MIKE DUBOE

Hey everyone. We just spoke with Tim Doyle, CEO and co-founder of Eucalyptus. Tim’s an old friend and one of the most interesting marketers I know. He always has interesting takes on brand versus performance, but doesn’t frame it in such a way. I think in his mind, everything is performance and really what drives performance is the ability to tell great stories and then also measuring it well, and I think Tim, the way he articulates what actually goes behind each of those, I find to be very interesting.

RISHABH JAIN

Yeah, one of the key ideas that he shares in this podcast is about the speed at which you catch the cultural moment being key. He tells this story about when he was working on a political campaign, and 24 hours before the election cycle, caught a key moment and he feels like was able to actually have a real impact in that election cycle. We are excited for you to listen to this episode and get some of his insights.

MIKE DUBOE

All right, Tim, you’ve been very vocal against VC podcasts, and so with that I’m honored that you agreed to spend an hour with us, and I hope I don’t regret this. We met a little over five years ago when you were described as the single best marketer in Australia, and you’ve become a trusted friend and one of my favorite people to riff with on just all things marketing and the evolution of D2C eCommerce. You’ve since scaled Eucalyptus into a leading consumer health brand and are one of the few who’s made this house of brands concept really work. And so, excited to spend some time on that with you today and just all things marketing. Maybe as a primer for our US listeners who aren’t as familiar, maybe give us the 30-second pitch on Eucalyptus and the initial thesis behind setting out to build that.

TIM DOYLE

Yeah, sure. So I was at Koala, kind of like a competitor to Casper in the golden days of the big Facebook arbitrage and took that and thought, well, these businesses can grow really quickly, but often there are inverse returns to scale with regards to CAC. And so maybe if I take what is the key skill that they have — which is slick websites and good marketing — and apply it across the portfolio, then there’s a chance that I could escape that CAC pressure and have a really strong house of brands in the style of any of the holding companies. And so I thought that was really smart. It turned out to be, I think as a thesis, quite dumb because I think all of those brands, when you run a company like that, all of the brands was exposed to the same risk, which was that as marketing gets more difficult and CACs rise, you don’t escape, you just get exposed to the same thing across your entire portfolio. So that’s why I never became, I guess, an investor or a capital allocator. But I think what Eucalyptus got right — and got luckily right, I think — was that we have an underlying piece of technology that differentiates us. So as we launch new brands, there is no Shopify in digital health, and so you can rely on your platform providing you with a competitive advantage no matter what category you enter. And so that gave us … As that got better and better, it got easier and easier for us to win new categories. When obesity turned up as the great new category of digital health, we were way ahead of the field in being ready to serve patients in that space. That meant that we caught the momentum the fastest and have caught it now across five markets, several hundred thousand patients, several hundred million dollars of revenue. It’s been a really wild ride, but I would say not on the original thesis so much, but on an evolving ability to capture lightning when it happened based on a really good piece of technology underlying that.

MIKE DUBOE

When was it, if you rewind back on the journey when the health business started to take off, you still had a portfolio of other brands you were managing at that point, was the allocation of attention and resources, was that decision obvious to you or how did you make it at that moment?

TIM DOYLE

My co-founder is an options trader by background, and so I was very into the idea of maintaining the portfolio and essentially hedging, thinking that my original thesis would play out. But he was like, Hey, the main thing in options trading is when you are onto a winner, you back it to its full extent. And so I think what became … We went from a multi-brand arbitrage essentially to a multi-market play. And I think this is something that US D2C brands do quite poorly, which is build a range of countries and then have them growing at different rates and then be able to deploy into them at different speeds. And so the same capital allocation advantage we had across multiple brands, we then got across multiple markets. And that’s been really central to how we’ve been able to scale.

RISHABH JAIN

I’m curious, have you read “Only the Paranoid Survive,” by any chance?

TIM DOYLE

Yes, I have.

RISHABH JAIN

Okay. As you’re saying this, of course I can’t help but think about the analogy from the lessons of Intel where they see in their factories by virtue of the factory demand and factory output that chips are taking off relative to memory and then they shift the business accordingly. One of the things that Mike was sharing with me is people call you a founder’s founder. When you think about the strategy you deployed for your business, how do you think about helping other founders understand that strategy for how they should run their businesses? So if you’re advising another founder on like, Hey, here’s how we thought about it, here’s how we resource allocated inside of the company, here’s how we then made the decisions, here’s how we brought the team along. This is hard to do in practice. It sounds easy when you say it the way you said it, but there’s real work that happens.

TIM DOYLE

Being a good support to founders is actually primarily about parking your ego and realizing that you only get a couple of things right on the journey. So I would say, across … It’s so nice to be able to look back on a journey and be like, Oh, I made all these wise decisions, but in reality we made three or four correct decisions over the entire life of the business, but they were so right that nothing else really mattered. And so when I work with founders or spend time with founders, I’m like, Hey, what are the three or four things that you’re going to nail and really scale here? And then what are the things that actually don’t really matter? And I think about the context of Silicon Valley is such a mimetic place where so many people … Everything is just replication of everyone else’s things. These strategies that work for one business don’t work for others. I come from a marketing background where I’m like, well, the things that work for my brand are the only things that work for my brand, and so who am I to tell other companies what’s going to work for them? So I try and really make sure that it’s like, These are the three things that work for me. They probably won’t work for you, but I’m here to help you find your things.

RISHABH JAIN

And when you think about help you find the things, I think the meta strategy is still deployable, right? The meta strategy of, Hey, make sure your experiment … For you, it was market-based, right? And so different things are going to be true for different particular businesses. How would you recommend someone think about architecting the organization in order to be able to run their quote-unquote main strategy but not miss when something is sort of starting to catch them?

TIM DOYLE

I think the most important thing here that I talked about with people is your … When you’re building a team, you’re running kind of a sports team where it’s like you have a constraint on your total salary spend because obviously you’re limited and you’re trying to produce the best possible outcome. Where are you going to spend in order to produce the best outcome? So where are you superstars? Where are you 3nD role players? Where are you actually not going to invest at all? And I think that piece is the portfolio management piece that’s so important to get right, is to be able to go, Okay, what is the shape of team here and what is the shape of skills that is going to allow us the best chance of making this work? For example, Eucalyptus is fundamentally a marketing plus a operational-scale company. And so the two things we had to get right is we had to get great marketers. That comes relatively naturally to me because I can see them and I understand how they work, but then we had to get great generalists as well, because ultimately the thing that makes this business work is landing in complicated markets and grinding through and getting scale. And so we had to be really good at pulling people out of Bain early and young and going like, Hey, here is a set of tools that you need to get in order to make this market successful. There are the two things we knew how to do. And we weren’t a technology-building organization for the first few years, and we weren’t a brand organization so much. We were just the two things we were and we were super disciplined and we made it work.

MIKE DUBOE

Before we jump into the marketing topics, I want to spend another minute on this. So you referenced some of the DNA of early hires. I actually want to talk about being located in Australia and both the benefits and, I guess, cost that had to you in scaling a business. What insights did you have that were a function of being based in Australia that you might not have had otherwise? And how has that impacted how you built the business?

TIM DOYLE

Yeah, this is exactly the right question of where you have to be uniquely thinking about your own situation, because there aren’t a deep pool of the best software engineers in the world in Australia, but we actually export lifestyle and wellness extremely well. So we are very good at marketing health. Actually, in our cultural DNA, it’s very easy to export that: We have an idyllic lifestyle that people aspire to. And so how do you monetize that and use that as an advantage? That was something that was really important to us really early on is getting that right. And then other things that you see is, well, it’s a small market, but you’re going to have no venture-funded competitors in your spaces. You can be a local bully and then use your gross margin from your local bullying to then go and compete in other local bullying markets. We never do the US, we don’t operate in the US because you’ll all play your blood sports and we will figure out how to do Japan. American companies do a really poor job of figuring out Japan because the incentives aren’t there. It’s like, Well, I’ve got California to worry about. I don’t need … I have 25 million people in Australia, so I have to figure out Japan. I have to figure out Germany. Whereas, when the going gets tough for an American company — and I advised a famous American DC luggage brand for a while and I was dismayed to see the Thanksgiving campaign in the snow launch in Australia in November, and I’m like, Yeah, that’s the kind of cultural arrogance that allows us the opportunity to win. That was really, really important. Understanding the advantages that come from being in Australia becomes a really important thing.

MIKE DUBOE

Do you think this is changing? So Australia’s had some great notable outcomes recently or on their way to outcomes, and it feels like we are — at least as software investors — we are opening up our eyes to the greatness that can come out of Australia and the unique attributes of Aussie founders. Do you think this is changing?

TIM DOYLE

No, I don’t think it’s changing. I definitely didn’t want to say that Australia can’t produce great software companies, it just has to produce its own type of software companies. It can’t produce the type of software company where your pedigreed Series A Silicon Valley VC calls up everyone in their portfolio and forces you to be a customer. It can’t produce a YC-networked customer acquisition strategy, but what it can produce is real PLG discipline in the form of what we’ve seen from Canva or from SafetyCulture or from Atlassian, or it can produce really interesting brands, which hopefully is what we’ve been able to do or what Aesop’s been able to do. We don’t need to become part of the Silicon Valley megamachine. We have to continue to play to what our strengths are and produce great companies that way.

MIKE DUBOE

Okay, so we’re going to move to a deeper set of marketing topics, and I think I’m going to start with a cliche framing that you and I have discussed a lot and then let’s go a lot deeper onto it. So brand versus performance. We’ve spent hours discussing this topic. I’ve made the point, everything’s performance, it’s just a question of time horizon. I know you have some nuanced thoughts on brand as well. Maybe talk about Eucalyptus. What is the integration with these two disciplines look like today? And then let’s maybe go deeper on a few attributes.

TIM DOYLE

Your framing of it is the most eloquent framing, so I’m not going to try and reframe that, but I think … What I think is really important from … I just never say the word “brand.” We talk about narrative and we talk about measurement. The reality is the great … It is the golden age of narrative. There is more returns to good storytelling than there’s ever been if only the people who historically have defended brand advertising would shut up and actually learn to do measurement. The rewards for big ambitious story are better — storytelling — are better than they’ve ever been, and the grubby arbitrage era that I grew up in is largely gone. My view is that the challenge is how do you tell great stories and the bigger the risk and the more ambitious that you are with them, you have to be better at measurement. I think the great pairing of modern marketing is hugely ambitious, high-risk storytelling and great measurement. If you do those two things, then you’re doing performance marketing at scale and you’re doing storytelling. There’s no such thing as brand marketing. It’s not a thing.

Rishabh Jain:

It sounds like ... And I'm excited that you listened to the episode with Bryan Cano from True Classic. He's a friend of mine who I consider to be a pretty good storyteller. He used to work at a company called Nude, and then he now works at True Classic. And in both places he took quite risky approaches to the kinds of stories that he chose to tell. And oftentimes you're seeing this rhetoric around the risk in the story that the brand is willing to take. Do you think that great storytelling has to come with that level of perceived risk? Or how does someone action on this idea of "tell great stories"? I hear you, I believe you. What do I do?

Tim Doyle:

I think risking can take a number of forms. You can be brand risk in the sense of saying something interesting. And the greatest way to take risk in marketing is to say something interesting. It's the obvious and best way to cut through. That is the obvious thing that is the first opportunity there. But I also think risk takes the form of the scale and uncertainty that you're willing to take the bet in. So you might be the first one to do a channel that's entirely unmeasurable. I was talking to this company yesterday and they just bought a million fortune cookies, and I'm like, that is going to be so hard to measure, but it's so perfect as a new acquisition strategy. And so I'm like, risk is both how you tell the story and then where you place the dollars and the scale that you place the dollars at. I think good storytelling has tension in it, and so there is real returns to that, but there are many forms of risk. It's just how do you create a idea or a story that cuts through into new space and kind of breaks the pattern. And if you do that, then the attention is yours to do what you want with.

Rishabh Jain:

Yeah, Tim, no, that makes a ton of sense on how to think about the risk that you need to take. And then the second part that you were talking about was on the measurement front and just doing good measurement. Again, it's a topic that's easy to say and hard to do. When you think about what good measurement is ... There's a lot of debate right now — attribution, incrementality, MMM — and again, as an observer, there seems to be more and more consensus that incrementality is the quote-unquote best approach. How do you think about each of those three and what place they hold in the marketing stack?

Tim Doyle:

Yeah, so I think you're right that measurement is the main thing. I actually am one of the world's biggest normies on this in the sense that I think the most important thing to get right is to be directionally good at understanding what a certain piece of creative or campaign is trying to do and whether it does it. And so, okay, let's take the fortune cookie example. It's not going to be measurable. You're not going to see uplift in an incrementality test, but you need to do the hard work to go, What is the signal that we will get here that indicates this is moving in the right direction and I'm going to continue to take this bet based on that signal. So the silly version of that was put a QR code in the fortune cookie, but reality is the QR code conversion will be poor.

What is directional signal that we would believe in here? And then if we can get that, let's double down based on our ability to take a portfolio of bets as to what actually works. And I think the other really big thing about being a normie on measurement is I don't find there to be a very good answer for ... The challenge of modern performance marketing is produce thousands of ads. When I see channel-based attribution of uplift or campaign-based attribution, I'm like, Lol, we were having a good creative week or a bad creative week. That is actually not going to be that accurate. And so what matters to me here is having a wide series of tools that I can use in different situations to get signal. And then ... I use all of these things, I definitely use ... YouTube's the one where holdouts set up uplift studies are the best because they're the hardest thing to measure and it has the biggest impact.

But I'm also, [I] really drill into our team that the thing that you have to be best at is understanding the thing you're trying to move and looking for lots of different signals to do it, and then having a wide array of signals to get things right. And also, I think one of the things that people aren't doing enough of is getting really good at understanding what's working in the channel, having a good, really good picture of what events are working in Facebook, what different creative styles are working, and actually doing most of your measurement within the platform and then looking for directional signal between channels. And then also people are premature optimizing with measurement. I think we're seeing too many pieces of tooling, particularly from an MMM perspective, when ... I think the great struggle with MMM is it is directionally useful and people are using it as a precise tool and it is not a precise tool and it deals with creative really poorly. Yeah, I think the answer is use them all and use them as best you can, but start with a really, really good understanding of the thing that you're trying to do and the signal you're trying to capture and then move down into greater complexity from there. And the fool's errand of producing the perfect attribution system is a real trap for growth teams. And I think one that I see so many go after.

Mike Duboe:

Yeah, I am not here to shill portfolio companies, but you did just articulate the thesis behind Paramark, who I know you guys work with, which is a triangulation of all these different methods. I think the truth does not lie in just one.

Tim Doyle:

And they're smart on it. They're really smart on giving you an array of tools to play with and allowing you to choose your own adventure, which I think is really important.

Mike Duboe:

So I want to spend more time on this creative piece. You've talked about the alpha, that there is alpha just by doing creative well, and creative is not just asset generation, it's broader campaign ideation and really end-to-end creative is what I'm referring to. I think in companies and in agencies that's often a blend of process and creativity. And I think AI plays a role in both. Who do you think ... Well, so there's a couple of questions here. One is who do you look to is doing a great job maybe besides Eucalyptus? And then process and creativity, are these two synergistic or are they at odds?

Tim Doyle:

The first one is easy. I think we see a couple of the modern data... So, the brand that inspires me most from a creative perspective is Mischief. The fact that they can do product launches on repeat at scale, hit them, make sure they work, and then capture cultural moments at the speed that they do with the accuracy they do? Phenomenal. Truly the best modern marketing engine that I can see and understand and I just am so impressed every time. So that to me is the most inspirational thing. And I think an example of where you have performance-style principles in terms of moving at speed, capturing moments, scaling through virality, but real discipline in the process side to make sure that product launches actually work. Into the second question, yes, we have two ways to create alpha. We can be the best at producing, turning winning insights into winning ads.

And that can be a repeatable thing. And I think you have to get good at that. And the skillset that is good at that is very different to creating an environment where people who have big and broad range of ideas can bring those ideas to life. And so you have this constant tension between what I would call creative strategy and creative process. And I think managing that tension ... The most frustrating thing about managing that tension is you don't hold onto the equilibrium for very long. You either get overly processed and suddenly there's a million notion docs floating around or you lose all process and suddenly your ads are about something completely random. There's a constant tension that needs to be held between elite process and the freedom to be creative for your creatives. And so I consider it one of my primary responsibilities is to watch over and ensure that there is enough insanity and then that the system that allocates capital and time is as ruthless and effective as possible.

Mike Duboe:

And maybe one of the components to that system nowadays of course is AI, and you've kind of teed up this topic of marketers obsessing over variants with a T at the expense of variance, kind of bold swings. And I think if you think about the role of AI, it could both play a role in helping you swing bigger, really stretch you on out-of-box ideas, or — and maybe more likely — it could just get you down a rabbit hole of more micro-optimizations generating thousands of variants to kind of the same concept and you're more likely to hit a local maximum that way. How are you thinking about AI in this context?

Tim Doyle:

So something Zuck's been saying a lot — like, our dear beloved overlord — has been saying a lot recently is that eventually Facebook will handle all of the creative for you and you'll just give it your gross margin and pray. And I think that's the really scary endpoint of a lack of understanding of what marketing is actually about. I think we've been on this perpetual march, which AI has really accelerated, which I would call the face to camera, hold up the photo, hold up the product UGC era, right? Where it's like you're just getting more and more precise with exactly the small modification to the ad that makes it work slightly better. So I think that being an inevitable endpoint where AI just does the whole thing? Obvious. And that's where Zuck thinks that advertising will go. And I think probably half of advertising goes there because half of advertising is that soul-crushing UGC.

But I think what — and I think the True Classic guy said this in a really fascinating and interesting way is — great modern advertising is the ability to move at the speed of culture and inject yourself into that speed of culture. And I think the ability to use AI to move faster ... His Bigfoot example was bizarre but brilliant. I think the Sydney Sweeney example is as interesting to me, which is things happen in real time and then if you can respond and produce creative in that, you blend the line between news and advertising and culture and advertising, and that's when advertising is most effective right now in the digital world. The anecdote that I always come back to is I worked on the political campaign as my first big digital advertising job in 2016, and we spent $30-$40 million on advertising in the campaign.

And the ad that worked best was on the morning before the election, the opposition leader said, "I can't guarantee no cuts to health care spending." And we cut that straight away, turned it into an ad, and 5 million Australians saw it in the 24 hours before the election and it became a huge election issue. To me, the ability of AI to increase the speed at which you can move and then the breadth and then lower the cost of taking ambitious swings, that to me is where the alpha will remain. And hopefully Zuck's terrifying view of the future doesn't play out because that remains really important.

Mike Duboe:

I want to ask a quick follow-up here, you're talking about speed, it ties back to the process question. So really this is about speed and establishing a culture of speed. Maybe it sounds trivial, but how do you ensure that that trickles across every part of the org? How have you institutionalized this?

Tim Doyle:

So the big thing you need to achieve when you're trying to get creatives to move fast is you need to change the reward cycle from craft to the dopamine of performance and outcomes. So it's really, really hard to do, but our creative team runs on a — even though we're not a baseball country in Australia — we run on a single-double-triple-home run per month system. And so they get incentivized and rewarded on the number of runs they produce in the month. The reason we have the single-double-triple-home run system is because there are many ways to produce a run in the "Moneyball" kind of sense. You could be great at producing singles and four out of five shots you take is a single or you could be a home run person. And ultimately, it's better to be a home run person, but it's harder to be a home run person.

Changing the incentives to really care about getting the dopamine out of performance rather than craft is really, really hard and really, really important. And then once you've got that, then it's about reducing the risk, reducing the cost of failure — I'm sorry, not the risk — increasing the tolerance to risk by reducing the cost of failure. So any company that talks about brand guidelines in the year 2025 is in trouble, in my view. Insert yourself into the moment. If it doesn't work, we turn it off, who cares? And that's very ingrained in the company here at Eucalyptus.

Rishabh Jain:

Okay. I'm going to make a quick comment that I think you'll find interesting and then I'm curious about something. There's a company that I met — and I'm not obviously going to reveal who it is — four years ago where the person was describing such strictness on their brand guidelines that they were unwilling to do other things as a consequence of that. And I remember telling them like, Hey, I'm sorry we're not going to be able to work with you because of this. And I think that they were surprised to hear that a vendor is unwilling to continue the conversation upon hearing that, but I actually think that that way of thinking about brand is — you're totally right — is just not correct. And in fact, perhaps what you're saying is, instead of guidelines, it is more like, Hey, how do we show up when something in the culture changes?

So now going to the question, how do we show up when something in the culture changes in order to capture that zeitgeist? Mike asked a really awesome question around the risk tolerance in the organization. What is it that you do to incentivize people to keep on top of the culture in a way in which is sort of like, I would call it as brand additive as opposed to brand guidelines? What are the things that you do to keep people quote-unquote in the zeitgeist to be thinking in that way? Because again, easier said than done, right? There's a reason why certain companies are so great at it and other companies are not.

Tim Doyle:

Yeah, I wish I was better at this, to be honest. I think this is not something that I would consider myself an elite proponent of because companies do it better than we do it, and I think we've been a little bit timid at times in our ability to get it exactly right. I think where my head is at with it currently is ... The challenge is how do you be really clear about the types of things that you're willing to involve yourself in and engage in? And I think that's really easy to say, taking the positives some way, but I actually think great brands are the ones that are able to have the discipline to choose their moments in a disciplined and kind of selective manner. And then what are the things that we're not willing to do? And what are the things that we're willing to move against the tide on?

And I think those are what makes great modern brands, is the ability to know, not just to ride a trend, but to go, This is the thing that we insert ourselves in and this is the way we do it, and these are the values or whatever the term that you want to come up with there that allows us to do that or places us in those scenarios. I would love to get better at that. I'm doing a rebrand for one of our brands with the pattern brands team at the moment, they're called Little Plains now. And I think one of the things I'm really trying to work out is what does it mean for us to turn up in the weight-loss conversation in different ways? And I don't have great answers, but it really is something that's important. So I think I would urge people to think about it even if I can't direct you as to what the exact answer is.

The challenge is how do you be really clear about the types of things that you're willing to involve yourself in and engage in? And I think that's really easy to say, taking the positives some way, but I actually think great brands are the ones that are able to have the discipline to choose their moments in a disciplined and kind of selective manner. And then what are the things that we're not willing to do? And what are the things that we're willing to move against the tide on?

And I think those are what makes great modern brands, is the ability to know, not just to ride a trend, but to go, This is the thing that we insert ourselves in and this is the way we do it, and these are the values or whatever the term that you want to come up with there that allows us to do that or places us in those scenarios.

Rishabh Jain:

Do you think the challenge comes from, Hey, we just have not had this type of marketing before or it's just too hard to keep up with? And not to always pull the conversation back to AI and AI tools, but is there just an instrumentation issue that leads to the difficulty in being able to be good at this, right? Not saying that this is possible, but why not just understand, Hey, every day, just get a report of like, Hey, here's what people are talking about.

Tim Doyle:

Yeah, yeah. This was the great promise of social listening. I remember when everyone tried to sell you social listening tools as a thing and it's actually not knowing what people are talking about. It's deducing where that story is going to go and how to involve yourself and inject. It's not like ... There's a great documentary on Netflix about Jerry Springer and the producer is the guy that is this awful human but understands exactly how to press the buttons of the audience. And I think, to me, that's actually the skill. An example is something that I'm constantly struggling with is, people's attitudes towards weight-loss medication in Australia are largely, they consider them cheating because they say fitness is a virtue and health is a virtue and it has to be earned. There are constantly moments where someone says something ridiculous and belligerent about obesity. And so how do we choose where and how to show up? Not picking the moment to show up, that's not very hard, but figuring out the shape in which you turn up there is a really big challenge. And this is why it's the golden age of narrative because Dan Wieden or the great advertisers of all time, they're flying right now because it's like you have so many opportunities to tell interesting stories, but we have to get over the myopic "performance marketing is evil" view to actually land there.

Mike Duboe:

Actually, it's worth spending a few minutes here on consumer health. We are in this golden age. So much is happening in health care right now. I think there's more, much higher tolerance for experimentation, self-directed health is an important meta, and I also think the role of AI in health care is interesting in various ways as well as all of our biomarkers and everything are being fed to the models and we now have different levels of control over our own health. Maybe one question for you is, so when you're in a moment like this and you're clearly riding a wave or in part helping create that wave, but that wave is getting noisier with a lot of other players, how do you think about trying to do different stuff and stand out versus just riding and acknowledging that that's going to carry you pretty well?

Tim Doyle:

Yeah, you move fast to ride and then you get well set on the wave and then you plan your strategy for the rest of the wave. One thing I'm always thinking about is, by being chronically online, I tend to be really early to things. I often think I'm late and then I'm actually really early. And so having the discipline to realize that you're early and then just do the thing. Don't be overly ambitious, don't be over creative. Just do the thing. People always give me shit about Eucalyptus being like, It's like a D2C business, right, it's like a commodity product, blah, blah, blah. It's like, All right, man. Well, you build a $500 million business. Because the thing that I now have the opportunity to do is figure out the next phase. And so now that ... The thing about everyone in Silicon Valley has too much SaaS brain where you build the thing and it's the perfect shape and you're able to scale it and it's the greatest thing ever.

Whereas in health care, it's really, really hard to get to scale. So if you get an opportunity that is an ugly shape to get to $500 million of scale, take it and then figure out what to do with those patients. And so now I'm very much in the world of like, Well, I've got the capital, I've got the time. I can figure out product expansion, I can figure out where longevity is going, what longevity actually means in the context of the median patient. But for a long time I was like, Where is next gross margin dollar? And I think certain categories require that discipline. I often talk to people in our team about being truffle hounds for gross margin because if you don't, you die waiting for health care to change and it doesn't really happen in the way that you'd hope.

Mike Duboe:

Maybe sticking on this health care topic, so I guess the power users of a year ago are now kind of the normies today. This is all happening very quickly. People who are experimenting with the Huberman protocols or following Mark Hyman are now the mainstream. How do you think about balancing between the power users and the normies? And is that a delineation that you even use internally?

Tim Doyle:

Yeah, definitely, definitely. I care about the normies. I think our challenge is — and what we do really well — is we remove the friction to action. That's what our commerce layer actually does. If you think about extending the shop, we're an eCommerce business fundamentally. And so you can be an insights business in health or you can be a actions business, and we're an actions business. And so we exist to make it really easy for the median consumer to make the next good decision and do the next right thing for their health. It's very rare that I'm going to be like, Here's the sauna and ice bath protocol that you need to be in. It's like, How do you make sure that when you're hungover, you don't compound one bad day with another bad day? I think about our responsibility and our opportunities to serve the median consumer. And then as trends happen, I will make decisions about whether or not we adopt them into our median consumer platform. And that's because that's the wedge that we have. We have the median 45-year-old woman, and so I could be more grand with my view, but the reality is I serve the patients that I serve.

Mike Duboe:

So maybe jumping back to some of the marketing and creative topics. So agencies, you and I have discussed offline the role of agencies and what they're becoming and what the opportunities are. Actually, there could be a contrarian play here, actually the opportunity to create an agentic agency with high gross margins that really just deeply specializes in the creative process. It could be a good time for that. There's probably a variety of reasons why that's a bad idea, but how do you think about this? If you were not building Eucalyptus and you were exploring concepts of the agency world, what would you be thinking about? And then maybe within Euc as you scale, what is the role of agencies?

Tim Doyle:

As you can tell and know from me that I hate a contrarian take, and so it's not one that I'm going to easily engage with, but I'll try. It feels like to me, the golden age of agencies was very much when they were quite specialized. The holding-group, do-everything agency is a bad idea, and it's proven historically to be a bad idea. What I increasingly think is, I think of my job as a CMO, which I'll never escape, is I consider my job to curate a set of tactics to serve a strategy to grow my company. And so the more tools I have in my set of tactics that are available to me, the better I can allocate between those tactics. And if agencies can be hyperspecialized and world-class at providing one of those tools, I'll buy that every time. I think the age of the small hyperspecialized agency that does organic YouTube content extremely well, it records podcasts extremely well, it makes extremely ambitious TV ads. And unfortunately we've got enough of those agencies, please don't start another one of those, but you know what I mean? The hyperspecialized agency is back. I'm super, super anti media agencies. I just think, I just don't ... It's such a core competency for how I've always thought about building businesses that I just don't see how you could have a search agency. There's no greater negative signal in a company for me.

I consider my job to curate a set of tactics to serve a strategy to grow my company. And so the more tools I have in my set of tactics that are available to me, the better I can allocate between those tactics. And if agencies can be hyperspecialized and world-class at providing one of those tools, I'll buy that every time. I think the age of the small hyperspecialized agency that does organic YouTube content extremely well, it records podcasts extremely well, it makes extremely ambitious TV ads. And unfortunately we've got enough of those agencies, please don't start another one of those, but you know what I mean? The hyperspecialized agency is back.

Rishabh Jain:

It's interesting, I'm trying to square how you're describing the hyperspecialized agency and serving the median marketer with this tweet, if you don't mind if I read, that you put out a few days ago. "The best way to launch martech biz is to aim at the dumb and lazy median marketer. Problem is, people eventually churn when they realize the product doesn't deliver value, much harder to build something top 10% of growth people want to use and then scale." I'm not sure I have a great way to describe the difference between a martech tool and an agency. Ultimately, these are things that a marketer buys to accomplish something. You can say that in the agency's case, you're signing up for an outcome and in the SaaS case you're signing up for a tool, but with AI, most people are describing SaaS as selling outcomes. And so that line is getting thinner and thinner. How would you describe, what is the right way to think about building a business, building a martech company for example, or an agency given this view?

Tim Doyle:

Oh, so I think the difference between what I said in the tweet and what I would advise to building agencies is agencies don't need to have the aspiration to be massive companies. They don't need to build deeper into organizations. And so you can limit your ambition. You can do one thing, serve a normie, serve a great market, or do whatever because you don't have to continue to scale. But if you're going to be a marketing — and churn doesn't matter. If a patient churns, get a new patient. If a customer churns, get a new customer. In the agency world, it's less important. Whereas if you're going to build a great SaaS business, you have to be able to deeply embed yourself and continue to deliver value and scale endlessly with increasingly complicated clients. I think the best way to do that, it's really easy to build something that it looks shiny to a modern marketer, whether it is a large animal, sea animal-based attribution platform, or it's a brand-tracking tool or whatever it is, it's really easy to make something that makes ... give somebody an apparently useful answer for the media market.

But as you get more complicated, it stops being useful. And so people will get bigger budgets, they'll get better, and then they'll churn. Whereas if you're going to build something that's really complicated but serves a really difficult use case, which I think both Haus and Paramark are trying to do in the measurement space, then you have a much bigger challenge. But it's really hard to sell because the media market is going to go, This is too complicated for me to use. I don't want to do it. There's so many examples of this. There's Credo, there's that Finnish business that used a layer on top of Facebook ads. It's really easy to build something that is a veneer that works for a short amount of time. It's really, really hard to build something that actually makes a difference to marketing teams long term.

Mike Duboe:

Yeah. Alright, so we should stay on this because this is ... The category that comes to mind here is the multitude of AEO tools that are out right now, SEO for answer engines. On the one hand you can make the case, Hey, these are very simplistic monitoring products that have a nice UI that are delivering insights that will eventually become commoditized. And that is many people's view or many of the top 10% marketers view. On the other hand, you could say, Hey, this is actually a pretty great wedge to get into these companies. And then over time, you layer on either services or other products that could actually help you influence results. One's ability to do that is still very much a question. But to the extent you've looked at this category, which of these camps do you think or where do you land on that topic?

Tim Doyle:

This is a great software story. It's like does the wedge turn into something valuable or not? This is a question for you in reality because this is the venture question. It's like, do these things turn into something useful? I'll sign up for all of them. Yolo, who cares? They don't cost very much money. I'll have a crack and then if they turn into something good, I'll do it. But that's actually not for me to have to worry about right now. My job as a customer there is to stay open-minded and then buy the value while it's there and I'll buy the insight when it's there. And then if they turn into something, great, awesome. If they don't, I'll churn in one second. That's the story of my engagement in most marketing tech tools.

Rishabh Jain:

And by the way, I think that that's exactly the right behavior for a user. One of the weird things that I've noticed with AI tools is the rate at which they improve is like nothing I've ever seen. So when I was buying software, I mean, we still buy software for everything. It's like, you don't need to revisit something for three years. Now, you kind of need to revisit things every six months. But as a marketer, you're inundated with an endless list of animal-based software, I guess. How do you think about the revisit cycle? So let's just say AEO. Let's just use the same toy example because it's a good one. Let's just pretend it doesn't work for whatever reason. When do you revisit it?

Tim Doyle:

I look at the channel, I think, mostly. I'm like, Is the channel maturing? Do I have a problem to solve at the channel level? And then do I need tooling to solve this problem? Because the channel will get better. A classic example here is the history of Facebook ads buying when there was Kenshoo and Smartly and all of these absurd tools that helped you do things a little bit better. And then the channels got better and I didn't have a problem to solve there anymore s I didn't have a revisit cycle on the tool. So the question will be is how does the Google search console for AI end up working? Who does that? How's it going to work? Do I need help managing that? And then that will drive my revisit cycle, I think, but hard to know yet. Have to figure it out as we go.

Rishabh Jain:

Yeah, that makes a bunch of sense. And then the second question that I have is you versus your team. So another thing that I notice is CEOs are surprisingly good at meddling with AI tools, and it's just not as obvious that the rest of the team has as much meddling desire. When you think about, Hey, here's the thing that I am finding has value for us, person X on the team, can you investigate further and figure out how we best use it? Things like that. How do you think about that? Because we're seeing this very weird thing where even for us as we sell to companies, it's actually easier to sell to — this is unusual in software buying — it's easier to sell to the C-suite than it is to sell to the user. Yeah, so I'm curious how you think about that for operating your own business.

Tim Doyle:

I wish I was better at this. I think the reason ... I that the thing that stands out to me about the CEOs being more curious here is it feels more existential. And so it feels like the right way to spend time because you're actually genuinely scared, and I think — or excited — but scared, certainly paranoid. And so I think that feeling is what drives the experimentation, whereas that feeling in the middle of the organization is often met with defense and also I think a little bit of blindness to what's coming. Creating a set of incentives ... Actually, I think less important than the incentives here is actually the, Hey, this is the right tool for now and this is the one you should go deep on. So actually being a little bit of a curator here and being able to be like, Hey, this is actually — I spent a lot of the first time after ChatGPT came out, being like, We all need to get better at AI.

Why aren't we doing this? I didn't do the dumb CEO memo where AI first, now that everyone seems to do as a LinkedIn marketing tactic primarily, but I did at least be like, Hey, how are we getting better here? And I just didn't see the rate of progress that I would've hoped for. But then I think what changed is I was like, Let's actually think about which specific use cases we care about right now, and let's try and go embed them in rather than being like, Here's a wide array of things that you might want to try. That feels like it is less sticky. But I wouldn't say I've done a fantastic job here. We're not on the front of this. I'm very, very slow.

Mike Duboe:

So Tim, we're running close to time. I want to ask one last question that zooms out a bit, and as I'm reflecting on this conversation, just our conversations over the years, I think you strike this really unique balance of just having a deep appreciation and kind of grounding in classic marketing foundations. Yet you're also not afraid to try anything and just blow up any assumption and you're risk-seeking. And I think that balance is hard to find. With that in mind, if you zoom out five years, how does your marketing operation look different? Or does it actually look more the same than different? What do you envision Eucalyptus' marketing machine looking like in five years?

Tim Doyle:

I think we are in a really great spot. I think it's like, there is great returns for doing compelling storytelling at scale in ways ... Like Mike, the era that me and you grew up in is like, it was so boring. Like, oh, is it exact match or is it broad match or is it this or that? Or is it like, oh, it turns out that most companies don't know how to do retargeting. There's an incremental $10 million. Great. Terrible time. Great. And it rewarded the worst type of marketer, which I was, completely uncreative, grindset arbitrage. Now I think what the reward is is really, really working hard on the creative side and rewarding that, and then working really hard to measure behind how that works so that you can allocate capital in a really interesting set of ways. And I think the new channel arbitrage is more interesting than it's ever been because you're like, How do fortune cookies work?

Or how do skywriters work? I hate the "billboards in San Francisco" thing. The reason that worked for some people is because it was an arbitrage at the time. It's now not a thing. Sorry, that was a segue. But I think basically finding new channels, doing interesting storytelling in those new channels, and then being really analytical behind that is such a clear set of priorities that a team has to be able to do. And you just evolve and evolve and evolve and get better and get better. And then maybe there's a hundred new channels we don't know about yet, the content will get faster and different. And so I'm so excited. I think it's a great time to be a well-rounded marketer. I look forward to doing that. And then I think we talked about potentially going through a channel that I want to see not exist. I want to see LinkedIn go away, please. That's my LinkedIn content organic posting. That's my number one. My God, that feed is hell. So that's how I wanted to close out.

Mike Duboe:

Tim, it's always a blast chatting with you. Thank you. This was great.

Tim Doyle:

Thanks guys. Thanks for having me on.