Making Smarter Marketing Decisions with Clickstream Data

Will Francis (00:06)
Welcome to Ahead of the Game, a podcast brought to you by the Digital Marketing Institute. I'm your host, Will Francis, and today I'll be talking to Eli Goodman. He's the CEO and founder of Datos, and they are a company that help you understand

your customers and your data. They specifically specialize in clickstream data. And we're gonna talk to Eli about what that is, how it's useful and his journey as a founder of a company that's since been acquired by a well-known tech firm. Eli, welcome to the podcast. It's really good to have you.

Eli Goodman (00:37)
Well, it's great to be here. Thank you for having me.

Will Francis (00:40)
Yeah, so let's just start by setting the scene. What is Datos?

Eli Goodman (00:48)
At its core, we're a clickstream data provider, founded back in 2020. And really in the end, there's so many different ways you feel like you could describe your business or what the uses are. But the foundational element is really that we're in the ingredients business, right? I sell sugar to other people that bake cakes, right? That's it.

Will Francis (01:07)
I love that. Let's talk a bit more about that sugar. Okay, so what data do you gather and how specifically is that useful to me as a digital marketer?

Eli Goodman (01:18)
So the core data that we have been gathering for many years, we're in the clickstream data business, which is as it sounds, the stream of clicks when people are on a desktop or mobile web browser, all of that data is anonymized, but think about it, it's really a collection of breadcrumbs that help you better understand at a mass scale, like what's happening, right? As far as pattern recognition, how and why people are doing things and whatnot. But we also have branched into additional types of alternative data.

We recently launched into e-commerce and transaction data. We have some partnerships related to in-app data, which is also can be very different from how people interact with the internet, right? Within apps and clickstreams. So we continue to add more arrows to the quiver, if you will, for our clients to be able to make these.

Will Francis (02:07)
And are you talking about clicks coming into a website as well as through it and out of it?

Eli Goodman (02:13)
Yeah, so think about it. It's even broader than that, right? Like that is one small subset of it. But clickstream as a definition is we passively observe and objectively our gathering of data. You know, we have an opt-in panel of around 15 million people worldwide spread out across a couple hundred countries. So imagine it's not just about web analytics data, which everybody is pretty familiar with, right? Your Google Analytics or Adobe Analytics of what's going on in your own website.

Clickstream data as a service offering tends to be that type of click data on everybody's website, right? Every search term people use, product they buy, video they watch, content they consume, and stitching all of that data together to be able to look at what is like a longitudinal behavioral analysis, right? So it's not just moments in time, but connecting the dots.

Will Francis (03:08)
So being able to really understand people's digital life and the flows and patterns that make up the sort of patchwork of that essentially.

Eli Goodman (03:16)
Yeah, that is correct. You know, and on our particular angle for it is that, and this goes back several years, you know, when we founded the company and we're thinking about, you want to be able to set up a data company that is perpetually delivering data, right? So you're always trying to think about future proofing things and digital marketers, I think we all know, have been dealing with what is a very unpredictable landscape related to cookies and cookie deprecation or GDPR versus what's happening in the United States and so on and so forth.

So, from our perspective, we actually exist in a 100 % anonymized universe. And actually one of our taglines has always been, we exist for intelligence, not for targeting, right? So the basic idea there, and I go back to pattern recognition and where clickstream and other types of alternative data play, it's not just about like matching up and enriching known datasets, but using the information at, again, like an anonymized objective level to identify broader

behavioral patterns, right? And we'll certainly talk about use cases and whatnot as we go, but it's a, from my perspective, the future of things, right? That like PII and individual targeting just becomes more and more precarious, if you will. So just, you know, really trying to set up and think of different ways that marketers can use data in order to be successful.

Will Francis (04:40)
What's a typical on the ground decision that a marketer might make based on the kind of stuff they get from Datos?

Eli Goodman (04:49)
Well, the first one to me is, let me like right size the opportunity, right? When you think about it is most of the tools that exist out there for digital marketers tend to be results oriented as in like you jump in, you're running reports, it's delivering you what has happened. You set up campaigns and what's happening with campaigns and whatnot. But some of the problems that exist within that in, let's say the classic way it's been done

is things can be limited when it comes to it's sort of a pretty linear fashion, right? Like I ran a search or, you know, I showed up on a search engine result page. I clicked on something, you know, I then what was my next step in that journey and so on and so forth. And eventually things again, like the cookies or the different types of MAIDs or AIDs or whatever is being used, IDFA.

You know, and those, by the way, are all the ad acronyms. I'm sure everybody here probably understands those, right? Like AAID and Android advertising ID versus an IDFA, which is what Apple uses and whatnot. But anyway, so what I'm getting at here is clickstream data, right? When you separate the trackable identifiers from it, it now becomes a way to think about nonlinear impact of campaigns, right? So imagine you're exposed to ads or you watch content or

you know, somebody's how to video or whatever, SponCon (sponsored content) on Instagram, maybe you didn't take an immediate step right then and click on it or go directly to whatever that page. You just don't, right? There's the issue.

Will Francis (06:17)
Well, you almost certainly didn't. That's the bloody problem with content marketing, isn't it? You can very rarely

draw a line between those two things.

Eli Goodman (06:27)
That is correct. So that's exactly when you think about clickstream. We talk about what is the nonlinear connection between things because it is I talk about longitudinal. You see what happens over time so you could start to gather up and say, well, everybody that watched this YouTube video that was related to a particular product like let's pretend it was around an automobile of some kind. Did they eventually show up and start shopping for that particular ID or that particular car later, right? The brand of it

the generic version of it, but they eventually type in like BMW xyz and their zip code, right? Which is certainly pretty down-funnel, meaning I am thinking about buying one of these or test driving it.

Will Francis (07:07)
I need one now.

it's interesting that isn't it because it is a perennial problem so you're you're trying to solve that and give people a way of proving that the non-obvious stuff worked Is that partly about helping marketers get buy-in for up-the-funnel marketing internally?

Eli Goodman (07:27)
So there are, listen, the use cases, the permutations of this are somewhat unlimited. But I think your point there is go back to early internet, right? Everything was direct response, right? That was pretty much the nature of it, right? Buy now, click here, so on and so forth. And then that direct type of tracking. Over the course of the past couple of decades, right, the upper funnel branding and awareness type of marketing, the classic brand metrics, if you will, when you go back, I'm old enough to remember pre-internet and how things were done, right? As it related to

advertising and effectiveness and whatnot. So if you think about all of that, yes, there has been a movement back towards that upper funnel. How do you quantify and to be frank, qualify that spend. And that goes back to not just getting what Google is telling you or Facebook or Meta is telling you and so on and so forth, but your ability to really connect the dots between, well, we took all of these actions days, weeks, months ago.

And we can actually quantify the lift, the measurement, the attribution of what happened later, even if it wasn't a direct connection. Now, I want to be clear, right? When you start getting into the data of it for the marketing scientists out there, a lot of it could be correlation versus causation, right? There are any number of factors that people that impact, like you saw ads on television or your friend bought the product and is telling you about it. So it's not that clickstream in and of itself can.

just answer every single question magically, but it can be an incredibly powerful and at-scale collection of data and going back to the fact that it's passively observed, right? Once you start, I think it's Schrodinger, right? Once you start observing something or Heisenberg, can't remember which one it is to be frank, Heisenberg, once you start observing something you inherently chang it right? So therefore, once you start asking people questions and surveys, which are

Will Francis (09:12)
Heisenberg principle, yep.

Eli Goodman (09:22)
excellent at answering any number of questions, but the moment that you start asking people to recall what they had for breakfast this morning, you're going to get wrong answers, right? So now stretch that out over time related to things like aided recall, favorability, likelihood of purchase and so on and so forth. But clickstream is deterministic. You can actually see what happened, right? So again, it tends to be a very nice quantitative collection of pattern data that you can play with to help you

Will Francis (09:42)
Hmm, yeah.

Eli Goodman (09:50)
cross that divide between 'I have no idea' to at least correlation onward to causation.

Will Francis (09:56)
Well, you know, one of your customers is SparkToro.

Eli Goodman (10:02)
Yeah, we work with Rand

Will Francis (10:04)
You work with Rand Fishkin who we talked to on the podcast and held a webinar with and yeah, he's obviously got, he's quite ambivalent. He's got mixed views about marketing attribution as a thing. And I hear a lot more mixed views about the idea of marketing attribution, whether or not it's a realistic aim in itself to even try and, you know, build that picture.

What do you think? What's your take?

Eli Goodman (10:36)
Listen, as an overarching piece, and I think I mentioned this earlier in the podcast is it could be marketers in general, because there are so many tools, if you will, there's this push and has been forever, whether it be about SaaS or any number of  technologies that just say, well, I just feed the campaign in and then the tools tell me exactly what happened. Right. And there could be an over-reliance on the outputs that come from these

let's say attribution modeling tools, right? Let's all just collectively call them that. And the methodologies are all very different and panels are all very different and there's bias. Am I ambivalent about it? No, but I think that there needs to continue to be innovation here because

you have so many marketers that have been so caught up in what the tools tell you that to me, and this is why I am in the data business as opposed to the tool business. So going back to sugar and cake baking is I believe that it is very important to be able to separate yourself where you're not just trusting one tool or the vendor that you bought is delivering your ads or what have you. And you need to have effectively a second or third opinion. So this is where we start to get into

marketers and trust me, I started my career at BBDO long ago, right back in the 90s. So I certainly understand. it's almost like a paradigm shift where you just got to open your mind a little bit to be able to say, OK,

Will Francis (11:53)
famous

Eli Goodman (12:00)
The tools will tell me X, Y, and Z, but I need to understand what the limitations of any individual or collective tools are. And that goes back to the data. And that doesn't mean everybody is magically a marketing and data scientist overnight. But I think you'll hear me continue to hit on that point around nonlinear analysis in order to be able to define what is the value of the campaigns that have come before you. And the other part that I'll mention on this real fast is you also have the ability to look at 
other campaigns that you're familiar with that you weren't involved with? Like, did that work? How valuable was that Super Bowl commercial? Like, how helpful was that particular viral campaign as it related to driving outcome later? Even though you don't have access to that information, in, you know, wherever they were running the ad campaign, it's a nice way to be able to help you battle test things that you're thinking about in different angles to be able to...

 

to run with. Anyway, that's where I talk about why the data is so important is the, not just the tools and the outputs, but the inputs and looking at both of those to be able to gauge how that could impact or work for you.

Will Francis (13:09)
Hmm, that's interesting. Not thought about that really yeah, the ability to look at stuff you didn't work on and try and get a bit more of a holistic sort of multi-channel, omni-channel view of the marketing and how it might be having impact. That's really interesting. I mean, to me, my read on this has always been that we just like simple answers. People just, we're busy and we're so...

We're so seduced by this idea that one tool, one platform, or even if it's just the Meta ads dashboard, the Google ads dashboard, give me my number, give me my cost per acquisition, boil it down, make it simple. And the answer really is, it really isn't ever simple, is it? we ultimately, but I think we're drawn in a little bit by that. Answer me this, what are the most surprising things that your customers find out when they look at this data?

Eli Goodman (14:01)
The first one I always think of is that if you sat down and asked any given marketer, no matter what product or service that they sell, and you would say, how long do you think the decision-making journey is, right? Related to your particular product. Everybody has an idea of that, right? They'll say, well, it's this many days or weeks or what have you. And almost every single time

is that the decision making or purchase journey of things is much longer than anybody even concedes And I'll give you an idea in the automotive world, right? When people are thinking about buying cars, that automotive journey tends to start six plus months before you actually walk onto a car lot and test drive. And it can start out with generic searching, which is just new car, used car, right? And they start to work their way through. And the more specific people become related to

the actual manufacturer, like the OEM, or the brand of the particular model of the car, the year of the car, and so on and so forth, they can get very specific. The deeper and deeper that you get in that, and the more specific it is, the closer and closer you are to eventually going and test driving three cars and then deciding to buy one. So a conversion in the online automotive world is effectively when somebody types in and says, either

schedules a time at the dealer or gets directions to the dealer, right? Like that's really like an online conversion, if you will, short of the, where people are like Carvana and people are just buying cars online. So that's another way to be able to look at it. But that's just, you think about, yes, that's a big ticket item, but even for small ticket items, you know, people will spend a lot more time and research playing around before they eventually show up and book.

But say a hotel on Expedia or direct on a hotel supplier site. So that's something that always jumps out to me is that the difference between what marketers think as far as the length of the journey and the actuality that you can see deeper in again, our type of question.

Will Francis (16:06)
It's interesting that isn't it? Because again, we think that, well,  a linear scale there between the cost of a product and the time people take. know, if a car is 50 grand, well, that takes 50 times as long for people to think about as a product that costs $1,000 or something like that. But it's not because like you say, even when we're spending literally maybe $150 booking quite an inexpensive hotel room for a night, it's not $150 that's at stake.

It's us having a terrible night's sleep or just not enjoying being in that city or being in the wrong part of town. There's actually a lot at stake, isn't there? And I think it's down those low that lower end where I would imagine people are most out of whack with how long people think about

Yeah, $20, $30 purchase.

Eli Goodman (16:52)

Well, it's, I mean, listen, it stretches it out and certainly from a marketer's perspective, and you go back to, you know, that  which is simple or easy, right? Is in the end, it's what is an individual or a group of marketers at a business? How are they judged? Right? Is it purely on sales? Is it on downloads of something? Is it on how many people viewed something right? What are the success metrics that the CEO or CMO are putting to their team, right? In order to say, go, go and do this. So, you know, the flexibility that any individual or group of marketers have in a business is going to be limited by somebody that's sitting on top and actually cares or doesn't care. So what I'm getting at here is that if you are interested in expanding that spectrum of that which you believe is important to your marketing efforts, and this goes back to just where the data comes into play, you need to be able to make a case for it, right? To be able to say,

This matters. It's not just about the last click attribution and it's all, let's go spend it all on Google or let's go spend it here, spend it there. It's also being able to say that these impressions matter. Like the way that people interact with your brand, maybe it starts on Reddit subreddits six months before they show up and they start to be able to find out how important or not important, that particular scandal was related to something about your brand, or maybe you had a launch and it didn't go well, right? And then you pulled back and changed direction. Is there a way to be able to actually quantify the impact that that had on the success metrics that you're being judged, even though they're not in your immediate view? So that's where I go back to the, we all have agency, right? Like in our lives.

in our personal lives, I hope everybody has some degree of agency, but in our work lives as well, right? Everybody's got a boss. I got a boss, I got a board. You know, like it doesn't matter what your title is. You need to be in a position to be able to, you know, innovate, but then also prove like, why are you suggesting this particular course of action? So in the marketing sense, it can't just be, well, this is what Meta is telling me I should or shouldn't do, because they're not necessarily an objective player.

Will Francis (18:45)
Yep.

Eli Goodman (19:08)
in this decision making here, right? Like things are going to be geared towards, look how awesome we are at delivering your end result. And then I think in the classic pre-internet days, there was no CMO ever got fired for buying television ads, right? Like that's just the nature of it.

Will Francis (19:21)
That sounds like you may have spent a good bit of your career, most of your career being a champion for the top half of the funnel because you've gone from working at BBDO. Was that in New York in the nineties?

Eli Goodman (19:36)
That was actually in Atlanta back in the late 90s, right? But then I quickly then turned after that, because I found it to be interesting, but to be honest, I couldn't afford to work there. It was not a very lucrative entry-level job advertising in the 90s. And just by happenstance, and this was early internet, I got a job in sales at a big tech research firm, a place called Gartner, and that set me on a path 25 years ago.

Will Francis (19:39)
interesting.

Eli Goodman (20:05)
to where I am today, right? Because I love math. I believe it. could geek out. I love statistics. Like, it's interesting to be able to gamble, you know, playing cards, right? Things like that. So that entire universe is appealing to me. And it was really just sort of, it made sense even back in like the late 90s and early 2000s about where things were going as the internet was taking off. And it really, yeah, certainly, you know.

Will Francis (20:26)
But what was that pivot point

that led you towards specifically data gathering and the value of data? When did that become sort of blindingly apparent to you?

Eli Goodman (20:37)
Well, like anything, right, better lucky than smart. Like you kind of start there. And I think that it became very apparent to me as because I didn't major in statistics or anything like that. Like I went to a liberal arts college and it was like a lot of history and, you know, like literature and things like that is different from a lot of mathematics that I had pushed in high school. And so what happened, it was it was almost like a

A love story that began right back in like 1999 and began moving was when you really were working with analysts and people that were using the data in order to make decisions. And I thought it was fascinating, like the ability for certain people versus others to be able to look at the exact same piece of information and assess it and come up with different answers, like different ideas, different strategies as a result of it. So that was to me, I thought this is very powerful.

Like I began to then learn about not just what was tech data at the time, which was really about how many cell phones were being sold versus different types of software that companies were using for ERP systems and whatnot. But I began to then learn about more marketing science, right? Going back to the, in the 1970s when UPC codes were on the rise, right? As a way to be able to analyze what was being bought and why at grocery stores. Onward to what was the classic way that television and print advertising were bought and sold, right? Related to Nielsen ratings or MRI, which was doing analysis around like, you know, ads and magazines and how much they mattered or didn't matter. And I just found that type of analysis, like that direction, very interesting. And that was really what set me on the path because the internet was so young, everybody was trying to apply like these classic offline advertising principles to online.

And it worked for a hot second until things began to exponentially explode across not only the US, but the world. And new ways of analyzing data had to be considered, right? It was not as simple as like Nielsen television families, right? Like you really needed to innovate. So I think that was really the turn for me was when you saw the explosion of the amount of data and what broke

between the old way of doing things and the new way of doing things. And then eventually watching those two things like bifurcate, but then have to come back together because they both apply. So I think that was watching that journey over the course of the early 2000s was radicalized me, if you will, as it related to data and what I believe it could or couldn't do.

Will Francis (23:06)

It's interesting that, you know, it's nice to have seen that progression. I'm maybe slightly more junior than you, but I did start playing around with this stuff at the end of the 90s. And yeah, for sure it's been an interesting, this is a really random, silly question. But if you took a marketer from today, like a really data-led digital marketer, right, from today, plunked them in the mid 90s in BBDO in the TV department, but you also did a swap. You took an old school mid 90s TV.

ad creative out of there and drop them in today's marketing landscape. Who would your money be on at succeeding best?

Eli Goodman (24:04)
My money would be on the current market or simply because you have everything behind you, right? And the ability to understand the internet in totality because it didn't even exist back then. Television still exists today. People do still pick up magazines and newspapers, right? So you can at least conceptually understand what that's about. Because here's the issue, right? The core in that question is that everything back in, let's say the nineties and before it, it was all based off of sampling and projection.

Will Francis (24:22)
I so.

Eli Goodman (24:33)
You would take a sample, you extrapolate out to a universe, and that is representative. And that's because the viewing points or consumption points were limited. Your television, it only had so many channels. There's only so many magazines you could read in a day. There's only so many properties. And there is effectively an unlimited amount of information on the internet today. And it requires a completely different way of looking at the data in order to get to the answers, right? And then I say this because you could certainly per analysis by analysis, there's too much, right? So you then also need to figure out like, what if this is junk versus not? So yeah, I would put my money on the group that is more familiar with how to deal with like census level data, like data from the source in the billions or trillions, right? Like that type of thing. That's it. I will go back and say that like the 90s people

didn't have all of the technology tools to just kick out the answers, right? So there wasn't actual like you're at the ad agency and you had a straight up research department that you would get together with. Like I was in the account executive department and the account executives and then you got to talk to research and they would come in and you would actually have all of these discussions and meetings about this is what the data says or the research says or the surveys, know, whatever you may call it, the focus groups.

Eli Goodman (26:00)
And then the account executives and creative like sit there and figure out which way are we going to go. So the human interaction of that, I would also say is very different than oftentimes everybody's just pushing data points around and be like, well, the data says this. And the human aspect, particularly in marketing can get lost, which is I don't care how much data you have, is also incredibly important. People have to make the final decisions.

Will Francis (26:13)
Yeah, you said that. I noticed I picked that out a few minutes ago. You said something about, when you talk about earlier in your career, you loved that thing that where you could see data and people would make their own decisions based on that and see different things. And I suppose that's my question to you is, know, what is this, you know, to people to talk so much about being a data-led marketer, what is that set of skills?

Eli Goodman (26:53)
Sure. Listen, the first things first to be a data-led marketer. And let's say that those things can be paradoxical data-led and marketer together is to be successful. In my opinion, you need to be able to straddle that fence, right? And I say that between the quantitative and the qualitative. The quantitative end of things is the data-led being that there is a lot of information that exists out there. And there's going to be tons of information related to your campaigns or your products or your industry and whatnot.

And you have to look at that. It does matter, right? You cannot just say everything is gut feel, right? And that's how we do things. And oftentimes that's how things were done, right? Like back in various times, and even today, any number of creative decisions that are, I feel like this is right. And that is important because instincts do matter. But I go to that as a data-led marketer versus data only is an important factor here, right?

you do have to have the ability. And that's why I say the qualitative. What's the quantitative? Who, what, where, when, and how? Like you just count things. Like I'm in the data business, we count things. Like that's it. But the qualitative is the why, right? And then what is the recommendation? What should we do? And although there are any number of like tools to be able to help you assess and like, you know, give you direction and like all of that is fine. Again, in the end, success for me is that you need to be able to understand at least the basics of what the data is saying.

And then make a qualitative call based on any number of like experience or we said instinct or expertise or what have you. Because you know not everybody is going to be an expert in women's clothing versus selling software versus selling whatever right? It could be anything so. That's where I think a success, data-led means that you need to have the data and you need to know how to be able to speak and understand it, but also understand like methodology of data, right?

Eli Goodman (28:50)
Where does the data come from? Why is it trustworthy? So that if you see something in the data that is an outlier, you can question it. I'd be like, this doesn't make sense. Because sometimes it's wrong, right? That's the thing. Like something got, somebody moved the decimal point in something, it doesn't make sense. And then you ran out there and made decisions on it or publicly did something. And you're like, and this is, I'll use a, at least an American term for something that we will talk about. And I would apply this to data-driven marketing.

Eli Goodman (29:20)
Measure twice, cut once. You cannot unring a bell. So before you go and you cut that wood in order to place it in the thing, take the time to understand what you are doing before you implement. So that's, if I go back, that's my opinion on how, at least how I would describe the data-led marketing and what's required for success.

So there's a constant battle of ideas or principles in marketing, the creative saying, we shouldn't listen too much to the data, you it's all about creativity. Ultimately, it's a creative discipline. You know, it is a creative industry. It's part of the creative industries. And then you've got data people who are like, look, if a decision is not based on data,

then it's ultimately a bit of a fumble in the dark. So where do you stand on that sort of creative versus the data dichotomy?

Eli Goodman (30:14)
So I'm going to throw

in a third component here, right? There is data, there is the creative, and then there is also, and this is just to be fair, there is the financial aspect of both of these. Okay, so let me give you an example, right, of what I'm talking about, is that back at BBDO, it was, you 1999, we were working on a campaign for a new lawn and garden like weed killer

that was going to be coming out. Creative and then research had been going through and putting together. Creative basically goes and just comes up with ideas. They just sit there, they talk with the client, what do you want out of this and so on and so forth. Then they go into the creative lab and they come back with really interesting ideas. Then everybody gets together and you start to whittle them down. It came down to at the time, there were two campaigns that

And this was then where we had used the data in order to be able to determine like through focus groups and whatnot. One of these two was going to be the play. Okay. And imagine this was about weed killer. And at the time the, and this is the brand by the way, was by called Bayer Advance Like, know, like Bayer pharmaceuticals was launching their way into into the sort of weed killer lawn care universe. And it came down to two campaigns. So the argument between the data people

right, versus the creative people was that the two campaigns were: one was license The Eagles' Peaceful Easy Feeling so that like, you know, the idea of when they would put together all the creative, that would be the music that would be playing and it would give you like this warm, my, my, my look, I love my yard and look at how we spend so much time here and my family's running around and birds are chirping and so on and so forth versus

There was another camp that said it's about comedy, that there was a comedic end of things and they wanted to do like sort of an animated version that would have celebrity voice. So would be like, if you go back to classic weed killer commercials, if you remember, it was like the weeds were being attacked and it was all this type of stuff. So it was funny. So it was, was it peaceful, easy feeling or was it comedy and what it was going to be? And the research were telling different things.

Eli Goodman (32:38)
And then Creative had different ideas. Creative absolutely wanted the peaceful, easy feeling, right? Like as far as they were concerned, this was the way to go because that was just what they believed. And the account management and research side had said, well, there already exists, like at the time, what was Scott's Weedkiller, which is another one of the major brands. And they had this little Scott's guitar that would play and they felt that peaceful, easy feeling was too close to that, right? It was like not going with their own brand.

But the third angle, and this is why I mentioned this for digital marketers, it came down to cost. Even back in 1999, and this was so fascinating for me as somebody coming out of college and learning about advertising at the time, the amount of money that it would cost in order to license The Eagles' Peaceful Easy Feeling for the commercial was a million dollars. And the amount of money, and this is where they ended up, the decision ended up being the cost, right? This was the final, this was the third thing.

Will Francis (33:28)
Wow.

Eli Goodman (33:35)
It was they went with a comedy animated angle because it's much less expensive. And the voiceover was done by Bobcat Goldway. If anybody who's old enough to remember he was a comedian, he used to be in the police academy movies. He was funny in the 80s and 90s. Still doing his thing, but like, you know, back in the day was somebody that was notable, notable enough for the commercials. So I mentioned all of that to be able to say you absolutely will have these competing factions, right? And they're all important. They were all very helpful that.

I also referenced that we also have financial limitations that any marketer has to deal with, right? You don't have unlimited money. So that's why the data can help guide the creative so that everybody could figure out like where we want to land. Because creative was happy with both of these, but love peaceful, easy feeling. See what I'm saying? that like you also then the data in the end, let's call it, it was like a 50-50.

Eli Goodman (34:28)
the financial end had to come in and like make the final decision because it just was like, how much are we spending on the making of it versus the placement of it? So anyway, just one example of like all those different, the competing things. So yes, creative versus the data itself, which again at that time was research. So I know that's an offline version, but that still applies today.

Will Francis (34:38)
Yeah, yeah, that battle plays out, is playing out right now in agencies all over the world and does do constantly. Well, let's throw something else into the mix. So that was obviously a while ago and that's basically been how it's gone forever, right? But very, very recently we've seen this thing sort of like the meteor, I hope it's not like the meteor, the asteroid that killed the dinosaurs, but a few years ago AI landed.

And AI seems to have disrupted pretty much every industry in some way or is certainly going to or threatening to. And I'm interested, you you're a CEO of a modern company and you're having to think about this stuff, you know, from that point of view, but also you're no doubt considering this stuff and its impact on creativity and on how we analyze data more effectively. So it's a big question, but how do you see AI having impact in marketing today and in the near future?

Eli Goodman (35:54)
Yeah, well, I have any number of ideas, but let me distill them down, you know, so that for the sake of time, we can certainly highlight, but on a personal level, the first one, if I just look at digital marketing and anybody that has a job with it or responsibility, the most valuable resource that any of us have is time. None of, nobody has enough time, right, to be able to do anything. Like we all have personal lives, you have work lives, everything is always an emergency or a fire being lit and I need this yesterday and so on and so forth.

So there's never going to be enough time. And certainly AI as not only an idea, but a practice helps cut down on time, right? Like, if you think about that and how valuable that is. And so that is a benefit, right? In any number of ways. But when you talk about a challenge that it could bring, certainly in marketing, that then does require that you have to oversee whatever you're using for AI to make sure that it is active.

right, that it does actually speak in the brand voice that you want. It is, because remember, AI overlays of any kind are only as good as the data that goes into them and the models that output them, right? So AI hallucinations, for example, could be a major problem that people will have because the data was bad or the model was bad that went into it and whatnot. So even though you might've saved time, you have to remember that you were going to edit. This goes back to the measure twice in a couple of months because, and I say this in both print,

like let's say actual words that are being written or messaging or spelling or something like that. It's also about the creatives themselves, right? Because people have begun to work with any number of tools to just create creatives and so on and so forth. you're going to have to, it may save some time on the front end, but you would better build in time on the backend to be able to assess what the AI has done to be able to see if it is actually what you and your clients want. That's one

Eli Goodman (37:51)
that I would say is uber important. You cannot trust the AI to take care of it without you overlaying with it as well.

Will Francis (37:58)
And that's the same on the creative and the data side, isn't it? I I personally, I'm always trying to push people more towards the data analysis side because that's machine readable data. It can save a huge amount of time sifting through thousands of rows. Whereas the creative output, yeah, it still needs a lot of guidance. It's like having an intern who's having a decent stab at it, but ultimately it's just, you know, never gonna be good enough for prime time in reality. So do you agree? you think?

Do you see AI being as useful on the creative side, the data side currently?

Eli Goodman (38:30)
I think that the obvious where we are today, it's more helpful on the data side, right? Cause what is AI at its core? It's machine learning, right? And going back to what does machine learning, what is the purpose of that as far as I'm concerned, and certainly where the data plays, it's about identifying patterns within data and especially very large data sets that human beings would have a much more difficult time doing on their own. It were impossible time, okay? So the obvious...

if you're waiting, it's going to be weighted very heavily towards AI helping you analyze the data to be able to find that which is good or bad for a particular marketing campaign. But I would also argue, now your point about it never being able to take over on the creative. I don't know if that's necessarily a case that I would support based, and I say this not as the creative, which I am not. I say that as,

Think about like Hollywood studio executives that are trying to be like, you know, how much do we really need to pay people? Because we could just have the AI create the visuals and like the actual output, which I don't believe in, right? It's still the human element here can never be removed in my opinion. But that doesn't mean that like, there aren't going to be any number of decision makers that are going to try to remove the human element because of pure cost. That they're like, you know, how much are we paying these people? Because the margin

Eli Goodman (39:51)
on human beings and services and labor and effort is never going to be as strong as it would be with AI. But that doesn't mean that like, you know, that's the right way to be able to do it. But I would, yes, I would say that AI as it relates to the data and the analysis and like, you know, how to be able to organize and identify patterns is the play right now. But the creative end of things is also under attack, right? Like, I mean, every other day they'll show you a new example of

Will Francis (39:59)

It is.

Eli Goodman (40:17)
Here's Will Smith eating spaghetti last year versus Will Smith eating spaghetti this year, if you saw that one recently. And you're like, you really covered a lot of ground in the past year about how well that works. So it can be scary on both ends. Yeah.

Will Francis (40:20)

Absolutely. Of course, there'll always be people trying, there'll always be people using it, whether it actually passes muster is the true test. yeah, no, that's like I say, it's advancing at an amazing rate. Well, okay, we've got a few minutes left with you. Thanks for your time. It's been very interesting, but I feel like I'm learning a lot, but I'm so fascinated by your kind of founder story.

So we've talked about your early career and all that, but as a founder, what have been the kind of hardest things about going from literally founding this, presumably in a coffee shop or your bedroom or something, or an office, spare office somewhere, to where you are today? What are the kind of hardest lessons you've learned there, I think?

Eli Goodman (41:23)
Well, so first of all, it was from the couch, right? To tell you exactly where it happened with myself and my two co-founders, Serge and Stan, and it was mid 2020, so it was COVID. It was during lockdown. Like this was done where Serge and Stan, we never even physically met each other for 18 months as a result of COVID. Like we had met basically over LinkedIn and then had some mutuals in common amongst us that led us together with Semrush as

Eli Goodman (41:53)
our mutuals between us and our initial seed investor. So that's one, right? It was definitely done from the couch in my condo here in New York City, all virtual right from the beginning. So a few things I could say, because this is also, you know, it's fascinating to me as well, because I look back on it. I am not a serial entrepreneur. It's not as if I've been entrepreneurial, if you will, for the past 25 plus years. But sometimes, you know, an opportunity presents itself. And I would say that

It's one thing to be in a position to be able to recognize an opportunity when it presents itself. But more importantly, when it comes to founding a business, you have to be in a position to be able to like actually take advantage of said opportunity. And through any number of like decisions and successes and failures and all of these things that blended together leading into mid 2020, I was finally in a position on a personal level to feel that I could take this chance, right? So that was one,

is that past is prologue when it comes to founding a business, right? Like, you know, not everybody is just sitting on a trust fund or like, you know, people got bills to pay, you got stuff to do, right? Like it's not, you have rich benefactors to be able to take care of them. So that's the first one that I would say to anybody is that your ability to be able to found a company and then to be successful and whether what's going to be any number of ups and downs, you need to be able, it's what's come before you, right? It takes time.

Right? Like diamonds are made out of like pressure and time, if you will. So, so that's the first thing that I would say is most important from there. First things first, every meeting that you have is going to be, when are we out of money? Like start there, right? Is that I don't care about raising money, not raising money. That's your first thing, right? Is that we need to keep the lights on. So our entire tact was not about

Eli Goodman (43:46)
just R &D forever and like raising money and like building moats and like all of these different cliches, it was what is the purpose of this business? And it is to build things that we know sell, right? We look through a sales lens as our number one thing here is that we have to focus on selling things, revenue. And we moved into profit. Yeah, like, it's just not at all cost, but you you need to like sit down and think about what it is that you're selling and find a path to revenue.

Will Francis (44:05)
revenue as soon as possible basically.

Eli Goodman (44:15)
you know, as quickly as you can, right? Like I think that, you know, everybody, especially like how things were done over the past 15 plus years related to raising money and, you know, the gold rushes at various times related to crypto or AI right now. And like, don't worry about profitability, it's revenue growth at all costs and so on and so forth. Well, that was also all during like what was effectively around the globe, a zero interest rate phenomenon environment, if you will.

ZERP is what would be called here, zero interest rate phenomenon tends to be the shorthand that we'll use here in the United States for it. During ZERP times, nobody cared about profit, margin, things like that. So that's a second piece when I think back to how we went about founding and running the business was we took that tack that the salad days of low interest rates were going to end at one point. So we needed to make sure that we had an idea that was going to get us to actual revenue much faster than it might have been back in the 2010s.

Will Francis (45:19)
Interesting, interesting. And then you were acquired by Semrush.

Eli Goodman (45:25)
That is correct. That was December of 2023.

Will Francis (45:28)
That's pretty good going. Three years, just over three years and you get the acquisition. What was the first thing that went through your head when Semrush first mentioned, used the word acquisition?

Eli Goodman (45:41)
Sure. Well, I remember it very specifically because we were out to dinner here in New York. I've known them, they were initial seed investors. So I certainly have a relationship with Semrush and I've known them also as clients many years before that. So again, going back to the how it passed this prologue in your life, in your career, it's not like lightning strikes. These are things that build up over time. So anyway,

I knew that we were going to be sitting down to have a discussion. I didn't know it was exactly this. It could have been in any number of ways. But listen, to be honest, the first thing that goes through your mind is the money. You immediately are like, how much? But you get over that real quickly because it matters. Money matters in the world. But put that aside for a minute. Is that the real question that you start with is why? Start there. Because

The selling of a business is about not just, it's like planting a tree. It's not what it looks like today, but what is it going to look like in another five or 10 years, right? Like that's sort of the landscape architecture of Central Park, if you will, right? What's all this going to look like in 50 years? So on a much shorter scale, the why is you start with, it's not just about the money. Remember that, know, Datos...

Eli Goodman (47:02)
is a collection of people and everybody that works here and their families that depend on them and like investors and all of the wealth that gets built. It's not just about like financial wealth, but about like the perpetual ability for people to be able to succeed. So you sit back and you say, well, why does Semrush want this? Why would this be appropriate for us? Why would this work for everybody? in, because again, when somebody comes to acquire you, they don't just say, well, here's the money, thanks and peace out. Like they're looking at

we will acquire you and then we need to continue to build, right? Over the course of the next however many years, right? Because there's more money to be had. There's, you know, you have like earn outs, you have all these different things. So again, it's not about just what the business is today. The value of businesses is also about their worth many years from now. And then not only the founders, but the team and the product and all these things that you've built. So it goes back to, again, you then really like, once you get past your like,

Eli Goodman (48:00)
The monies, you're just like seeing dollar signs or what have you. You get past that real fast and then you really have that, you start to just have a call to come to Jesus moment, right? If you to use that, is just why are we doing this? Like, why does this matter? And then that begins a set of discussion to me that is much more fruitful, right? Because

Eli Goodman (48:26)
just because somebody comes up and says, hey, we're thinking about acquiring you or making an offer. You got many months of like diligence. it's now going to like, everybody is now going to dive in. It's all just based off of like surface stuff. But now we're talking about a deep dive into everything that like your business and you personally have done and all these different types of things. It doesn't happen overnight, right? Like it takes a long time. So just because somebody makes an offer, it doesn't mean deals close, right? So that's what I say by

Eli Goodman (48:57)
You put the money aside because that doesn't matter. That'll be the end result of what is a successful acquisition. But you need to start to get your head around about why this is important to the acquiring company, why it's important to your company, and then figuring out how to be able to circle the wagons around what's important to you and your team and your employees and your business. Because it's easy to get swallowed up right by somebody else and then not have, and as we talked about earlier in the podcast, agency. Not to have.

the agency to be able to do what you need to do because you get acquired, you're giving up control, right? Like that's, you have to come to grips with that. So there's a degree of protectionism that you have to take towards the exercise, which is slow down. Yay money, but slow down and think about like why this is, why now? Like why not later? Like, so that to me is really the core. the, once you get past the emotional reaction,

Eli Goodman (49:55)
the thing that really matters is that.

Will Francis (49:56)

Yeah, interesting. mean, it sounds like I was going to ask you about, you know, advice for aspiring founders who dream of an exit via acquisition. And you kind of answered it there, but is there anything else quickly to mention there to founders who dream of that exit?

Eli Goodman (50:15)
Yeah, there's actually a very tactical thing that I would recommend to everybody, which is when you found a company, there are that's that's all great. You're like, hey, I found something. But there's a lot of like legal and financial things that go into. Founding a company, right? Am I going to do, in America you would do, am I going to do a limited liability corporation is a partnership is that a C Corp, which is like a corporation?

What are the accounting rules I'm going to apply to what we have going on that impacts how you pay taxes and so on and so forth. But I mention all of that because if your intention is that you are starting a business, that you are eventually going to or would be acquired by a big, a much bigger company, as soon as possible, you need to start setting up your business in the way that big companies set themselves up when it comes to finance, accounting, legal and whatnot. Because if you set up as

an LLC and you eventually take on investors, they're going to want to reorganize here in the United States as what's called a C-Corp, right? That is a corporation that has different, that's like stock structure and that governs like the taxes that you pay and like any number of things. Accounting, right? Here in the United States, you can set up and you have what's called cash-based accounting, which is you just run it based on money in, money out.

Right? Be like, we made profit this year because we brought in more money than we sent out. But in big companies, especially public companies, they have what's called gap accounting, generally accepted accounting principles. And that is like very formal ways that they look at accounting and they call it a cruel basis. So there's a big difference between what a cash-based business would call profit versus what a gap accounting business would call non-working capital. Right? And I can explain that in detail.

Eli Goodman (52:09)
But I'm using that as examples here is that if you want to sell your business, right, the sooner that you can start getting it set up in ways that big businesses look at things, it's going to be able to help that transition, the ability for them to acquire you. Because that type of thing is going to come out in due diligence. And then you have to figure out like, how do we all get together and agree on the value of the business in the end?

Will Francis (52:10)

Yeah, that's true that in fact, funnily enough, some of the things you sort of mentioned in there take me back to when I started an agency, I co-founded an agency in London and I went out and bought, literally bought a business accounting for dummies from the four dummies range of books. But because I thought I really need to know about things like gap accounting and EBITDA and stuff because I just, someone's got to know about this. The other two guys weren't interested. So I thought, yeah. So I think,

Eli Goodman (52:46)
Run!

Will Francis (53:01)
A bit of financial literacy as well is probably quite good, isn't it? Going back to the data bit, I wanna just ask you for some mega practical tips to close. What are your three top tips for being a more effective marketer through data?

Eli Goodman (53:18)
Yeah, well, the number one is you have to start thinking in a nonlinear fashion, right? Is that like data and the right data will help you better connect the dots between things that aren't obvious, right? And I think that is one of the number one things that I recommend for a data-driven market, right? Because we all have like a long history in marketing of the obvious things, right? I did this and suddenly I sold more stuff, okay, right?

So it feels like a causation, right? I put these ads out and we sold a bunch more, but your ability to then also correlate other factors that might've gone into it, right? Like those are very important. The second is I highly recommend if you're going to be a data-driven marketer that you educate yourself as best as you can about measurement, right? Because data, and this goes back into some of the nonlinear thinking, there are a variety of ways to be able to measure success

and your ability to be able to explain to somebody why this worked or why we should do this particular strategy, implement this strategy, or not do this other strategy is that you need to be able to measure it. If I can't measure it, didn't happen, because otherwise it's just luck. I think that's a second piece when it comes to data-driven stuff, because if you don't figure out how to corral that data into something that makes sense,

And you understand how to be able to identify things that's going to be difficult. the third and the last thing that I would say is, and this is also important in data, data can be wrong. Data can be misinterpreted. So it is okay to fight back. It is okay to be able to lean into your instincts and your creative skills and call it the more qualitative things that are to marketing. Those are just as important as your ability to analyze data.

Eli Goodman (55:13)
Because yeah, data could be such a benefit. I work in it. But it also could be an albatross around your neck, right? If you don't also remember that your instincts matter, right? Your gut feeling does matter. So it's a bit of the art and science, if you will, the cliche of that. But that's the third thing that I would say. Do not only depend on the data. Because otherwise, think about it. The data is the GPS telling you to drive off this bridge into the water.

Eli Goodman (55:42)
And you're like, OK, that was

that's it. Like I thought, well, follow the data. The data told me to do this. And suddenly you're halfway buried in the lake. Right. So you remember, use your eyes, use your ears, you know, use your brain.

Will Francis (55:46)
Yes, indeed. Interesting, very good. Good advice there, Eli. Well, look, thank you so much. We've learned so much in the hour that we've talked. I really appreciate it. Just remind our listeners where they can find you and connect with you online.

Eli Goodman (56:10)
That's easy, right? Datos.live, right? That is our website. Or you can also find me on LinkedIn anywhere, right? I mean, that's very simple. But those are, I would say, the easiest ways to be able to find us. But feel free to reach out, like myself, my team. We'd be happy to engage with anything that anybody has going on, if we could help.

Will Francis (56:13)
There you go.

Yes, cool, great. Well, thank you so much again. And look, I hope to chat to you about this soon. It's been fascinating.

Eli Goodman (56:37)
Well, it's been great. I appreciate you having me on.

Will Francis (56:39)
You're welcome.