Mastering paid growth | Jonathan Becker (Thrive Digital)

Jonathan Becker (00:00:00):
There's a lot of different ways that we are beginning to use AI to do more with less, basically. The effect ultimately that we've seen from a human capital point of view is displacement. We have more people now than we've ever had, but the nature of the work that they do is more strategic. It's more about modeling, validation, asking the right questions, being focused around creative levers. And less so the like trench work of implementation and bid modifiers at the keyword level on Google search, and some of the really hardcore manual analysis we had to do.

(00:00:35):
On our creative group, we can come up with mockups, in literally, 1% of the time that it took. And so you still have to understand what questions to ask of the AI and be capable of iterating, but these rough drafts that you might show the artwork of to a client to say, "Do we like this more or do we like this more?" That's AI generated. It's really interesting.

Lenny (00:01:01):
Welcome to Lenny's Podcast, where I interview world-class product leaders and growth experts to learn from their hard won experiences building and growing today's most successful products. Today my guest is Jonathan Becker. Jonathan is a legend and an OG in the world of performance marketing. On this podcast, we've done deep dives into the many aspects of growth, including SEO, sales, conversion optimization, retention, product led growth, product led sales. But this is the first episode where we get super deep on paid growth. For the past decade plus, Jonathan and his team have planned, built, and executed more than $3.5 billion in paid acquisition budgets for companies like Uber, Asana, Square, Masterclass, Tempur-Pedic, and many more. And they've built their agency, Thrive Digital, into one of the preeminent independent digital marketing agencies. In our conversation, we get real deep into all things paid growth, including what's changed with recent privacy shifts, why focusing on creatives is the new biggest opportunity within paid growth. How to think about attribution, and what's changed there. What to look for in people you hire to run paid growth for you, how AI is already changing how paid growth teams operate, and so much more. Enjoy this episode with Jonathan Becker after a short word from our sponsors.

(00:02:16):
This episode is brought to you by Brain Trust, where the world's most innovative companies go to find talent fast so that they can innovate faster. Let's be honest, it's a lot of work to build a company. And if you want to stay ahead of the game, you need to be able to hire the right talent quickly and confidently. Brain Trust is the first decentralized talent network where you can find, hire, and manage high quality contractors in engineering, design, and product for a fraction of the cost of agencies.

(00:02:43):
Brain Trust charges a flat rate of only 10%, unlike agency fees of up to 70%, so you can make your budget go four times further. Plus, they're the only network that takes 0% of what the talent makes, so they're able to attract and retain the world's best tech talent. Take it from DoorDash, Airbnb, Plaid, and hundreds of other high growth startups that have shaved their hiring process for months to weeks at less than a quarter of the cost by hiring through Brain Trust network of 20,000 high quality vetted candidates ready to work. Whether you're looking to fill in gaps, upskill your staff, or build a team for that dream project that finally got funded, contact Brain Trust and you'll get matched with three candidates in just 48 hours. Visit usebraintrust.com/lenny, or find them in my show notes for today's episode. That's usebraintrust. com/lenny for when you need talent yesterday.

(00:03:35):
Today's episode is brought to you by AssemblyAI. If you're looking to build AI powered features in your audio and video products, then you need to know about AssemblyAI, which makes it easy to transcribe and understand speech at scale. What I love about AssemblyAI, is you can use their simple API to access the latest AI breakthroughs from top tier research labs. Product team that startups and enterprises are using AssemblyAI to automatically transcribe and summarize phone calls and virtual meetings, detect topics in podcasts, pinpoint when sensitive content is spoken, and lots more. All of AssemblyAI's models which are accessed through their API are production ready. So many PMs I know are considering or already building with AI, and AssemblyAI is the fastest way to build with AI for audio use cases. Now's the time to check out AssemblyAI, which makes it easy to bring the highest accuracy transcription plus valuable insights to your customers. Just like Spotify, CallRail, and Writer do for theirs. Visit AssemblyAI.com/lenny to try their API for free, and start testing their models with their no code playground. That's AssemblyAI.com/lenny.

(00:04:46):
Jonathan, welcome to the podcast.

Jonathan Becker (00:04:48):
Thank you. It's a pleasure to be here.

Lenny (00:04:50):
It's my pleasure. What we're going to be doing with this episode, is we're going to be digging real deep into all things paid growth, which we've never done on this podcast yet. And normally I actually skip this part, but I thought it'd be actually helpful for you to spend a little bit of time to give us a little bit of background on your experience in the space of performance marketing, AKA paid growth, AKA paid ads. You tell us what the right term is for this genre. But yeah, just tell us what you've been up to in this area over the past decade and a half, I think.

Jonathan Becker (00:05:20):
Sure. No, that's a great way to kick things off. And again, thank you for having me here. If I think about it, my experience goes back about 15 years into this space. I started off as a web developer. And as I built and structured websites for people, I became pretty obsessed and fascinated with the fact that you could build landing pages or homepages, or whatever it was. Basically the content on a website, and structure it in a manner where you had the chance to surface in organic, so SEO results more prominently. And as I became a practitioner of SEO, SEO really being my first love of marketing, I started attracting attention and people wanted to hire me on as a freelancer. And what I noticed is that as people started asking me questions like, what was the ROI of our SEO campaign? Or, how do I scale this, or whatnot? The answers to their questions ended up being a lot more aligned with what then was the biggest driver in the paid acquisition world, which was paid search.

(00:06:25):
And so I started experimenting with paid search. And what I found was that it was a tangible format and lever through which we could basically give people the types of results they were expecting from SEO, but that were obfuscated in terms of Google's analysis algorithm from an organic point of view being somewhat intangible. And Google slowly removing a lot of the data early on that allowed you to guess and test more easily. 10 plus years later, what started off as this freelancer consultancy that I started running in a walk-in closet in my old apartment, became 130 people. And we manage about $500 million a year in ad spend for small and large companies, including Uber, Asana, Tempur-Pedic. We've worked with Lululemon. Very exciting companies, mostly from the United States, even though we randomly happen to be based in Vancouver, Canada.

Lenny (00:07:26):
There's a number of threads I'm going to pull on there over the course of our chat. But you mentioned Uber and you told me that you had a crazy story about how you actually landed Uber as a customer. Could you share that?

Jonathan Becker (00:07:35):
I had been running Thrive for a couple years, and it was a very excellent regional agency in Canada with really cool local clients. In 2013, I got invited to go down to the TED conference, which was in Long Beach, California. And my friend Andrew Wilkinson asked me to join him at a dinner that night. I didn't know anybody from the TED community at the time. So we have dinner, and then afterwards as it goes at conferences, there's an after party. And so essentially I hop into a taxi, everybody else sped off in their cars, or however which way they were getting there. And as we're talking to the driver saying, "Hey, we're going to this place, can you take us?" There's a knock on the window and the person outside says, "Hey, I think I'm heading to the same destination. Do you mind if I hop in this car with you?"

(00:08:25):
And so we're like, of course. And I turn to him and I'm like, "I'm Jonathan. I run a 10 person agency out of Vancouver." And he says, "Hey, I'm Garrett Camp. I started a company called Uber." And so ironically, I meet the founder of Uber, the company that is in the process of disrupting the entire taxi industry worldwide in the back of a taxi cab. And what happened next basically changed my career forever. We end up at this party. At the time I was spamming Uber's referral program. Kind of a long story, but essentially I was using paid search to camp out on their branded keywords. And as people would sign up with my confusingly similar snippet to Uber's organic snippet, I was essentially siphoning off referral credits. So I would get $20 every single time someone signed up.

(00:09:18):
And I ended up making tens of thousands of dollars doing this. And so fast-forward, I'm getting a drink at the bar next to Garrett, and in my head I'm like, should I tell him about this? Maybe I can land them as a client. This would be really interesting. And essentially I tell him, I'm like, "Hey, I'm doing this. I'm adding zero value, but this is a loophole in your marketing system, and someone should close it." He essentially is like, "I need to report this to the board, but here's my card, write me what you're doing and we'll contact you." And so I get contacted by a bunch of his lieutenants. If you've read the book, Super Pumped. All the people that we dealt with at the time are in the book, and [inaudible 00:09:58]-

Lenny (00:09:58):
I watched the show.

Jonathan Becker (00:10:00):
Yeah, exactly. And they were like, "Hey, this is bad, you have to stop doing this. But can we hire you to solve the problem?" And so what started off as me running projects for local bars in Vancouver, or clothing stores, or whatever it was. Turned into me landing early stage Uber as a client, and really graduating us from competent professionals to leaders in our sector. And so it was a fascinating project, and we worked with Uber for 10 years.

Lenny (00:10:32):
That is an incredible story. I love the arbitrage game you're running there. It's basically siphoning VC money out of Uber. And I guess the lesson there a little bit is just sometimes it's this interesting combination of hustle in terms of just make some money, and also taking advantage of this opportunity you were plopped into.

Jonathan Becker (00:10:53):
Yeah. I think people often talk about entrepreneurs who have been successful and they comment that they're lucky. Whereas, I actually look at that situation and I think that you have to make your own luck. I could have been like, oh, cool, I met this guy in the back of a taxi and that was it. But I decided to take a risk, being that I could get embarrassed or nothing could happen, or they could shut down this referral gimmicky thing that I was doing. I had very little to lose, I guess, ultimately. But a lot of people just don't make these moves in life because they're nervous, or they're worried too much about what the downside might be. And so I very much was like, I'm going to shoot my shot here. And you put yourself in situations where everybody has luck, but you have to capitalize on it basically. And so that was an example of being willing to take a risk, and it paying off pretty big time I think.

Lenny (00:11:46):
Also, being at Ted, that seems like a good move. Networking, paying off. Love that he was in a taxi, that's hilarious. I was going to ask you about that. And clearly, he's doing some research [inaudible 00:11:55].

Jonathan Becker (00:11:54):
Yeah.

Lenny (00:11:58):
Let's start diving into the world of paid growth. And if you think about just paid growth.. And again, actually, what do you refer to this area as? Paid growth, performance marketing, paid ads?

Jonathan Becker (00:12:08):
There's a lot of interchangeable terms. Performance marketing is a common term. Paid acquisition is a common term. Some people think of those two things as growth marketing, whereas I see growth marketing as a bigger practice area within which paid acquisition sits. And then of course there's subsets, there's social ads, there's paid search, there's programmatic. And so there's a lot of different ways of saying the same thing.

Lenny (00:12:31):
I'm going to use performance marketing. I like that term because it really describes what it is. It's like marketing with you can measure performance. Let's see how that goes. As a channel, it's such an interesting mix of, on the one hand it's this incredible growth lever that allows basically any company to spend money and understand the ROI and acquiring the users. This thing that never existed before. Essentially Facebook and Google created these platforms. On the other hand, there's this sense that it's this drug that you start and then you get hooked on and you can never leave. And there's a lot of advice of just, avoid paid growth. That's just not a good healthy way of growing, especially as a startup. And so my question to you is just, how do you think about that element of it? And then even more specifically, what products do you find paid growth as a channel is right for and not right for?

Jonathan Becker (00:13:24):
Yeah, it's a really great question. And so, I think there's a couple different things that need to be unpacked here. Paid acquisition or performance marketing-

Lenny (00:13:33):
Performance marketing. [inaudible 00:13:35]-

Jonathan Becker (00:13:34):
... you can call it, can be seen as a drug, I suppose, when you are entirely reliant on it to fuel the revenue of your business. The analogy that I try and use here is that it would be very dangerous if I was advising you with your life savings and I told you to put it all in a single stock in the stock market. Stocks can be volatile. And as a result of that, your net worth would fluctuate quite a bit in the short-term and the long-term based on a lot of things that you don't control, like the external markets, or things that are happening within the performance of that particular company that you invested in. When I think about all of the marketing mix, so email, direct mail, linear television, performance, marketing, whatever it is, I think about it as investing capital with the expectation of a return. And in the same manner that I would not take all of your life savings and dump it into a single stock, I don't recommend putting all of your money into a single performance marketing channel, and then somewhat exposing you to the volatility of fluctuating CPCs or changing market conditions. I would agree with you that it is a drug, in a sense, if you have all of your eggs in one basket, and that basket takes you on a very intense rollercoaster in terms of performance. But when I think about the fact that Thrive manages $500 million a year, I think of myself to a extent as a fund manager, we are managing people's money with the expectation of a return. And part of the strategy is to diversify across channels. And so we decrease the reliance of any individual project on a singular channel and its performance.

(00:15:26):
And similarly, I always say to people that the first rule of performance marketing is not to forget about offline marketing and the classic marketing that works for organizations. In other words, direct mail can really work, email marketing works beautifully, SEO can really work. There's all of these wonderful things at your disposal. The real crash and burn scenarios that I've seen are these, not fly by night brands, but brands that figured out... Just like with Uber, I figured out this weird tie-in where I could make free money from their referral system. Sometimes people find shortcuts, hacks if you will, to scale rapidly because of one specific nuance of the Facebook ad platform, or something like that. And what they fail to see is that those loopholes come and go.

(00:16:19):
And so if they scale massively and their entire business is predicated on the performance of this one loophole that they found, or investing everything in a single channel, and then the conditions change, they're not going to be very happy and the business will suffer dramatically. And so when I think about this, it's a responsible channel mix, diversity, and understanding that you can't be overly reliant on performance marketing for the success or failure of your business.

Lenny (00:16:48):
What about the second part of that question of when you think about when companies come to you. What do you look for to help you understand this is going to be a really good fit for performance marketing and this is going to give them a lot of opportunity to growth, versus maybe not, maybe it'll be a small sliver but it's not going to be a massive success for them?

Jonathan Becker (00:17:06):
Yeah. I would say that the answer to that question is different at different stages of a company's life cycle. Early stage, look at what the company is doing, look at your own company. Have you established product market fit? Is this an idea that has yet to be tested, and are they entirely looking to performance marketing to scale everything? Are they at risk of it becoming an over-reliance on performance marketing? At a later stage we look at certain criteria that they might or might not possess. Typically, that will come down to resourcing.

(00:17:43):
The question at a later stage is not, does it work? But, to what scale can it work? And so we are looking at things like, do they have adequate creative resources and buy-in? Does creative resourcing tie into performance, and can we create a feedback loop there around testing? I'll talk to you about that in a bit. Do they have professional marketers on staff? Are there people who have experience doing what we do, that speak our language so to speak? Or is part of this an organizational educational, and creating buy-in through stakeholders process that needs to take place? Do they have technical resources? If we say, "Hey, tracking and attribution is broken, here's how to solve that." Can you implement it? And so on and so forth.

(00:18:27):
There's no one magic formula for what works or what type of company will be successful on performance marketing channels. Just as evidence of this, Google, which I think had a down quarter reported sales this week, and it was $70 billion in three months. And Facebook similarly just recorded yesterday evening, and it was $32 billion. That's $100 billion on just Google and Facebook in three months, and the majority of that revenue is from ads. This works really well for lots of different companies, it's just a question of at what scale.

Lenny (00:19:06):
Just to pull an thread a little bit more. Something I've heard from other guests is that paid ads are best for products where you get basically payback really quickly, basically to feed the flywheel of spend so that you're not sitting around waiting for someone to buy something in the future, or it's a small trickle of pay. How important is that, I guess? Do you find that you could just do paid growth for any company, it doesn't matter their business model? Or is there something you're like, okay, this needs to exist for you to invest serious resources? And even hire that team that you just talked about.

Jonathan Becker (00:19:37):
It's always nice when there's a quick turnaround on investment and return. And that's wonderful for like D2C or e-commerce style businesses where they're essentially taking the revenue to fuel additional inventory and operating costs. However, not all businesses work that way. And so, in a B2C or B2B lead generation scenario, we have to undertake pretty sophisticated modeling around these abstract concepts, like lifetime value. Which is difficult, because most businesses are relatively new that we work with, and so the idea of lifetime value is a misnomer. They don't know what that is yet.

(00:20:15):
But we have to model things like LTV to CAC, so cost per acquisition costs versus lifetime value and the period within which the payback occurs. You end up getting to a pretty sophisticated place where you can build out things like a lead scoring model, which predictively can determine in a statistically significant way the likelihood that one lead will convert to revenue over another. And so there are ways around the slower payback period that still end up being pretty accurate based on what you're bidding on today versus the latent revenue that will be accrued to those campaigns through attribution.

Lenny (00:20:58):
Do you have any just rules of thumb for someone listening and trying to decide, is paid a real model for us, either on LTV or CAC, or payback period? Or something just like, here's what you probably should have, especially early stage for you to feel like paid growth is going to be a great lever for you to use, and maybe a primary lever.

Jonathan Becker (00:21:18):
Product market fit. If you know that your business sells into audiences. Let's say you are a social media influencer or you had a really strong email marketing game, or organically your content surfaces within Google search results. Or, you did a lot of direct mail and linear television and billboard advertising and that worked. If other things work, it is highly likely that paid acquisition will work. The issue for most companies is in this incorrect assumption that the data that is provided through paid channels allows you to have full end-to-end understanding of attribution. Which is wrong, it's never been that way. And the other aspect of this is the patience to understand that every business is unique, and these metrics that we know are important are different for every business. Lifetime value, like propensity to repurchase, ROAS, which is return on amount spent, CPA, CAC, all of these different things are different for every business.

(00:22:24):
Even if I worked with two hotels in the same city, they would have different results based on the nuances of their budget, their brand, the market that they sit within, the services that they offer, and so on and so forth. I think that everything else, the main problem here is that nobody should expect an overnight turnaround with performance marketing. It is a very difficult channel to manage, and that's why people hire experts like us to help them with it because it's a never ending problem with constantly changing issues. It's always been like that, that's not a new thing since pandemic, or whatever. And it will take some time to work out what works.

Lenny (00:23:07):
I have this framework of there's these four growth channels, basically growth engines is what I call them. There's paid/performance marketing. There's SEO, there's virality in their sales. And essentially there's some companies whose growth is almost primarily paid. A few that come to mind are booking.com, which we know well at Airbnb, which is almost all paid growth driven. Credit Karma comes to mind as a classic paid performance marketing. I keep coming to paid growth as my term, so I'm just going to stick with that. TikTok initially was very performance marketing, paid growth oriented. Wish was another one I think about.

(00:23:42):
And I want to talk about how much things are changing within this realm. But before we get there, do you think there's still an opportunity for startups to emerge where they get to scale almost exclusively through performance marketing? And this question actually came from Twitter, someone tweeted this randomly the other day. And I was like, oh, that's a great question for Jonathan. And by the way, her name is Liz.

Lenny (00:24:00):
And I was like, oh, that's a great question for Jonathan, and by the way, her name is Liz Georgie asked this question, so there you go.

Jonathan Becker (00:24:06):
I'll put it this way. Every unicorn from the 2010s era that scale did performance marketing, but not everyone during that time who did performance marketing scaled. So I want to remove the bias here that just because all the successful organizations did this didn't mean that it was a magical channel for everyone. We had plenty of projects that we worked on that flatlined during that period. And so the sense that there was a period of time where this was easy or it worked on any project is not correct in my opinion. With that said, we see companies that are spending millions of dollars a month on performance marketing channels like Google and Meta still, despite all the ups and downs that they have faced and they do so profitably.

(00:24:55):
And I think there's some really great examples of companies that have scaled in relatively recent times, almost exclusively through paid. Grammarly is a really good example of this. They have been good at solving for this problem that exists around understanding the cost per acquisition versus lifetime value, how sticky customers are, predicting how much revenue can come from a customer and backing out into therefore how much we can pay per click and per lead and so on and so forth. Athletic Greens is another good example. So Athletic Greens is actually a pretty old company. I think they've been around for 10 years. I think they have retail distribution. I think they have done a lot of the more classic marketing things that are important in terms of developing that channel mix. But I think the amplification of that brand really, really gained traction quite recently where now everybody knows what Athletic Greens is, and that's because they're buying loads of ads on TikTok. They're buying loads of ads on other social channels like Facebook. They're investing in podcasting partnerships. But this is all digital paid acquisition.

(00:26:07):
And so it had a wonderful effect on this really interesting business that they had already built. So yes, it's still doable. We still see people doing it, and I think that there's been a bit of a reckoning in the performance marketing industry pertaining to things like privacy and the changes that Apple made and people being very creeped out at how their data is being used rightfully so. And then obviously the economy in 2022, we had a terrible macroeconomic shift where interest rates rise and inflation's out of control. And so of course the first thing that people cut are typically marketing budgets and we see Facebook and Google and other ad channels directly suffering from that. So all of that said, these storms pass. And so when the economy improves, generally speaking, I imagine people will go back to trying to find as much inventory from a pay per click point of view that they can purchase as possible and figuring out the economics of how to do that.

Lenny (00:27:08):
I was definitely going to ask about that, and I love that you touched on it, just clearly a lot is changing in paid growth/performance marketing recently. You talked about the privacy stuff, you talked about COVID kind of shifted the way people spend and kind of dropped and then came back. So my question is who are you finding has the most success these days in performance marketing? And I will plant one seed, which from the examples you just shared, it feels like it's mostly companies that are very efficient. I think Grammarly, they're just super efficient as a business. And then I think Analytic Greens, they're a sponsor and their negotiations for sponsoring is just like, okay, here's the number that makes sense for them financially, and they're not going to go anywhere above that.

Jonathan Becker (00:27:49):
Yeah. Because they know, they understand how many impressions they'll get and on average what the quality of impression is and how many dollars they can put behind that before it has a cliff in terms of ROI.

Lenny (00:28:05):
Exactly. So broadly, who do you think is having the most success now with the changes and then generally, what should people know about what has changed in the past say year in the space of performance marketing?

Jonathan Becker (00:28:18):
I'm not going to really be able to point to this company is really getting it right and I think that you can do that, but it's like saying, a friend of mine who recently on Twitter tweeted this funny thing where he was like, this is the number that I used to win the lottery said every successful founder ever trying to give advice to other founders. And so in other words, just because they won the lottery doesn't mean you're going to be successful picking the same number.

(00:28:43):
To that extent, there is a bit of a playbook around modern day performance marketing, and that includes everything from really stringent and rigorous creative testing and thinking about that correctly to understanding the subjectivity of attribution and its strengths and weaknesses, doing a lot of work around validity of these campaigns. So really just the measurement and whatnot. Companies that can do those things and then understand their own marketing economics, in other words, quite basically how much can we afford to spend in acquiring a customer on any channel before that acquisition is no longer profitable. So really focusing on the profitability of the bottom line and not just break net growth, for instance. Companies that have those capabilities and see the world that way tend to be successful in performance marketing.

Lenny (00:29:39):
When you were giving your intro, you talked about how you initially started doing SEO, that's where you started, and then you moved to paid growth. Well, how do you think about those two investments as a founder trying to decide which direction to go? What would your advice be of spend your time here versus there if X, Y, Z?

Jonathan Becker (00:29:56):
When I'm asked a question of should we put money into organic search or paid search? My response is often that they're actually not mutually exclusive to one another. So it's a great idea to do both, and that backs into my strategy of diversification of channels. So don't build up exclusively in one area and create volatility within your marketing mix essentially. SEO is a wonderful marketing capability when it's built out correctly. I think the issues that you run into are cause and effect related. So one of the things that people really like about performance marketing theoretically is that we can spend a certain amount and then if we're modeling things correctly, we can essentially determine how much revenue is generated from the actions that we took. Finance teams love that, C-suite teams love that, they can build projections, they can budget around it. There's some degree of predictability around it if it's done properly.

(00:30:52):
Search engine optimization is different in that the attribution can be tough. It's difficult to determine whether ultimately the actions that you took contributed to a rise in organic traffic. You have to essentially correlate that. And the reason is because when you build clusters of content and it's grouped thematically and you're targeting buckets of keywords, whether they're long tail or head keywords, whatever it may be, you can publish all of that on your website. Google still has to crawl it. They run it through their analysis algorithm, which is comprised of 200 different signals, of which maybe 20 to 30 have been publicly disclosed. So it's a bit of a black box.

(00:31:37):
We don't really know ultimately what ROI comes off of that unless you're very sophisticated, like probably one of your other guests, Ethan, in terms of measurement. Whereas paid, the ROI is still a difficult problem to solve, but there was a lot more of a linear relationship as it relates to attribution. And so paid being tangible was the reason why I leaned heavily into paid and ultimately away from SEO. But I do think that if you do SEO properly, the payoffs are indisputable and it is certainly an important part of a modern media mix.

Lenny (00:32:18):
You mentioned this earlier, I've heard this more and more recently that one of the biggest levers these days in paid growth is around creatives. It's not tooling or smarter data, or you tell me if I'm wrong, but it's just getting better creatives. And so I'll let you actually explain, what are creatives? There's a term creatives that people outside the industry don't really necessarily get. And then broadly, what should people be doing to optimize the way they approach creatives?

Jonathan Becker (00:32:46):
Certainly. So when we say creative, we're referring to the assets that power typically visual programmatic or pay per click campaigns on social channels or display networks. And so literally the motion graphics ad that you see on Facebook, the user generated content that you see on TikTok or a static ad that you might see that has a pretty picture in it or whatever. And so I think that when we talk about creative as a big lever around efficiency and optimization, the underlying conversation there is that over time our industry has been heavily automated. So a lot of the levers so to speak around performance have been automated by Google and Facebook over the last seven to 10 years. That's because originally when you ran these campaigns, you needed to have a rocket scientist in front of them. It was so complicated and there were so many different things that you could get wrong, and their solution, the channels like the big tech company solution to this is figuring it out for you. So eventually Google wants you to say, hey, I'm Google, give me your credit card and I'll take care of everything else. Facebook-

Lenny (00:33:57):
And the URL to point people to, and then we'll do the rest.

Jonathan Becker (00:34:00):
Yeah. I don't know if that's a great idea for consumers, by the way. But in the meantime, there's certain things that have just been fully automated. In the context of creative, it's still one of these things that for now is not auto-generated in the world of AI and all the changes that we're seeing. Maybe that's something that will change. But for now, essentially creative directors and their teams are concepting and producing different types of assets. And so there's a bunch of problems that we typically see when people come to us. So number one, performance marketing and brand marketing in a lot of organizations are two different things. And the designers that occupy brand teams bandwidth and whatnot often don't have a sense of how paid acquisition works.

(00:34:48):
And so one of the pitfalls of working with certain companies or the mistake that they make is the design team will hand off a file full of random assets for paid acquisition without any sense of how the channel works. And what I mean by that is these days we're using the analogy, the classic analogy of the funnel to organize our thinking around creative assets. So you can think about this as generating intent at the top of funnel and capturing intent at the bottom of the funnel. When I think about an experience that I want a consumer to have on Facebook, audience targeting and creative, I think about us beginning a conversation at the top of funnel creatively with an audience, having that conversation change as we say different things, and the audience that we're targeting ultimately graduates through different behaviors on our website from one to another, and then ultimately it resulting in an end to the conversation where they take an action hopefully that the brand that we're working with is looking for.

(00:35:56):
And so there's a clear beginning, middle and end to that. And one of the major pitfalls that we see is that certain brands just dump one homogenous message into all of their targeting. You see the same ad over and over again. It creates banner blindness and it's a total lack of efficiency. The antidote to that is to have resources dedicated to paid and essentially iterate upon the creative assets themselves based on the data that we see coming from ad sets and campaigns in various channels. And so what that means is that you have to experiment. You have to take a bit of a scientific approach, although it's a bit of an art and a bit of a science. You have to try and isolate variables, maintain similar conditions across targeting, and then determine which style and feature of an ad performs best at which stage of the funnel versus which audience. And the results that we see are dramatically different from brand to brand. But if you are not undertaking rigorous testing in conjunction with how you are driving the iteration and design of your ads, then you will not make progress essentially.

Lenny (00:37:11):
Is there a specific example that comes to mind here as something we did that just dramatically changed the impact of a change to a creative? Or if you can't think of one, are there just specific tactics that you can suggest for people to improve the way they approach creatives?

Jonathan Becker (00:37:28):
Yeah. So I think from a testing point of view, let's say I was running ads on Meta, beneath the campaign level when I create a campaign, the structure that my testing might take would be that I would have an audience, a single audience at the ad set level, and then I would have two nearly identical creatives within that ad set. The only thing being different across those two creative assets is a single variable. So it might be the copy, it might be an image or whatnot. That allows us to isolate a lot of variables and really test into one singular change across two creative assets. There's a lot of nuance to this. So sometimes the ad-serving algorithms, even when we set up a test structurally in that way, we'll serve one ad a different number of impressions than the other ad, in which case we then have to say, what is a leveling factor that allows us to look at these two ads equally, even though one received dramatically more impressions than the other?

(00:38:31):
This becomes where it becomes subjective in terms of how you want to determine success. But a good example is looking at the click-through rate, which is essentially a ratio or a metric like impressions until conversion, which is a leveling metric that allows us to determine, even though in a scenario where two ads facing the same audience within the same campaign received different numbers of impressions, we can still measure the efficiency or effectiveness of one ad over the other using metrics like that. And so from a testing point of view, I think that this is one way that we might look at trying to assess ad performance so that we can gather learnings, send that back to a creative team and say, hey, it turns out that when this copy is used at this stage of the funnel, it converts 50% more frequently than this other copy.

(00:39:26):
So let's now take that copy, use it as our base copy and challenge it with a different type of copy and see if we can continually iterate and refine. So that's a very practical example of how ad creative testing might work on a channel like Meta. You asked for specific examples where we've seen an unlock. There's two that come to mind. Several years ago we realized that highly produced ads from brand teams, and there's nothing wrong with brand teams, we work with them all the time. They do amazing work. I think we're just trying to work as a singular unit as opposed to being fragmented. But a lot of the brand guidelines of different companies would end up yielding these highly polished assets. And when you launch those on Instagram back in the day or something like that, what we found is that they would always underperform next to user generated assets.

(00:40:23):
So a brand that essentially has an influencer in front of it that says, I tried this product, I love it, I'm filming this ad from my iPhone. Look, check it out. Here's the product, here's me, I'm better off for it. Whatever it is that they're talking about. The unpolished iPhone, mobile phone creative, suddenly we realized massively outperformed these other channels because there was an authenticity to it. And rather than the brands themselves saying, hey, trust us, our product is great, here's a third party essentially validating what is so great about these brands, basically. So I can't speak to specific clients because I'm not allowed to talk about the work that we've done in large part, but that would be one example. And another example is we worked with a furniture company several years ago, which scaled dramatically, but they were having difficulty early on thinking through how to scale social ads.

(00:41:21):
And so paid search worked really well for them. Social ads, again, they had these highly produced styled rooms and one of the owners had their dog in the office all the time, and so they have these models sitting on furniture or whatever it was, and they would take these beautiful styled shots of their furniture in a room and one of the art directors was like, put the dog on the couch and let's take some photos of dogs on furniture, which made it more playful and approachable. That one change resulted in a total unlock of their performance on paid social channels. It would double or triple the ROAS that we were seeing type thing. So it's these minute differences that you can test into that ultimately can drive performance.

Lenny (00:42:04):
That reminds me of Airbnb. The photography team found, and actually the photography team was initially including people in the photos of listings, and that was the [inaudible 00:42:16] images of listing photos. And it turns out when there's no people in the listing photo, much higher conversion. And the theory basically people don't want to see other random strangers in a place they're going to sleep, and that was not expected.

Jonathan Becker (00:42:29):
That's a really great example. And so the question becomes, as a company and as a brand, do you arrive at those learnings anecdotally by accident? Is it the brilliance of one person on the team that the whole brand is predicated upon in terms of performance? Or do you put a very rigorous structure around testing and the iteration of assets and how to determine whether something works over another thing that is a process that yields the outcomes that you're looking for? And so a lot of times with agencies, people talk to us about the obvious hard skills that we have, but the soft skills are around process that can be deployed, that can actually be the difference between being highly sophisticated or unsophisticated. Internal teams do this too, but that Airbnb example is a great illustration of how if you put the right testing in place, you can yield an outcome that becomes an unlock for performance.

Lenny (00:43:28):
Getting even more tactical with that idea of testing a bunch of creatives makes me think about a story from chatting with a guy from Wish who's had a growth at Wish about how they uploaded hundreds of variations. And I think Wish is the epitome of the opposite of brand highly polished ads or it's just the most ridiculous looking product with just banners and numbers and prices and crossouts and all kinds of stuff, just anything that would take to get people to click. Is there a tool or process that you can point people to to help them do this testing on creatives or is it mostly built into the existing tools now and there's nothing really fancy about it?

Jonathan Becker (00:44:04):
We get asked this a lot and I think there are tools for different types of functions that we undertake in the process of running paid acquisition for different brands, but these days most of the testing that we do is within the channels themselves. So Meta ads, Google Ads, I keep touching on those two specifically because they're the most advanced in my opinion. They're really powerful. The good news here is that you don't necessarily need some exotic tool to do sophisticated creative testing. You can do it in platform and you can use the structure that I just kind of outlined as a starting point.

Lenny (00:44:43):
You mentioned Google and Facebook. What is your take on TikTok and YouTube and maybe Snap? Do you recommend people go there and there's a big opportunity? Do you think it's overblown? Do you think people should just take Google and Facebook basically at this point? What's your general advice?

Jonathan Becker (00:44:59):
So when I say Google and I say Facebook, I'm talking about the Google ecosystem of channels inclusive of YouTube. And when I talk about Facebook, I'm talking about Facebook, Instagram, and probably eventually WhatsApp, which I think will probably launch ads later this year. With all that said, some interesting opportunities these days really exist on channels like Amazon and TikTok as well as a number of other challenger channels I think. Snap I'll talk a little bit less about because I think really there's just fewer and fewer people potentially using Snap. There's no knock on Snap. I use Snap, I like it a lot, but at the same time, most individuals or organizations tend to want to place ads where there's a lot of views and engagement and the king of this right now is definitely TikTok. And so TikTok faces its own set of challenges and arguably there is as much an opportunity there as a barrier at the moment.

(00:45:58):
I love TikTok. I'm a consumer of content on TikTok, not a producer of content on TikTok. Their ads platform kind of reminds me of where Facebook ads was like six, seven years ago. It's not super sophisticated and attribution is not great yet, but you can get cheap CPCs because there's just fewer organizations advertising there. Fewer companies have figured out how to make that work. And same thing with Amazon. Amazon is a bit of a bespoke channel. It doesn't work for a B2B SaaS company. It's more specific to D2C I would say. But essentially those channels are wonderful. They're not at the point yet where they're kind of ubiquitous. Every single advertiser that has ever worked with us I think in the last 10 years is for sure on Google. Everybody does that. And then the question is, what else can I invest money in with the expectation of profitability?

(00:46:53):
And so TikTok and Amazon, and even if I want to break out YouTube, they tend to be a little bit more specific and they definitely work, but not for everybody yet. And so the question becomes, will TikTok become as prominent as a social ads platform like Meta Ads? And we just don't know yet. What we see working very nicely on TikTok, by the way, are these founder led brands where the founder is the ambassador and they're doing those handheld videos with their iPhone. They're talking about the merits of the products that they make. They're talking about the problems that they solve. They are great at cultivating a following. Their content is highly engaging. Those types of organizations crush it. Also, companies that got really good at partnering with influencers. So we haven't talked about partner marketing that much, but affiliate marketing companies got into this early, and then influencer marketing networks are all over this. But essentially having these third parties be an ambassador for your brand in an authentic way, partnering with them, having them generate the content. And that's the nature of the partnership. Stuff like that seems...

Jonathan Becker (00:48:00):
... And that's the nature of the partnership. Stuff like that seems to work quite well.

Lenny (00:48:05):
Awesome, I love those bullet points of just what is likely to work on TikTok.

Jonathan Becker (00:48:05):
Yeah.

Lenny (00:48:09):
Just broadly, would you say TikTok advertising today is underrated, overrated, or just right?

Jonathan Becker (00:48:16):
If you are e-comm based, if you're wholesale, retail based, anything direct to consumer, you have to investigate TikTok and try and figure out whether there's a way of promoting your products there, basically. It's a massive, highly engaged audience with somewhat specific demographics. It's a very exciting opportunity. Other brands need to pay attention to this and figure out what their entry point is going to be. And just because the innovation hasn't happened for B2B SaaS yet on TikTok doesn't mean it won't. TikTok wants that to happen, and so the question becomes how. And so someone will unlock that, and they'll make a billion dollars because of it.

(00:48:59):
But I wouldn't recommend that channel to everybody yet, and at Thrive for instance, we don't have a huge amount of people coming to us yet saying, "Hey, I'm spending $750,000 a month on TikTok, can you help me?" Type thing. The other problem with TikTok is that it's extraordinarily creative, asset dependent, and heavy. And so you are launching net new assets several times per week, it is hard for creative teams and brands to produce that much content. It takes a machine, and that's expensive, and so that all factors into the cost of not only just the ad dollars, but the creative resourcing and investment that's required to power that. So it can be done in a more nimble manner if you're spending less, but you can't really get to a million dollars a month on TikTok without having a huge amount of creative being pushed to that channel.

Lenny (00:49:56):
Awesome, that's really good advice. I want to chat out about AI soon, just because I imagine that could help with some of these things. But you mentioned B2B SaaS, and I wanted to ask, I imagine Google is the primary channel for B2B SaaS, and if not, or if it is, where else do you find there's value in spending money to drive growth through B2B SaaS, which channels and ad networks?

Jonathan Becker (00:50:20):
So I'm agnostic here in terms of where people should be spending money. I see this world in terms of impressions and clicks, and does that conform to the marketing economics of a project that needs to be achieved in order for it to grow. It's less about where to place your money and how to think about the placement of those funds. And so the common mistake that we see as people build out a funnel, so to speak, for B2B lead gen projects, or B2C lead gen projects, is they'll be overly reliant on the first of a sequence of metrics that ultimately yields a sale.

(00:50:57):
So typically it's something like cost per lead, marketing qualified lead, sales accepted lead, there might be a couple more metrics there, and then eventually a sale occurs. And so that the thinking is that, okay, for sales to occur, I need more leads at the top of the funnel. The more leads I have, the more marketing qualified leads I'll get, the more sales accepted leads get, and ultimately the more opportunities there will be and the more revenue will generate.

(00:51:24):
And so if I think about this world of lead generation on performance marketing platforms as a function of cost per lead in that sense that I just described, then the tendency is always to want to drive down the cost per lead, thinking that that's the efficiency, I can get more cheaper leads and that will yield more revenue. When in fact it turns out that if you flip that conversation on its head and say, not all opportunities and sales are equal, and instead of focusing upfront on a cost per lead, I now want to focus on what is a high value customer, so the cost per lead is actually higher, but the ROI of targeting those people is also higher. And so to get the higher quality leads, it's not a function of CPL.

(00:52:15):
And so that is a very common pitfall that we see when people come to us. The antidote to this, it's interesting, so Thrive developed an ETL tool, an extract transfer load tool called Thrive Stack. It's not commercially available, but if you wanted a commercially available version of this, you can use something called Supermetrics. Basically Supermetrics is a data connector. There's a world of different data connectors out there, but it allows you to pipe, via an API, revenue data from your CRM into a third party database that can then be joined with data from the channel itself. You can build tables within a database and normalize them together in a manner where you can start to determine the relationship between, again, the audience that a particular cohort of opportunities came from and whether a sufficient degree of revenue is derived from those opportunities.

(00:53:08):
And then you can build upon that what is known as a lead scoring model, which allows you to bid in real time on the audiences that have a higher likelihood to convert to high revenue generating customers. And so the magic there, we talked a little bit about rates of return and instant gratification in the performance marketing world, lead generation inherently is a slow gratification process. And so the problem is that if my pipeline is full of opportunities that yield revenue in two months, six months, sometimes 12 months, how do I determine how heavily to bid on different leads today in order to predictively have an outcome where I'm maximizing revenue? And a lead scoring model basically solves for that.

Lenny (00:53:59):
Today's episode is brought to you by Miro, an online collaborative whiteboard that's designed specifically for teams like yours. I have a quick request, head on over to my Miro board at miro.com/lenny and let me know which guests you'd want me to have on this year. I've already gotten a bunch of great suggestions, which you'll see when you go there, so just keep it coming.

(00:54:20):
And while you're on the Miro board, I encourage you to play around with the tool. It's a great shared space to capture ideas, get feedback, and collaborate with your colleagues on anything that you're working on. For example, with Miro, you can plan out next quarter's entire product strategy. You can start by brainstorming using sticky notes, reactions, a voting tool, even an estimation app to scope out your team's sprints. Then your whole distributed team can come together around wire frames, dry ideas with a pen tool, and then put full mocks right into the Miro board. And with one of Miro's ready-made templates, you can go from discovery and research, to product roadmaps, to customer journey flows, to final mocks, all in Miro. Head on over to miro. com/lenny to leave your suggestions. That's M-I-R-O.com/lenny.

(00:55:06):
You talked about attribution, and I thought this would be a great time to get into that. So the way I think about this is oftentimes the biggest challenge in driving growth isn't the actual work you do to drive growth, it's measuring what impact your work is having and understanding where dollars are spent effectively. And things are changing heavily in attribution land with privacy shifts and ATT and iOS 14.5.1, or maybe whatever the version was that came out that changed everything. And so my question is, what changed, is one, in terms of being able to do attribution. Maybe even explain what attribution means for folks that are just like, what the hell are you even talking about? And then finally, just what do you recommend people do now to do attribution well in this new environment?

Jonathan Becker (00:55:57):
Yeah, it's a really, really great and pertinent question in the industry, and it has a lot of focus on it now, but attribution has always been a conversation that's been very important. So just to kick things off, attribution essentially is how we determine the relationship between what we did and what happened. In the context of performance marketing this means what ads did we serve, what campaigns did we launch, and generally what was the revenue ultimately that was derived from these campaigns? And so there's a funny quote that I love, someone named John Wanamaker in 1919 said, "I know-"

Lenny (00:56:33):
I bet I'm going to guess what it is.

Jonathan Becker (00:56:34):
Do you want to guess?

Lenny (00:56:35):
It's, "Half my marketing dollars are wasted, I just don't know which half."

Jonathan Becker (00:56:41):
Exactly, I love that you knew that. He said this in 1919. The world was a very different place there, but oddly enough we still have these types of problems today and it's because the world of what I did and what happened because of what I did is very complicated. There's so many variables, some of which we know, and then some of which we don't know that we don't know, type thing. And so attribution, as it stands in our industry, is still an incredibly subjective art and somewhat of a science. It doesn't mean that it's impossible, and it doesn't mean that there aren't varying degrees of sophistication as it relates to attribution, but we're still in a place where we have to understand the business and its goals, and then start to work on an attribution model.

(00:57:32):
An attribution model is probably something that is never solved, but I'll give you a couple of examples. So number one around attribution, it's important to determine whether an organization, a company, a client, is focused on profitability or growth. Whether I use one attribution lens or another will drive those outcomes. And so these days, for a number of different reasons, you mentioned iOS 14.5, in addition to third party cookie deprecation, and whatnot.

(00:58:02):
There's no one way necessarily of approaching attribution. The most classic and commonly held version of attribution used to be a cookie-based form of attribution called last click attribution, meaning that the last click in the sequence of clicks that yielded a conversion would be attributed with all the revenue from that conversion. Other models in a cookie-based world involved first touch attribution, so it's actually the first click. So I launch an ad, it's in an audience that doesn't know about my service or my brand, and so the first click that I get should be given all the credit for that sale. And then there's multi-touch attribution, which can take several different forms, but essentially is saying it's not one way or the other of the two first versus last click options, it's somewhere in between. So I'm going to create a weighted attribution model where the first touch gets so much of the credit and the last touch gets so much of the credit.

(00:58:59):
This is very subjective, as you can see. I am essentially looking at my business model, determining what my goals are, and then I'm backing out into an attribution methodology that I think adheres to both the economics of my business and the capabilities of these platforms. And again, we live in a world where a single brand might do television advertising, they might buy media in magazines, they might buy billboards, they might get impressions from Facebook, they might do paid search, so there's a big argument over how to ultimately model this.

(00:59:34):
So there was something called an IDFA that Apple allowed advertisers to use, and essentially what that was, it literally means ID for advertisers, and it allowed Facebook, Google, and other advertisers, Snap, TikTok, whoever was essentially selling ads that were predicated on being displayed on a Apple mobile device or a desktop device, it allowed the advertiser to provide attribution metrics and certain types of customer match metric metrics in their own platforms. And so when Apple ultimately launches its privacy changes, I believe in mid-2021, overnight it allows users to say, "Well, I don't want to share that information, I'm not going to share my IDFA anymore." Or, "Yes, I do want to share my IDFA." And so what you see as a result of that is less of the core data that Facebook in particular would've required to make attribution more airtight and ultimately validate its advertising.

(01:00:43):
So because we can no longer validate attribution on Facebook as seamlessly, we are in a situation where we're not sure any longer which audiences to target, and we're not sure how to run all of our creative testing, and ultimately we can't even determine the degree of revenue that's coming from particular campaigns that we've launched, and whatnot. And so it created quite a stir in the world of attribution, which is obviously this core discussion to is it working or not? People want to answer John Wanamaker's question, but the modern methodology now is through a number of different approaches to validation.

(01:01:25):
So the cookie based attribution, which had been probably the most popular version of attribution in the 2010s, really up to this point, is now one of the ways that we would look at this, but we would also include things like various forms of statistical modeling, customer surveys, population surveys, there's a number of different ways to the same place here. Statistical modeling, by the way, one thing that's very popular these days, which I think was originally created in the 1950s, is a form of statistical modeling called media mix modeling. It's essentially regression analysis, and it is trying to determine the causal relationship between, again, what you did and what the effect was on revenue. That's very topical in the industry these days. If you're looking for a tool off the shelf that can help with this, Recast is something that I see people using. But a lot of organizations build these highly customized bespoke models that ultimately feed the algorithm inside of an MMM model in a customized manner.

Lenny (01:02:27):
What is it, or is there a recommendation of just like, hey, if you're trying to do attribution today, here's what you should do. Is there a clear, do this, piece of advice, or is it super personalized depending on what you're trying to do?

Jonathan Becker (01:02:42):
The advice is that there's no single source of truth. Anyone who claims that they are a single source of truth, whether it's an individual, a model that they've created, or a tool that they've created, is not being accurate. I believe that the approach that works to attribution is that it's an ongoing investigation and it never stops, and essentially what you're doing is looking for evidence that validates the outcome of your performance marketing campaigns one way or the other.

(01:03:12):
There have been plenty of incredibly sophisticated attribution models that we've helped build and invalidated with MMM or other types of tools that have indicated that the campaigns do not work. So I will repeat that, we, a very sophisticated organization, have stood up campaigns that we measure and ultimately determine do not work, and that is an important finding in terms of an organization's determination of whether they want to continually invest, take budget out of these channels, or invest into other areas.

Lenny (01:03:49):
Awesome. You mentioned a company called Recast. I don't think you knew this, but I'm actually an investor, and so obviously a big fan, so check it out. I don't know what the site is, recast.com maybe. I think if you Google Recast attribution, we'll also have a link in the show notes.

Jonathan Becker (01:04:02):
Cool. Yeah, they're working with a couple of our clients right now and it's interesting, I like their approach.

Lenny (01:04:06):
Yeah, they're basically, they're building the model as a SaaS tool to do attribution in a really clever way. Amazing. Okay, just a couple more questions, one is around AI, one is around agencies, and just how to start down this road. How has AI in any way impacted the work you're doing today, or the work of paid growth, performance marketing, and then where do you think this will go in the next few years?

Jonathan Becker (01:04:29):
Interestingly, and I think I alluded to this earlier, but our industry has been influenced by AI for over a decade. So Google, Facebook, even Microsoft, these are some of the organizations that historically have been at the bleeding edge of artificial intelligence, and their goal was always to automate as much as possible within these platforms. And so the effect ultimately that we've seen from a human capital point of view is displacement. So we're actually, we have more people now than we've ever had, but the nature of the work that they do is more strategic, it's more about modeling, validation, asking the right questions, being focused around creative levers, like some of the things that we've talked about, and less so the trench work of implementation and bid modifiers at the keyword level on Google search, and some of the really hardcore manual analysis we had to do.

(01:05:19):
So I think if we are the canary in the coal mine for other industries, what will be interesting is I assume that if you work in an architecture office that you'll be doing less of the drafting, and it's already been a change that's occurred in those offices, but you'll be doing less of the drafting and more of this strategic problem solving around what kind of building is necessary for the client in this scenario, type thing.

(01:05:42):
And so on a practical level, though, some of these conversational AI models are very interesting for us in that when I think about creative testing, for instance, we can have ChatGPT come up with all kinds of variants of copy that we would not have necessarily thought of. It can do a lot of drafting of things like RFP responses, so we can feed ChatGPT 100 previous RFP responses that we've done and have it spit out net new responses to net new questions in a net new RFP, and it's like 80% good and still requires 10 hours of work to massage to the point where we can send it off to a client, but that replaces a week of work with five or six people that it would've previously taken.

Lenny (01:06:33):
And that's already happening today, you're saying?

Jonathan Becker (01:06:35):
That's happening now, literally right now, type thing, all of what I just said. And there's a lot of different ways that we are beginning to use AI to do more with less, basically. So that's how I see it influencing not just our industry, but other people's businesses as well.

Lenny (01:06:55):
That's amazing. So just to repeat what you said, your team now, because of ChatGPT is able to spend more time higher level, and leverage specifically ChatGPT to do this kind of, and RFPs are basically proposals, people are asking you to pitch to work with them, is that?

Jonathan Becker (01:06:55):
Yeah.

Lenny (01:07:12):
Okay, yeah.

Jonathan Becker (01:07:13):
Some of these AI-driven organizations that allow you to type in a prompt and it spits out an image, on our creative group we can come up with mockups in literally 1% of the time that it took. So you no longer have to draw the initial mockups for art. Suddenly what might have taken one person a week of work on a campaign takes an afternoon, type thing. You still have to understand what questions to ask of the AI and be capable of iterating, but these rough drafts that you might show the artwork of to a client to say, "Do we like this more or do we like this more?" That's AI generated. It's really interesting.

Lenny (01:07:55):
What's the tool your team uses for that, is it MidJourney or Dall-E or something else?

Jonathan Becker (01:08:00):
It's Dall-E and it's MidJourney, so you got it.

Lenny (01:08:03):
Got it. And so what you use it for is you're thinking about a concept and you just come up with a mock, coming up with a prompt to pitch what this could be.

Jonathan Becker (01:08:11):
Yeah, the interesting thing there is that these concepts often live inside of someone's head, and we do this thing as humans where we verbally discuss it, and I try and put the image in my mind's eye, in your mind's eye, which is an inherently inefficient process and actually hurts our ability to be creative leaders. Someone might just see it slightly differently than you do. Now that person theoretically can use Dall-E to output what is inside their head, and they can refine that easily through subsequent prompts in Dall-E or MidJourney, or whatever it may be, and then just say, "Do you like this? This is my idea." And so in a way it's helped humans connect more easily in that context than we were capable of doing before, and then in the context of pitching a client on ideas, we are capable of generating more ideas faster and iterating upon them live. So I can have a Zoom call with you, I can present my screen, and if you don't like the output of Dall-E that I just prompted, I can re-prompt based on your feedback in real time.

Lenny (01:09:16):
There's this sentiment that AIs going to take away the fun stuff of jobs, like maybe being creative and coming up with a new concept and photography, setting up a whole photo shoot, whatever.

Jonathan Becker (01:09:30):
Yeah.

Lenny (01:09:30):
Do you find that your team is excited about uploading this to Dall-E, or are they just like, goddammit, this is what I loved, and now I'm just sitting around in docs writing prompts.

Jonathan Becker (01:09:40):
I am not an incredible artist, but now I can take those thoughts in my mind and output them just as well as someone who's pretty skilled in artwork. So in a sense, I'm excited because this technology democratizes the ability to do this across everyone. Anyone who can think a thought can generate an image on an AI image generator. How many scribes do you know?

Lenny (01:10:04):
Zero, I believe zero.

Jonathan Becker (01:10:06):
Because we invented something called the printing press, and the printing press took over what was a cottage industry for very specific individuals that allowed them to control the flow of information. Yet today it's not like we have fewer writers or creative thinkers, they're just using different tools to arrive at the same place. You could say the same thing of the loom. There used to be people who would stitch your T-shirt together, or your sweater together, and none of those people have work anymore. There's still people who knit, I suppose. But your clothes are made in factories with machines, and the quality of the clothing is arguably a lot better than it was, maybe I'm wrong about that point, but you can make more of it and more sophisticated clothing.

(01:10:46):
There's no question that AI is going to displace and replace certain industries of people, certain roles. I'm not going to pretend like I know whether that's a good or bad thing, and I'm also not a poster child for AI, I am really just using the technology that's available to us to run our business the best way we can. We haven't let anybody go yet because of these efficiencies. The people who used to do the artwork are the ones using these tools, and I think they're excited that they're more productive. Is there going to be nuance around that? Of course. Are some people going to be sad about the changes? Yes. I think human nature is that people generally don't like change, but I think over time we've seen change has been a productive and positive thing for society, not a negative thing.

Lenny (01:11:35):
Awesome. Yeah, so it sounds like the team is excited, generally, and that's a good sign, and that's what I find generally. I think some people will be really upset, but most people will be like, amazing, and they never wanted to do this part.

Jonathan Becker (01:11:47):
This generation of people will be upset, and then there will be a subsequent generation that never knew anything different. So the way that I think about this is the people growing up these days have always had the internet, whereas you and I probably remember a time before the internet existed. AI-

Jonathan Becker (01:12:00):
You and I probably remember a time before the internet existed. AI is a platform like the internet and just as when the internet was launched, you couldn't conceive of amazon.com or all of these wonderful or social networks or whatever it was. The permutations of what AI will become, even in the current GPT 4 context or Dolly context or whatever it may be. We don't know what people are going to create with this and so it's difficult for the generation that knew what life was like beforehand, before iPhone, before internet, before AI, but there will be people who know nothing but this eventually, and to them it'll just be normal.

Lenny (01:12:42):
Until we're all just replaced by AI and then it's AI writing ads to get other AI tools to buy products from each other.

Jonathan Becker (01:12:42):
Yeah.

Lenny (01:12:49):
And then we're just sitting around watching Netflix and that'll be great. There's always this question when you're starting to invest in paid growth, performance marketing of should we work at an agency? Should we work? Should we hire someone junior to figure it out? Should we hire someone senior?

Jonathan Becker (01:13:03):
Yeah.

Lenny (01:13:03):
What's your general advice for, say, an early stage startup when it makes sense to work with an agency versus bringing someone in-house?

Jonathan Becker (01:13:12):
Mm-hmm. We need people in-house to do our jobs properly. Agencies are not mutually exclusive to personnel in-house. If we don't have a point of contact, for instance, at an earlier stage company and we're supposed to report to the CEO, CEOs tend to be very busy people and we often just can't get the information and approvals that we need to be successful. If you think that growth marketing is an opportunity and that performance marketing is a subset of growth marketing that you want some focus on, the reality is you're going to need expertise and you probably should start in-house and then hire an agency.

(01:13:49):
Again, there's different stages of companies and different examples of what makes sense at different stages and the nature of the work that we do at different stages changes. Very, very late stage, we're kind of doing a lot of this sophisticated production work. We're implementing sophisticated testing, we're working on attribution, we're building beautiful reporting, but in a sense, we're maintaining this machine that already exists. Earlier stage, we're doing a lot more of net new experimentation, trying to discover what is working and what sequence of audience targeting versus creative asset display will work, what channel mix works, and all of that. A lot of times the body of work that's required to nail that and the inputs to literally out front what questions to be asking isn't something that a small in-house team possesses and so they will work with an agency to scale faster, to get where they're going quicker, to bring in more resources quickly, basically. Then later stage, sometimes it's a permutation of the same thing or we're really just managing a whole bunch of different capabilities that they have because it's very complicated and difficult to staff.

Lenny (01:15:05):
That is really interesting. So it's not like agency versus in-house full-time person. It's you need both is what you're finding.

Jonathan Becker (01:15:11):
I don't think we have any projects that we're working on where there's not a professional marketing person as our POC and generally the people that we report to have a background in what we do, whether they ended up doing something different and now they're a VP or a director level and not directly managing channels themselves is one thing or another. But yeah, we work with people that have in-house expertise all day long.

Lenny (01:15:35):
I see. Sometimes it could just be a generalist marketing person that becomes your point of contact.

Jonathan Becker (01:15:35):
Yeah.

Lenny (01:15:39):
They don't have to be experienced in paper.

Jonathan Becker (01:15:41):
There's one more thing. As agencies, we're constantly solving people's problems in the market and as a result of that, I have a fairly large team of 130 people that think about performance marketing all day long. It's the first thing we think about when we have our morning coffee and it's the last thing we think about as we're leaving the office kind of thing. We have very sophisticated and built out capabilities as well as all kinds of software and processes and capabilities, I guess, that we've created that most in-house teams just don't have because they're smaller or newer or someone might be really sophisticated but they don't have the resources to implement it and stuff. In the same way that a really sophisticated organization might go to Boston Consulting Group or McKenzie to say, "Hey, we did this, but you do more of this all day long and we need advice on how to be more efficient," people come to Thrive saying, "Hey, we're top of our game. We have really strong people. They absolutely know what they're doing," which is always the case by the way, "But we're looking for extra sophistication and additional capabilities that are very difficult to create in an in-house organization." It's why we worked with Uber for 10 years. It was tough to replace us.

Lenny (01:16:59):
Yeah, that's quite the retention. 10 years, 10 years going, I imagine our net dollar retention has gone up too.

Jonathan Becker (01:17:08):
Yeah.

Lenny (01:17:09):
When you're looking to hire someone full-time to drive the paid growth channel and/or work with an agency, what do you recommend people look for specifically, especially things maybe they're not likely to think about when they're hiring someone that people often miss?

Jonathan Becker (01:17:26):
There's a couple of competencies that I think are very important. I found that, so I was a nerd as a child, if you can't tell, I don't know.

Lenny (01:17:35):
Nope.

Jonathan Becker (01:17:35):
Super nerdy, geeky kid. I would take apart-

Lenny (01:17:38):
[inaudible 01:17:38].

Jonathan Becker (01:17:39):
I would take apart my parents' computers, I'd try and rebuild them. I was fooling around with bulletin boards before websites existed, stuff like that. Then I became a web developer and as I transitioned into a performance marketing role, I realized that being technical and having technical aptitudes was extraordinarily helpful in terms of being on the front line of what we were doing and solving problems. If you're making your first hire, you probably want them to have the ingenuity and capability to solve a lot of the problems that you're going to run into within the context of tracking, attribution, data visualization, and then campaign management. Number one, do they have a technical background? That can be multidisciplinary.

(01:18:22):
At Thrive, we have someone who is a nuclear physicist that left discipline and became a performance marketer. We have people who have left mathematics and become performance marketers, people who left finance, full stack engineering, all kinds of stuff. Those are technical fields where people tend to be good at math and good at problem solving, and so that's a strong piece of evidence that they might have a strong aptitude for performance marketing and managing this.

(01:18:52):
Another thing is just literally do they have experience in this discipline? Have they managed on any level Meta ads, Google ads, TikTok, Amazon, some of the things we talked about? Do they have an appreciation for some of the inputs that we just discussed today that ultimately create the conditions for success within a performance marketing environment? Do they understand the role of creative? Do they understand the role of proper attribution and are they capable of understanding the mechanics of your business and so on and so forth?

(01:19:29):
I would say just fundamentally, those are some of the areas that when we are interviewing we're looking for. As an agency too, we are looking for the ability to work with people and be excellent at client services. One of the questions we might ask, I actually borrowed a Jeff Bezos question from early stage Amazon where I asked people without much warning in an interview, I'll ask them whether it's okay to do a bit of a thought experiment together and whether they are comfortable responding to a logic question. I'll ask the question of, "How many windows are there in New York City?" There's two things that I look for in this scenario. One is can they think on their feet? I'm not actually looking for the correct answer. It's a pretty difficult question to solve. But number one, do you realize that windows are not just building windows and windows are also in cars and trucks and stuff like that, so there's more than one type of window. Then can you just do the arithmetic of saying there's so many buildings per hectare or square mile or whatever it is and it's division multiplication type stuff.

(01:20:33):
But also from a client services point of view, I'm trying to see whether in a scenario where I've asked them something that they obviously probably weren't prepared for, that is a strange thing to ask and probably makes them feel a little bit uncomfortable, how they react in that situation. Do they remain composed or do they lose their composure and get frustrated or even angry that I asked them something that's pretty weird to ask? Their reaction there is often a good leading indicator of whether or not in a client scenario working with our clients, they can compose themselves in a difficult situation.

Lenny (01:21:10):
Very tactically, what level of experience do you recommend people look for in their first performance marketing hire year or two, four or five years, longer?

Jonathan Becker (01:21:23):
I've seen people who have 10 years of experience in the industry not be that great and then I've seen people who have one year of experience be absolutely phenomenal. I would say that again, the years of service under their belt is one clue or piece of evidence to their potential aptitude. But really, you want someone who can clearly demonstrate to you that they can competently run the channels I suppose. If what you're asking is literally at the level where someone is managing pay per click and programmatic media buying, then you want to make sure that they can just do that and I would say it's less about the years of experience. A lot of times people only want to do those types of jobs for a couple years before they're like, "Hey, what's your plan for me? I'm tired of running ads and building campaigns. I want to manage the people who do that now." The people who tend to really do that are often somewhat earlier on in their careers. That's just a reality that we see.

Lenny (01:22:25):
One more very technical question. What do you recommend the title for this role be? Growth marketer, something else?

Jonathan Becker (01:22:31):
I've seen so many different names for this. I think again, if the intention is to grow out a growth marketing capability, which again is not exclusive to performance marketing, then you're looking for a manager role of growth marketing manager. I've seen paid acquisition specialist for people who just manage channels. I've seen paid acquisition manager, performance marketing manager, media planner and buyer. I've seen all kinds of different roles. But when we are launching a new role these days, there's such wonderful tools like everything from Glassdoor to LinkedIn to PayScale where you can essentially just look at what other people are calling it and borrow from their role descriptions and then create your own role description and then determine what role is meaningful in the context of your own organization.

Lenny (01:23:23):
Awesome. Last question. We started with the story of how you landed Uber, and I know you also have an interesting story which I haven't heard of how you landed Snap/Snapchat as a client, so maybe tell that story.

Jonathan Becker (01:23:38):
It's a fun story and I think it's about ultimately confidence and knowing your craft. I believe it was like 2015 or 2016, we were essentially contacted to participate in an RFP down in Los Angeles. RFPs are difficult processes and essentially a department of a company is saying, "We want performance marketing and here are all the questions that we're going to ask you and exercises you need to complete in order to vet you as to whether you are the right partner or not." Sometimes RFPs are incredibly thoughtful and they're run by people who have either done a lot of RFPs in the context of what we do or they know specifically what the outcome is. Then in other cases, they're very experienced marketing departments with senior people and an approach that they may take is every stakeholder within that department gets to submit one or two questions to a long list of questions and the respondents have to respond to all of this.

(01:24:38):
The RFP process for Snap was somewhat complex and they asked a range of different questions and we find ourselves in LA at a hotel the night before, it's 2:00 in the morning, my business partner and another team member that we had brought down are sweating and we're like, "How are we going to build a presentation out of this? They're asking so many well-meaning but difficult questions that are zigging and zagging." In a pitch situation you want to tell a story, it's important, and that story has to have a beginning, middle, and end. It was just difficult and yet here we were and we had worked with all these really cool organizations and I was very confident that we knew precisely what we were doing. I kind of had this eureka moment where I was like, "Here's what we're going to do." It was Brent MacArthur, my business partner and I kind of arrived at this together, but we're kind of like, "Here's what we're going to do. We're going to have a completely unique take on how to respond to this and we're going to build a deck that's bespoke around X, Y, and Z."

(01:25:54):
We work till 6:00 in the morning, I think we get a couple hours of sleep and then by 10:00, we're sitting in their offices. Essentially the pitch is something like this. I stand there in front of about 20 different executives from their organization and I say, "You have thoughtfully curated an RFP for us to respond to today, and you have asked us to respond to about 20 to 30 different questions. With the utmost respect, I am not going to answer any of your questions today. I'm standing here because you are looking for a marketing partner and you need expertise in this area and instead of telling you the answers to the questions that you asked, I'm going to tell you what I think you need to do to get where you want to go and if you agree with me and my logic and what I talk to you about here today, then you should hire us. If you do not agree with my logic and you think that we're out to lunch, then you should not hire us."

(01:26:47):
You could hear a pin drop in the room. My own team I think was a little bit surprised. We had a whole bunch of technical difficulties that day. Someone who was supposed to Zoom in couldn't make it and so I have to ad lib part of the pitch that I wasn't prepared for, lots of zig and zags. They thank us for RFP response, lots of handshakes and smiling faces, but we don't know whether we're going to land the client or not.

(01:27:12):
Then about two hours later, I was driving to San Francisco with a friend, immediately leave LA for SF because I had a bunch of follow on meetings, and I get a phone call and they were like, "We loved your approach. Congratulations. You're hired."

(01:27:28):
These days I think Snap is still a really interesting organization, but at the time they were kind of bringing the future of a social platform or what a social platform could be in the same way that we think about TikTok right now. TikTok wasn't on the scene yet and Facebook was very worried about Snap and there was tons of engagement on Snap, so this was the it client of the day and it felt amazing.

(01:27:55):
Essentially the lesson was that you have to trust yourself and understand your strengths and weaknesses. In situations where someone is well-meaning but may not understand what they need, be brave and tell them what they need rather than just conforming to what they're asking for because sometimes inadvertently that leads you down the wrong path. We took a big risk and it had a wonderful payoff and we have, in addition to this story, we had a wonderful relationship with them for a couple years.

Lenny (01:28:28):
I love that story. I love how dramatic all your new customers, your way of acquiring customers and I imagine not all of them are this dramatic.

Jonathan Becker (01:28:38):
Not all of them are this dramatic, but there have been some pretty dramatic ones. I think part of running a business and running an agency has been the curation and experience around all these fun, crazy stories that have accumulated as we run this really interesting services business.

Lenny (01:28:55):
Awesome. Well, with that, we've reached our very exciting lightning round. I've got six questions for you. Are you ready for the lightning round?

Jonathan Becker (01:29:03):
I'm ready. Let's do it.

Lenny (01:29:05):
Okay, great Looking, very excited. Okay, first question, what are two or three books that you've recommended most to other people?

Jonathan Becker (01:29:13):
As a marketer, Storyworthy by Matthew Dicks is a wonderful book. He is a world world-class storyteller. I don't know if you know him. He wrote this book about how to structure and think about the art of storytelling and every marketer should read Storyworthy. It was a huge contributor to how I think about what we do at Thrive. As an entrepreneur, I love startup stories, so Shoe Dog, which is the startup story of Nike. It's basically Phil Knight talking about all the ups and downs, and it's crazy. Nike shouldn't be a company based on all the turmoil and challenges that they overcame. Fascinating read about taking risk and doing what you think is right and ultimately succeeding and building Nike. Then a book by Nick Bilton called American Kingpin, which is the story of the startup of the Silk Road, which was like this dark web marketplace for some pretty bad stuff. But it is half tech story, startup story and half gangster kind of kingpin story. It honestly should be a Hollywood movie. It reads like a Hollywood thriller. So awesome, great book.

Lenny (01:30:30):
Favorite recent movie or a TV show.

Jonathan Becker (01:30:32):
I don't know how recent it is, but I love The Big Short, just love it.

Lenny (01:30:35):
Not recent at all, but it's still great.

Jonathan Becker (01:30:37):
It's still great. The reason why I like it is because someone found truth in the data they were analyzing and then capitalized on it, which is really an analog for what we do in performance marketing. Then recently I have fallen in love with White Lotus for no particular reason. It's just raw entertainment.

Lenny (01:30:54):
What's something you've changed in the way that you all operate that's been relatively minor in terms of how hard it was to change, but that had a tremendous impact on your ability to execute as a team or a company?

Jonathan Becker (01:31:08):
The tools that we use in performance marketing change constantly. The channels themselves change and therefore what problems exist and what SaaS companies or different types of platforms offer as solutions to problems change. I don't think it's one thing necessarily, this might be a bit of a roundabout answer here, but I mentioned that change itself earlier is something that humans seem to not love, whereas in our industry, and I don't think we're alone here, but there's a culture of change. A welcoming of change, aptitude around change, having a playful mindset with new technology and tools rather than being upset about the fact that AI exists or third party cookies are being deprecated or iOS 14 removed IDFA and trying to be creative around those types of solutions has been like a constant for us. I'm kind of sidestepping the question a little bit, but it is the cultural willingness to adapt that I think has been a strong suit for us as a company.

Lenny (01:32:12):
Speaking of tools, final question. What is your favorite most underrated tool for performance marketing work?

Jonathan Becker (01:32:18):
It's Thrive Stack. It's a tool that we built in-house that allows us to pipe third party data and anonymized customer level data into a database and then ultimately visualize it in a format like Data Studio, Google Data Studio, which I think now is called Looker Studio. That has been a profoundly powerful platform from which we're capable of delivering insights that are built upon the data as opposed to just regurgitated data, if that makes sense. We have not released this technology yet. It's basically a tool we use in-house with our clients. But to a certain extent, I mentioned Supermetrics, which is another ETL and has some of the same capabilities but isn't quite as powerful.

Lenny (01:33:05):
Amazing. Jonathan, this is by far the deepest I've ever gone into the world of paid growth. I really appreciate you making time and sharing all of your wisdom with us. Two final questions. Where can folks find you online if they want to learn more, reach out, ask some questions maybe, and how can listeners be useful to you?

Jonathan Becker (01:33:23):
Thank you for having me again, by the way. This was a lot of fun. I'm available through Thrive, so thrivedigital.com. You can contact us through our site by accessing the contact form. Personally, I'm on LinkedIn under my name Jonathan Becker or Twitter, JZBecker. Then things that would be interesting from the audience, always looking for feedback in terms of whether people agree or disagree with thoughts and feelings that we have about what we do. Obviously always looking for amazing people ultimately to join our team at Thrive. If you're a practitioner of growth marketing or performance marketing and you're looking for a top place to spend some time, would love to talk to you. Also, I suppose if you need some help making this stuff work within the context of your own organization, we'd love to be at your service.

Lenny (01:34:12):
Awesome. What is the URL again?

Jonathan Becker (01:34:14):
Thrivedigital.com.

Lenny (01:34:16):
Sweet. You asked for feedback, be careful what you wish for. We get some hilarious C2 comments. We'll see. We'll see what comes in on this one.

Jonathan Becker (01:34:24):
Sounds good.

Lenny (01:34:25):
Jonathan, thank you so much for being here.

Jonathan Becker (01:34:27):
Pleasure. Thanks, Lenny. Thanks for having me.

Lenny (01:34:29):
Bye, everyone. Thank you so much for listening. If you found this valuable, you can subscribe to the show on Apple Podcasts, Spotify, or your favorite podcast app. Also, please consider giving us a rating or leaving a review as that really helps other listeners find the podcast. You can find all past episodes or learn more about the show at lennyspodcast.com. See you in the next episode.