-market motion. And that feature in this case is, you know, creating a taxonomy and maybe an AI agent or a script that you can query your GPT API to say, hey, where did this person come from? They just filled out the, you know, how did you hear about us, you know, empty text field. And now I need to standardize that data. We did a version of that at Marketing Ops. We just branched, we took, we looked across all of the information that came in on our field, right? The, the how'd you hear about us? And we said, all right, GPT, tell us generally, like some categories here.

Yeah.

Now we're looking for keywords. We created different branching logic so that if it contains a particular keyword in HubSpot, it does this. This was Before AI became even more accessible. And now I could literally just turn that whole branch off and say, stop looking for specific keywords and just let the AI do it for me.

Right.

But again, that's something that we had to go build in order to pursue the next sort of phase of, are we directionally accurate in where we want to go now?

Right.

Because when— When we signed up for HubSpot, it didn't come with this stuff. So, all of this is to say, I love what you're saying. Establishing these sort of boundaries and these rules and these systems and these processes, to me, means that you're ultimately like building out products. But I'd love—I know this is a tangent—but I'd love to hear your thoughts on like. How do you feel about that?

Like with, you know, maybe, maybe we can frame it in this, this way, because I think we're going to have to wrap it up is like how, like if someone's listening and maybe they're newer in their career, this is new stuff for them, like, or they're in a place where they see this process as broke, like how would you start to build out this as a product within your, your tech stack or your, your environment? Yeah.

And just to be clear, like when we think about like this, are we talking about in regards to the regards to classifying, how did you hear about us? And like, Or just ultimately, like, how do you push this stuff forward? Because like, I think one, you should be in most companies, the area where they struggle the most is being able to actually build something unique, which is not currently done, whether that be working with your development team to build a microservice or anything that's going to do these things, right? Like going back to the, we couldn't classify this data, so we built a keyword parser, right? And a phrase parser. With AI, we're able to build products really, really fast. I literally saw I can't remember the gentleman's name who's building a competitor to us, and he's doing it entirely using AI bots. And I was like, I think that's rad. Like, keep doing that. So I think naturally you need to be building product and tools to be able to better serve yourself. But with AI, it just makes it a little too easy where there's a lot of random tools there. There's going to be a big trash dump of Wall-Es of AI tools in the next 12 months that we're just going to have a ton of these crap products that we tried to build, we tried to solve something, and are going to die. 100%. So, there's a balance that you have to find on that because in a company, when you build your own product, and I can speak specifically for ourselves, we built our own meeting agenda tool because all the meeting agenda tools we used kept getting acquired and then getting shut down. So, we're like, forget it. We're just going to build our own. But we could never keep up on the innovation spectrum like a product like Fellow, fellow.app, which we use for all of our meeting agendas and meeting tools. Their innovation is way faster than ours. So you, when you build a product, you have to realize somebody's gonna outpace you in innovation eventually. So I don't know if that answers the question, Mike or Michael.

No, I, I, so I was actually even thinking like, 'cause we talked about it at the very beginning, right? Like having, putting tools and process tools and technology in place on something that's already broken is not gonna get you make, it's just gonna make that bad process faster. So I'm like, yeah, how do you do, like, what do you do for, what are the first steps they should be taking before maybe even investing in technology or changing technology that you already have? It's kind of where I was going.

I would always revert back to a Google Sheet. I'm sorry, there's not a time in my life that a Google Sheet hasn't solved 85% of the problem. And then once we have that Google Sheet and then build off of there, and what I think is really cool even with the Google Sheets now is like with some of the stuff we have an advantage of with the extensions that we're integrating into Google Sheets from AI, being able to leverage that data in many, many other places has been a lot of But I'm, my first go-to is always how do we use a Google Sheet for everything? But that's, I sometimes I feel like I'm old because I use a Google Sheet for everything.

No, no, you're not old.

I use a Google Sheet for everything too.

So you'd like define your taxonomy and all that kind of stuff in there and then build out maybe even something formula-based and then go from there?

100%. Like especially if you're speaking in regards to UTMs or any of that type of stuff, first you're gonna wanna start with using a spreadsheet, doing concatenation of multiple fields. Multiple dropdowns. I mean, Google Sheets do give you the ability to kind of lock some things down. If you want to get fancy, do it in an Airtable. I've seen some really cool things in Airtables. We actually have a, like, Notion databases. They do not work well for UTMs. I want to be clear. Like, I've attempted to build some things in there, but anything that's a simple spreadsheet is going to make it a lot easier to be able to be flexible so you can stalk around and find out, which is like the most popular saying in Florida, which I'm happy is starting to catch on all over the country in America now. But yeah, get a spreadsheet, fuck around, find out, keep iterating, keep iterating. And then when you feel like, hey, our attribution reporting is getting better, it's not perfect because I still have some nuances. That's when you should really look at a tool. So, because I never tell anybody like, hey, if you have no UTM experience, you probably don't want to start with UTM.io, right? You definitely don't want to start with us. Like, go get a spreadsheet. That's going to be way easier to use before you come to us. Which, don't get me wrong, we replace the spreadsheet with no problem. But for most customers, most companies who have no idea how to use UTMs, a spreadsheet's gonna be way, way, way easier.

Yeah.

Yeah. Makes sense. Dan, I wish we had more time, but we're gonna have to wrap it up here. So if folks want to learn more about this, get in touch with you, learn more about UTM.io, what's the best way for them to do that?

Yeah, for sure. Definitely go feel free to check out UTM.io. We do offer all the people at marketingops.com a 30% lifetime discount. discount, which, Mike, we found out existed the other day unbeknownst to us knowing that still existed. But that's there. But at the same time, if you want to check out any of my stuff, just go to LinkedIn, look up Daniel McGaw. Happy to be able to chat with you online through LinkedIn. That's my most active network. I don't use X. Once they switched from Twitter, I got confused. I couldn't find the app anymore. So just on LinkedIn now.

Yeah, you and me both. What?

Hey, you have any speaking gigs coming up, I'm sorry, what'd you say, Mike?

I said, do you have any speaking gigs coming up by any chance?

I speak in London in 2 weeks at some exit event, and then I don't know, I think I'm gonna make a cameo at this event called Spring Fling.

I might do a cameo there.

That'd be great. I do hope you can make it. I might make it. I think I might be speaking at Twilio Signals event in May as well. But yeah, no, I'm trying to lay low. I moved to Barcelona this past year, so I now live in Barcelona. So, it's finally starting to get dark on this side of the world. So, I'm doing a little bit less speaking because I'm trying to spend more time with my family traveling Europe and getting to meet all the interesting people across Europe. So, it's been a ton of fun.

That's awesome.

Fantastic. Dan, tons of fun. Seems like to me an old topic, but I've learned something new, I think, even through this. So, appreciate it, Mike, as always. Thank you. And to our listeners, thanks to you for continuing to support us and suggesting great topics and guests. If you have suggestions for topics or guests or want to be a guest, reach out to Naomi, Mike, or me. We'd be happy to work with you. Till next time, bye everybody.

See ya.