20Growth: How to Use Influencers to Scale Growth Insanely Fast | How to Optimise User Onboaring for Growth | How the Best Growth Teams Create Organic Growth and Community | Why LTV/CAC Models are BS with Ketty Slonimsky

This is 20 Growth with me, Harry Stebbings. Now, in 20 Growth, we analyze how the best growth leaders build growth teams, run growth experiments, and execute at the highest levels when it comes to growth. Today, we have one of the best joining us in the form of Keti Slonimpski, Chief Growth Officer at Palta, the platform behind apps like Flow, the number one female health app with 77 million MAU, Simple and Zing AI, where she leads a centralized growth function across the portfolio. Before that, she was the first VP of product and growth at Helios X, a 900 million pound ARR, bootstrap D2C health tech business. And she's also advised companies like Sondermind, Runner, Guardio, Cheddar, and many more. But before we dive into the show today, SecureFrame empowers businesses to build trust with customers by simplifying information security and compliance through AI and automation. Thousands of fast-growing businesses, including NASDAQ, AngelList, Doodle, and Coda, trust SecureFrame to expedite their compliance journey for global security and privacy standards, such as SOC 2 and ISO 27001, CMMC, NIST standards, and more. Backed by top-tier investors and corporations like Google and Kleiner Perkins, the company is among Forbes' list of the top 100 startup employers for 2024, G2's best software awards for highest satisfaction products, and a recipient of the 2024 Cybersecurity Excellence Awards, something I love. I definitely never got in school myself. Learn more today at secureframe.com. And once SecureFrame locks down compliance, Pendo makes sure your product truly lands with customers. Whether it's the software you build or the software you buy, your tech stack should be creating results, not creating roadblocks. Well, Pendo's no-code software experience management platform makes your software better, with tools to see where users get stuck, guide them with in-app messaging, and constantly improving your UI. It's so easy that over 14,000 users and businesses use Pendo to increase revenue, lower costs, and reduce risks. Businesses love the control. Engineers love the freedom. Everyone wins. Start for free today at pendo.io/20product. - You have now arrived at your destination. - Ketty, I am so excited for this. I've heard so many great things. I'm also an incredible fan of some of the apps that you've been a part of building, so thank you so much for joining me. - Thanks for having me. It's a pleasure. - We were talking before about growth being kind of this weird, discipline. I just wanna start with like, when did you realize you were in growth? And what was that, ah, I'm actually a growth leader and I love it. - I started in marketing, switched to product-led growth before they even called it product-led growth. It was called a conversion optimization like 15 years ago. Then I did monetization. Then I did some core product roles. And at some point I started to go wide and broad. This is where one growth guru gave me an advice. You can do marketing, and product. You should fit to one of them. Like, and luckily I didn't take that advice. And I said, I'm gonna do just wide and broad roles. So I started to take user acquisition, product, analytics, product-led growth as like one big discipline. And I called it growth. And I don't care how other people call it, but I call it growth. You can grow through product. You can go through marketing. You need data, you know, to get all those insights to grow through both. This is what I call growth. - 'Cause before we were just chatting and you said a lot of people say they're in growth, but actually they're kind of in marketing. So for you, just so I understand, growth is an user acquisition, it's monetization. - I'll call that product-led growth, which consists of activation, onboarding, monetization, pricing, packaging, paywalls, all that stuff around. Serum, which is like a part of marketing, and product engagement could be easily a part of growth. In my previous job at HelloSex, I owned product, as well. - My question when I hear that in terms of the onboarding, the paywalls and everything that you mentioned there, is growth the game of incrementality, moving one to 2% in each of these different levers, or is it a game of we should try something totally crazy and totally move the ball out of the park? - I think it really depends on the stage. At the very beginning, you need to nail your core use case and build a growth machine around that core use case. Then you start the optimization phase, those small optimizations, they're gonna compound, and you need to make it really at speed. At some point, when you feel you nailed that core use case, you need to build defensibility, and defensibility you cannot build through those small moves. You really need to make a moat of this defensibility. Build a great product with clearly defined, broad need that can expand into multiple use cases. - Do you think anyone has defensibility on day one? A lot of VCs are like, "Oh, I'm worried that X could do this or Y could do that." Yeah, of course Google could do it on day one. Do you think anyone actually has defensibility, and is it built through process and speed? - I think it's built through process. Think about current companies in AI space, right? They built much of the defensibility through data loops. They sharpen that within communities, within their product. Users make prompts, they create things, they sharpen the mechanism of how this whole AI thing works. And if you nail that use case, you're gonna become defensible, 'cause other companies need to run after you and after that data. And if you make it at speed and at scale, you can win the game. - When we go back to this conversion and optimization, and you said data multiple times already, I think very much of science, being in the weeds and being a very technical mind. When we chatted there just now, and I was like, "No, no, we need to actually just like, go to the show because I don't wanna have this chat outside of the show." You said something really interesting about the profiles of people that can succeed in growth. Who do you think can succeed in growth, given what we just said? - I believe in people who are hungry, smart, proactive. They want to win, they're self-reflecting, 'cause you can miss lots of opportunities if you don't do that. Smart people can figure it out. I think growth is about being analytical and salesy at the same moment. - Why salesy? That's interesting. Think about onboardings and web onboardings that Palta nailed years before this space got overcrowded. It's basically a long funnel where I warm you up to purchase a product that you have never seen, and only after that you download the app and start playing with that, right? So what is it? It's a long salesy process up to 100 landing pages, right? You need to nail sales in order to warm up those cold leads to make them purchase. - What do you think most people get wrong when they think about that process? - Most of the people, they just copy what exists and think that works for them. That probably gonna work for us as well. They don't think about the value and the uniqueness of their specific audience and product. - It's so interesting. I hear so many times, and we chatted about this again before, like first principles and frameworks in quite a, mature, well-articulated industry, like consumer subs and growth both are in different respects. People think that you can take frameworks off the shelf and just apply it. To what extent is that true versus not true? - Good luck with applying frameworks to mature products. It won't work. The way Palta operates, for example, we're commonly gonna take something that works for one of our companies, apply that in like, just the company we're incubating currently, but then, we're gonna test the hell out of that to make it work. - Palta's this kind of weird beast in a respectful way. It's amazing. But like, what is Palta for those that don't know? And then we can dive into the machinery of how you build incredibly successful. - So Palta is a builder. We are a venture builder. It sits in the intersection of parent company and venture capital. We create the ideas, and then we partner with founders to make those ideas happen, and we give them everything from the very beginning to make it scale really fast. So we give them growth, analytics, data infrastructure, finance, legal support. We build such category leaders as Flow, Simple, Lavi, and Zing. Palta is going to hit 550 million this year in revenues. We grow each year 50% and even more. We raised around 200 million from top tier investors. One of them is VNV Global, and Palta is Avocado, in Spanish. So we are building health tech B2C mobile subscription. - Got you. Okay, fantastic. And so when we think about that, I'd love to know, you go through lots of different ideas when you're creating new ones. How do you know, and you said about testing, really incessantly, how do you know when to really go all in versus discard and let's move on to another idea? - So the way Palta operates, we look at red ocean categories where there are already players that are doing more than 100 million in RR, and then figure out how to outplay them and to outbuild them. So we're going to create great products and scale really fast, put lots of capital so they can become the leaders in the space. Each company that cannot hit less than 100 million, we're not interested. So at the peak, we had nine apps, and at some point we killed some of them. They either didn't nail the product market fit, or they were too narrow and couldn't hit 100 million. So now we stayed with four, leaders. Can we go with one that didn't hit product market fit? And what did you learn? Let's go into a different example. Let's do the example that hit product market hit, but was super narrow. Let's do it. So one of them was called Weatherwell. It's a product that could basically foresee how you're going to feel based on the weather around you. It can be mood swings, headaches, everything. So we had terrific long-term retention, but we nailed just a very narrow segment of those who suffer from really tough migraines and cannot get out of the house for three days because of the weather. We couldn't scale it ever to 100 million. It had the product market fit. We were frustrated of killing something with rather high retention, but we needed to do that. Okay. So many things to unpack there. Number one, respectfully, why do that business? That doesn't feel like there's many categories. You said Red Ocean, people making 100 million. I don't know anyone making 100 million telling you about kind of emotions tied to weather. We said weather is huge. Health is huge. Let's blend. Why the 100 million? That feels like a bit of an arbitrary number. If you can get it to 10 million and it's super efficient and profitable, why is that bad? Because each business is going to be compared to flow. And when you have flow in the portfolio, which is a. Yeah. More than 75 million now, 6 million active subscribers, second year retention of 80% and more than 65% organic traffic, the numbers that companies can just dream about. You need to match those numbers with the new kids that you're raising, right? We have Simple. Simple is a weight loss management tool. They're going to hit 200 million in ARR. We have Zing, which is a powered fitness coach. With hyper-personalized and super engaging workouts. They're going to hit around 50 million in ARR. They grew four times since last year. And we have Lavi. Lavi is an AI cosmetologist making face care accessible. They're going to hit around 20 million in ARR this year. They're growing 30% month over month. These are actually really interesting products and really interesting stories. Zing, for example, again, sorry for like going off schedule. I'm just too interested. It's such a crowded market. And I'm always pushed all of these different products. And, you know, I'm pushed ladder to the extreme and 50 others. How do you succeed in a market where there's seemingly complete commoditization and no differentiation? So it comes to the way how we basically launch products and how we run the whole play. The way we operate, first, we're going to UXR the area and see like if there is something there worth our attention. Then we're going to do some more tests. We're going to send SCAC willingness to pay, how the unit economics might look like at scale. If it's there, we green light. We'll bring a tiny team of two founders, commonly a tech founder and business or product founder. We're going to put a lean crew behind them. We're going to launch an MVP and build the core use case. This is where a part of growth comes in. We're going to literally scale very fast the first use case. How we do that? We start with paid user acquisition. Commonly one channel. It's going to be meta. We scale to three, five K daily. This is where the algorithms start to optimize. And this is where you see true unit economics. Once it's there and we nailed it, then we're going to expand to the next use cases. We commonly start with web onboardings because you avoid 30% fee. There is attribution there, which makes user acquisition more efficient. But there are things that people commonly don't talk about. The nature of credit card payments drive higher retention and on web, you can charge more. This is how consumers perceive pricing on web versus mobile. Now there is a catch around funnels. If you overinvest into the funnels and start obsessing about that instead of all obsessing about product, it gets you to the funnel business. And that's fine, but most VCs won't back that business. It's probably not a business that's going to go to IPO. And build a multi-billion valuation, right? Though if you're really great at that, you can run killer UA machines and funnel set scale and get to tens, even hundreds of million in RR. How do you think, if we go back to the start of that where you said, hey, we like to go to one channel, we like to do paid and we like to see what works there and like just test out that unit econ in that first testing phase. I'm always taught that actually, if you need to grow your product early through paid, it's not going to be that easy. It's not going to be that easy. It's going to be bad. It's not good. It's not a good product. It should be word of mouth. It should be viral. To what extent is that complete bullshit? I think times have changed. So when Flow was born, they ran on purely organic. We do just product. We don't do monetization. Monetization UA came years after. Within the current products in the red ocean market, I don't think you can live just with organic. And organic is very hard to get. So you need to start with paid and bring users in, see how the whole unit can look like. And if you may even scale that use case, and if you have PMF and then see how you grow that business further. What is good unit economics? I know that's a really hard question to ask, but like, you know, is a two to one LTV to CAC ratio good? Does it need to be five to one? What impresses you versus doesn't? Again, it comes to like, do we need to match flow numbers? Oh, we're talking about standalone businesses. Yeah. If you imagine me being like an early stage founder, thinking through this, needing your help. That's how I think about it. I think when you just run paid, you need to think about payback around three months, and you can run with 140, 180% for us. This is probably what I would be looking for. If you just do paid at some point, and very early, you need to get to organic. And this is what we start to discuss. You better build that and anchor into shareable loops. Well, for example, why flow partner mode exists because women want to share the insights with their partners, right? Why running app might be super successful as they basically built around the habit of running marathons in communities. If you thrive into tribes and communities, you amplify organic far more than with single use apps. Think about natural retention. So flow has that natural retention, where users come back month after month, since being teenagers to like, peri and post menopause. And this is huge. This is how you become billion business. It's also unique, though. And again, I'm kind of going off script, but in the way of like, the use case is enduring, for obvious reasons. Dating, on the other hand, sucks. Absolutely. Do you agree, then, that there are just inherently shit categories, like dating, where you go in, out, you get married, now, I mean, you're gone? You want to outsmart, think about categories and product first, and then think where and how you build your products, instead of just, you know, fantasizing of building a business where there is no natural retention. What are other categories that have this enduring retentive feature that people don't think about? I think, currently, it's coming to workflows. If you're embedded into the workflow, and each piece of your workflow, think about your GPT, think about other companies in the space. This is what builds your defensibility, and that really tremendous retention. You are becoming very hard to replace, think about work or something, right? And if you see it, like, in your email and search, and all the different tasks that you are performing during the day, and you use just that one tool, which is hugely embedded, knows everything about you, and sits deep inside your data, this is what's going to bring your retention and defensibility. How do you think about the painkiller versus vitamin, when we think about those? You know, you said there about kind of weight loss management app with, what's the name? Simple? Simple. Simple. People don't like losing weight. It, generally, is like diet, exercise, change of habit. It's much easier to sell chocolate than it is carrots, in many respects. It's much easier to sell GLP-1s. You know, I'm board advisor at HelloSex, which is GLP-1 company. And when you compare GLP-1s to habit management, it's unbeatable. People are lazy. People want to sit on their couches, to eat junk, and to become lean. It's a sense of trying people to work out and to eat healthy. You need to have this inner motivation that most of the people don't have. So we were talking about the paid in the early stages there, and actually the payback and the ROAS, fantastically helpful. I'm a founder listening to this, and I'm going, "Okay, but I'm in a Red Ocean category, and I'm competing against a Simple or a Flow or one of these apps." You don't have a chance. Yeah. We built on fair advantage. We built growth, 100 million hour plus machines from day zero. What the edge of Palta is cross-pollination synergy between the apps. For example, simple playbook will double Floss conversion. The shared acquisition strategies that we use help us to scale new ventures with the same ROI. We get fantastic terms with all ad networks. Braze, PayPal, it's something that small size and mid-size players cannot even dream about. So you have cross-referral networks between the apps. So basically you get cheaper customer acquisition because a Flow member will become a Simple member and a Simple member. No? It doesn't work like this. The way it works, so basically we can take tactics and playbooks from one app to another. Think that Flow is running thousands of experiments per year. We're going to learn out of that 1,000 experiments and apply them to other apps. Flow is rather big and they don't take big risks. So we can do something more risky. In a small company, take those findings and roll out in the apps in the group. Is there an example that comes to mind of one way it was a risky experiment and it worked and you took it to another one? We introduced upsells, second subscriptions and third subscriptions and we rolled all of that in the whole group. By the way, second subscription and third subscription, I don't mean here tiering. In the world where CPMs are very high and user acquisition is super expensive, you don't make a living from one subscription. You introduce additional revenue streams embedded into the product like extra value that you sell a second subscription and third subscription. We do lots of upselling as well and we use all those strategies in all the apps in our portfolio. So just so I understand, I'm a venture capitalist, so we're inherently naive and ignorant, which is why we do what we do. A first subscription is like, okay, I'm a premium Flow member, okay? And then a second subscription is adding my partner to it, adding insights. What is that? So for example, you're subscribed to get personalized workouts, right? But we also sell a body scanner and you can get a subscription for a body scanner. In addition to various upsells of workouts, of recipes and other stuff that you're going to purchase from us at the very first minute, you open the wallet. Got you, got you. So basically your first subscription does the payback and makes me neutral in terms of spend. And then your second and third is basically profit. Absolutely. Got you. How much of revenue do you want to be from second and third? Not sure I can answer that, but let's put it in a different way. All the upsells are going to add around 20% to the LTV. And you don't worry about like, I feel a little bit pissed off when this happens. Sorry, Katie, but it's like, you know, I've paid to come. It's like private members clubs. I've paid to come in. It's some ridiculous amount. And now you're charging me a ridiculous amount just to eat there. The restaurant's free to go in. Do you know what I mean? But I'm doing that after a long web onboarding where I warmed you up and you're ready to open the wallet and you're already there. And just a small another feature that I want you to purchase in order to be successful. Do you think people are doing the second and third subscriptions well today? No. Why? Palta has nailed all those tactics around web onboardings, payment infrastructure, and the whole sales piece of paywalls, intrapricing. We almost don't do free trial. We do intrapricing, which is different. Upselling bundles and second and third tier subscription. Not many players in the industry are doing that. Some players that are, Noom, they nailed that as well. But for me, and Welltech. Sorry, you said you don't do free trials. Why not? With free trials, you bring free trialists, which is a very weak signal. You want to bring payers, people who have high willingness to pay. And this is how you want to optimize your algorithms. But I'm really put off by the like, enter app, paywall. And I'm like, well. It's not an app. We do web onboardings. How much of net new subscribers is web? On boardings. So 80% will be coming through web in all our companies apart from floor where it splits 50, 50. That's amazing though. Well done because the cut is just brutal. I mean, the Apple cut is insane. Absolutely. We invested in a company that does like rooting away from Apple just as I said, it will be a great company. There are some pros and cons to do web onboarding. Think about web onboarding. And if you don't have a free tier, you might never grow to become a floor. So it's like. A game with pros and cons. It's a game of you can bring payers with high willingness to pay and nail this whole unit economics and user acquisition, but you won't get now. You won't have that word of the mouth. So better to blend both tactics in both worlds. By the way, Floyd's still doing the free flow is different product. I'm so interested. You don't do free trials because it brings like lower quality. We don't have the willingness to pay when we see the paywall. There's many different options for pay. Do you have like a strong opinion around it's good to give people four options, just give people one option highlight one option and put most recommended on it any lessons around paywall optimization or preference the best lesson going to be put a PM nails monetization from day one and test the hell out of that you need to test probably to run like 1000 experiments on the paywall to get what's right for you. So you use paywall. Paid to get a shitload of traffic in, see what the traffic does. We apply best practices from palta portfolio, and then we test the hell out. They pay, they come in. And the trouble with most AI apps today is if they do get there, they just churn within like seven days. Is there anything that you've learned in terms of onboarding and retention in those first user actions that people don't think enough about? I think it comes to the nature of the. Product you need really to solve the need to do that in a fascinating way, where you nail the first use case, you build a whole onboarding and activation around the use case and not something fluffy and broad and only after that you expand to more use cases, the second one going to be the whole activation, what most of the teams many times forget, I just sell premium and then what I care about is how to improve cancellations. And churn, but what happens in between you need to activate your users on the subscription users don't remember what they really managed to buy. They don't remember your features. No, we run a super interesting survey at flow where we asked our paying customers if they use feature X or they use feature Y, if they engage with that feature, if they enjoy that feature. And you know, the answers were around, I don't really distinguish between the doctor's report and this and that insight. Oh, I couldn't find it. Oh, I didn't know it really exists. So if you don't do a good job around taking the user and really explaining them why you should be enjoying your premium and what you paid for, then don't be surprised why the users churn. And now let's do like something truly amazing to get them back when they just are pissed off of your product. I think it's one of those ones where it's a bit like email newsletters where people often don't send them very much because they're afraid. They're afraid that when they do, they're going to churn. And it's like people don't like to actually show you the depths of the product because they're afraid the more that they hassle you and go, hey, look how amazing it is. The more they remind you of the subscription that you have. And that's the worst thing ever. I find when it's like, oh, I don't want to remind them of the subscription. It's like, ah, that's not a good thing. You want to remind them that they have a problem and we are the best solution for that problem. This is what you want to do, really. I totally agree with that. When we think about going back to this kind of, hey, we test and we iterate and we use performance marketing to get a lot of people in. And then you said it's about that ability to transition to shareable loops, organic shareable loops. How do you know when it's the right time to go from paid, let's smash Meta with three to five, well, I'm not smashing Meta, to be honest, with three to five grand a day. But you're spending on Meta with three to five grand a day to, hang on a minute, we need to shareable loops and it needs to be more organic. It needs to run parallel. Once you nail your first use case, you'll have that like economic. You start to scale. You need to start working on your product and those shareable mechanics within the product. But you cannot get any mechanics if the product is weak and if it's not built on the natural loop. Otherwise, it's just the road to nowhere. What do you think about growth loops? Everyone talks about a growth loop. And I'm like, is this just another Silicon Valley-ism of frameworks? And what does that actually mean to you? And if I'm a founder. Trying to build a growth loop. What does that actually mean? You know, I'm a big fan of that theory. When I work with founders, and I work with founders for like more than 20 years and with different founders, I find it very hard to explain. And I find myself very much talking about funnels and creatives tailored to the use case. And this is what easy and simple basically to deliver the message that I managed to deliver. And at the end, it doesn't matter how you call that. It's like building that UA engine that can scale. build a UA engine that can scale most for the last few years or 20 years has been SEO. We're seeing that really move and change. Given the dominance of web for you, how do you think about the transition away from SEO and how that changes your strategy? Part of it onto SEO. First of all, the best organic you can get is word of the mouth. And this is where you should aim. It's like building the great product that solves the problem and brings other users in. This is what Plur does. With the rest, I believe the current opportunities lie within communities. Building those communities where the users are, Discord, Reddit, YouTube following. I'm not a fan of talking about SEO, AO, if one's going to replace another. Probably yes, probably no. Not all the companies are built for SEO and for good SEO. I'm not a fan of that. You need to be in a specific niche probably to build some programmatic SEO. For B2B SEO or AO can work. In our niche for health tech, this is not a strategy to go. Word of mouth does wonders for your unit econ. When we go back to LTV and trying to understand that unit econ, the hard thing is we're an early stage company, as you said, we're testing one channel three to 5K. I don't have LTV. How do I figure out unit econ when I'm really early and when you're testing new ideas? You don't have 12 month cohorts. What do you do then? We use the cohorts we have, apply Paltos LTV models and try to figure it out. But we keep that space, meaning I know that if currently I'm looking at, let's say, Rolls 140, tomorrow when the real numbers come up, it can become 110 or even 90%. You need to think about how you make it really smart and don't like try just to be at the surface of 100 because tomorrow it can be much lower. Can you take me to a time when you use Palto models, but way overestimated where it would be? And what did you not see? So we did it with one of our companies. We thought that we are someone in the area of 114. The real data start to come. We realized very quickly we're around 110. We start to bring different channels in and different channels can change a lot your LTV. For example, TikTok that brings like very different type of users that we realized in a few weeks can go under 100. And we start to rethink how we make this whole thing work. You said TikTok brings in very different users. We find this too, just for our core shows, given they bring in a lower quality, much more high churning user. And actually often have zero impact at all. We have videos with 25 million plays and they bring in like 12 subs. I mean, like, and we have videos with 2000 plays that bring in 50. I mean, is it not worth then doing this kind of junk food, short form of low quality? I think it's worth Palto commonly works with Meta, Google and TikTok. Those are three channels we try to nail within each company. Sometimes it takes us a while to nail that. For example, with Floyd took us a year to unlock TikTok, which we scaled tremendously this year. What did you do to scale it? And why did it take long? TikTok is a game of creatives. With TikTok, you need to launch around 1000 of creative per week to make it successful. Well, of course, 1000 creatives a week. What do you mean by that? We are playing a very different game in order to scale the way Palto scales and Palto spends hundreds of millions. As a group on user acquisition. So as a centralized growth function, we build a creative factory where we launch around hundreds of creatives per week. And with AI, it becomes 1000. Can you talk, what does actually hundreds of creatives a week mean? Does that mean hundreds of short form videos? We're probably going to split it into 70-30, where 70 going to be like the permutations of the old concepts that work. 30 are new form creatives. Fascinating. It can be videos. It can be static. Don't underestimate static. For example, flow is 50%, I think, static. What do you mean? Static creatives, like static ads, not videos. Wow. And it's much easier to produce those. Much cheaper as well. And you can produce that with AI. That's so smart that you do the permutations on what worked before. Of course. Yeah, I've done this on LinkedIn a lot. We basically have like my top 20 LinkedIn posts. And then you can just do, I mean, obviously put it in ChatGPT and say, hey, give me three variations, and then schedule 60 days worth. And it's basically guaranteed viral-ish bangers. So basically, the creative and the ad is the hook, the body and the CTA, right? And you can play with the hooks. So you can just take the body and think about 30 hooks for the same body, and then like 30 permutations of the CTA. And this is how you get to that crazy number. So it's. Then you get like all the. All the mentions possible. And this is how you win. So on TikTok, you need to have three times more creatives than on Meta. Wow. And you need to work really with like TTC or their creators, marketplace, the creatives in-house, blend them, and this is how you win. How do you use AI to take 100 creatives and use to get 1,000? So basically, we work with like, you know, MeJourney, VR, and it changes a lot. It does. Of course, like, you know, on a weekly basis, you can replace the tools. Basically, the idea is to have a very strong team that amplifies those tools. Someone who is like on it, someone who is saying, cool, so now let's test this. Let's try this. Let's see if that brings us into a different speed. Those are modern growth teams where you have like people who are AI native, and this is their job. You know who I'm dreaming to bring into growth teams? Prompts. Like those people who are going to just prompt and build growth machines, like literal growth machines as, you know, probably in two years' time, it's not going to be relevant. So you build a growth machine where it can learn the trends on TikTok of the last five days and your performance on TikTok and recommend you like what you should apply tomorrow in the creative and watch your campaigns performance. Blend it together with creative performance and tell you what to do. I'm worried that in a world of AI and creative, the value of content goes down because the supply is infinite. And so when you go from 100 to 1,000 and you use Veo and Midjourney and 5 million other tools that there are, it goes down in value. Do you share that concern? No, because I think it's about diversity, velocity and volume. At the end, the creatives will find the way to the right. And if the audience is not right, then your creative sucks. Do you use influencers? We do. We never managed to scale the channel. It can go to, let's say, up to 1 million spend. And for us, it's like nothing. Do you think people waste a lot of money on influencers today? Absolutely. But I do believe when talking about organic, there is an opportunity in the market to hack it, to bring like, you know, again, TikTok AI native person. We're going to like launch 30,000 videos per day and see if he can get to great influences or build brand out of that. Probably this window won't be open for a long time, but it can be super interesting. How has your role changed as a growth leader in the last 18 months? You've been doing this for 20 years. Yeah, I think we should talk a bit about Palta and like how I got the role. Sure. Four years ago, Palta decided to build centralized growth. So growth function and they decided to centralize even more functions such as data analytics, finance and legal. And I was brought in to build the growth. Palta is a decentralized ecosystem with independent companies where they have their own leadership, their own teams, their own growth teams and roadmaps. And there is no top down control. Yuri, the CEO, said, hey, this is Kathy. She knows growth. She's going to support you. No mandate. No citizenship. It's at the table. And I love those challenges. I love ambiguity. I love chaos. I love building from nothing. Big and porous. So I understood that I need to treat myself as a product and to deliver value fast. Is it hard that when you're not involved in the company day to day because the team's like, well, you don't know what I'm dealing with. You don't know what I'm going through. You don't have the data that I have. Absolutely. I took a plane. I spent weeks with founders and their teams to understand. What are they doing to dig into their data, into the experiments, into their ways of working. And this is where trust start to click. They basically invited me to do like bigger bets with them together. What I try to do is not to tell them what growth is and how they should be operating, but basically solve their immediate challenges and help them to scale, help them to win. Do you think growth. Operates best as a separate function or as an integrative part of a product team, a marketing team, a sales team? you name it? It really depends. The way Palta operates, at the very beginning, we incubate those companies and we launch the products. We hire one UA manager and one PM, PM that nails the funnels. Basically when the company hit around three, five million RR, before that Palta runs the show. So up to that point, just so I understand, Palta runs the show, we're doing performance marketing, we're getting to see if we have PMF, and then we're hitting three to five million. Absolutely. And then we hire a UA and PM. For every 10 ideas that you have, how many get to that stage? Do we have a lot of fails? Nope. We also don't tend to launch too many ideas in one quarter. So it can be like one idea or even one idea in six months. Okay. Got you. We never wanted to become a factory of apps. So we never wanted to have more than, let's say, 10, 12 kids. Otherwise, it becomes unmanageable. And you want that cross-pollination and synergy. This is the edge of Palta. So we get to this UA. We get to the stage. And this is like your first lean growth team, which is commonly sitting under the founder. At some point, like when they scale to, let's say, five, eight million, I want to hire the leader for that team. I don't hire for titles. I don't hire fluff. I don't hire managers that just can manage. So I'm going to bring someone who can lead the team and be hands on. At the same moment. At that point, they still can operate like a separate team sitting under the founder. They're rather lean. Once they scale, what the rule of thumb is one UA manager per one channel, matching the number of growth PMs, and it becomes like a department. At that point, Palta becomes fractional. It means that I'm still deep in their numbers, in their experiments. I know everything that is going on there. I approve their board plans. I run with them weekly meetings. But they become like a standalone and independent beast. Where do they sit? I don't think it's about the perfect place. As at the end, it's about people, company's DNA, product and marketing leaders' skill set. And you cannot bucket people into frameworks. It's all about efficiency and the processes. At Flow, product growth is a part of product. Where marketing is a separate function that consists of marketing, growth marketing, brand marketing, product marketing. At Simple, growth function is a standalone function that works great with product and marketing over the years. And Zing and Lavi, since they're like the youngest kids, they're still a part of growth marketing, including funnels, commonly sitting under the founder. When we think about building out that growth team, there's a lot of founders who will be listening to this. And I'm thinking of several of our portfolio companies actually will be listening to this. And they're going, okay, I need to probably, I've hit that five to eight million stage. I need to start thinking about this as an independent function, or it's like an independent role, so to speak, within the team. Should I hire a head of growth, a stellar Silicon Valley badged person and spend a lot of money on them? Or should I try and hire someone young, someone fresh, but maybe who doesn't have that experience? How do you think about that? I probably would not hire, that Silicon Valley. Why? I think what you need to bring to the table is hunger, is lots of dignity, lots of dedication, lots of experimentational mindset. And you cannot get there when you basically worked for best-in-class companies, seeing how growth runs there. And you come into a small company, what are you going to do there? You don't have the motivation to work in a lean team. You don't have the motivation to do the dirty job, basically, and to run those basic experiments to build the infrastructure. I believe it should be someone young, fresh, with the skill set of being a builder and doing that over and over again. Because in one quarter, you build and optimize for X. In another quarter, you need to rewrite the whole thing and rebuild what you have built because it doesn't stick anymore. This is the mindset. You need to be ready and open for the changes all the time. Why do growth people most often not work? I think it depends who you hire. At Palta, they work. How many people are in the centralized growth function within Palta? So basically, the way I built the centralized function was around setting common standards, cross-pollination, making the process efficient. And I looked at three pillars of growth. User acquisition, product-led growth, CRM and engagement. I put a leader. So what's the final one? CRM? CRM and engagement. Ah, CRM and engagement. And then I put a leader for each pillar. This is how we run it. We integrate with the companies, and we can act as fractional CGO, fractional VP UA, execute the project inside the company, or even outsource that. And with small companies, we are basically incubating them. We are currently five people. But it goes to a direct level, and we don't hire PMs, copywriters, and basically the execution functions inside the centralized function. I would never do a better copy than Flo can do, of course. Can I ask you, I'm always struck by, does CAC get better or worse over time? CAC can get worse over time if you scale, not efficiently, but if you build a growth machine with creatives and funnels, tailored to your use cases, you can keep CAC stable. What has been the biggest growth decision you made that turned out not to work? One of the biggest decisions was to start centralizing the UA function, where we analyzed that each company is very unique, that creatives are unique, and we cannot build it as like a holistic function that's just going to provide like a chunk of creatives and assets and campaigns to the independent companies. So we pivoted very quickly and decided that all growth should stay within the companies, and we're just going to support the scale. And so you're like, hey, there's commoditized functions like finance, like legal, like. Absolutely. We even were thinking about doing that with CRM function and realized it won't work either. We can do economies of scale for data, vendors, resources. Resources, talent, but we cannot really replace the growth functions in the companies. One thing we haven't talked about, which is probably criminal of me, but it's actually retention. You know, we've mentioned Alex Schultz before the show. He says to me all the time, retention, retention, retention is the only thing that matters, the only thing that matters. And, you know, I think Duolingo has been so successful because they have such good retention numbers. I think they're D360, it's like 35%, which given the category they're in, well done. How do you know what is the right retention metric to go for? Is it D7, D30, D360? At the beginning, you optimize for D7 and you look at D1 and D3 because you can't really optimize for D30 as you don't understand like what is happening in your activation process. So let's say the first six months up to 12 months, you're going to be optimizing, you're annealing your activation. And onboarding and only after that, you're going to be watching like your further metrics. Having said that, you should have in mind, you need to build that defensibility and you need to see that flattened retention curve, which we are seeing with the products that have a very strong PMF in our group as well. Yeah, a hundred percent. You see that with Flow, I'm sure. With Flow, we've seen that with other products as well. How much do you care about like D30 retention, say, versus usage? Do you get what I'm saying about that? So it's like for a Flow, for example, people may not need to check it every day. Absolutely. And so it's like not an engagement game, but retention is fundamental. Absolutely. I think, again, it comes to the nature of the product and what are you solving for and what is the solution. Does that differ across your portfolio? Absolutely. You aim probably for a couple of workouts throughout the week, right? But from like a woman who is tracking her period, the natural use case, it's going to be monthly. Does that suggest success, like checking monthly? Is that a close enough gauge of whether you've got a successful user? No, I think this is what can tell you that you nailed your natural use case and now you need to start building on that. Meaning you need to build content and content loops. How basically I bring women to check in Flow, like the content, the guidance, to see if today, for example, it's safe to have sex. If today I have a desire to have sex with my partner. This is what many Flow users are coming to. They're not just coming to track their period. They want to get the insights about their health and the whole life around their period. Push notifications are a kind of drug for engagement where it's like, hey, we can just push notify and boom. And we saw that in the first 10 days. 10 years of probably iOS and app engagement. At scale, by the way, it can be very expensive with Braze and other tools. Talk to me about that. Is push notifications a tool? overused overhyped way to engage an audience absolutely when we just start and scale and build the products we commonly have like basic email flows and not pushers within the time like let's say six months after we start to build basic journeys with push notifications around the user journey when the company really scales and like they nail the unit economics and the use case we start being harder on crm push and email together the more is not the best way to go here but if you don't have enough it's probably not gonna work as well you need to find that balance and it's just about testing do you find there's a wide variance in conversion depending on copy absolutely copy timing you and it has nothing to do with creative and design so you can just screw it and anything on timing that's interesting absolutely like you need to find the hook the real the best timing it's not about the hour of the day it's about the best timing for the user to get that notification you should it should be relevant it should be tied to some kind of event this is how it's going to work and not just send a 10 a.m push that today is going to be a great day for you why did you not decide to build your own braze or push notification technology within palta so flow does that and within the rest we use brace which becomes very expensive at scale yeah to build our own solution it's also super expensive for example we did it for payments and it took a year and a big team to make it happen though we have like one of the best solutions in the industry listen i'd love to do a quick fire round so i say a short statement you give me your immediate thoughts does that sound okay yeah sure so what is the most overhyped channel today well everyone's super excited and you're like meta there's many channels within meta at the level of the channel we don't really distinguish if it's facebook instagram or whatever and i think it's overhyped as it's getting really tough to get to good user economics there you need to try really hard to get to good user economics there you need to try really hard to get to good user economics there you need to try really hard to get to good user economics there you need to try really hard to get to good user economics there you need to try really hard to get to good user economics there you need to try really hard to get to good user economics there really really really hard with creatives and with the funnels to get it right at scale and when we're to get it right at scale and when we're to get it right at scale and when we're talking about hundreds of millions in talking about hundreds of millions in talking about hundreds of millions in spend what's under hyped why are not many spend what's under hyped why are not many spend what's under hyped why are not many people youtube absolutely you see great people youtube absolutely you see great people youtube absolutely you see great return on youtube return on youtube return on youtube i think youtube can be great when we think i think youtube can be great when we think i think youtube can be great when we think about google ecosystem so about google ecosystem so about google ecosystem so google search is like overhyped the whole google search is like overhyped the whole google search is like overhyped the whole demand generation piece of google is demand generation piece of google is demand generation piece of google is underhyped and underhyped and underhyped and youtube is a huge part of that we just youtube is a huge part of that we just youtube is a huge part of that we just started entering youtube with like started entering youtube with like started entering youtube with like short-term and long-term videos and it short-term and long-term videos and it short-term and long-term videos and it can become huge it can beat all discovery can become huge it can beat all discovery can become huge it can beat all discovery channels so funny the people who've done channels so funny the people who've done channels so funny the people who've done it so so well it's actually monday.com it so so well it's actually monday.com it so so well it's actually monday.com absolutely yeah i had a founder on he absolutely yeah i had a founder on he absolutely yeah i had a founder on he was it was a real game changer for us was it was a real game changer for us was it was a real game changer for us and at the scale of monday that's pretty and at the scale of monday that's pretty and at the scale of monday that's pretty fascinating but i think b2c companies fascinating but i think b2c companies fascinating but i think b2c companies are underestimating youtube it can be are underestimating youtube it can be are underestimating youtube it can be great it can be like really big what great it can be like really big what great it can be like really big what have you changed your mind on most in have you changed your mind on most in have you changed your mind on most in the last 12 to 18 months i think the last 12 to 18 months i think the last 12 to 18 months i think how the growth team should look like how the growth team should look like how the growth team should look like unpack that one i think if we were unpack that one i think if we were unpack that one i think if we were talking about user acquisition managers talking about user acquisition managers talking about user acquisition managers and growth pms you need to bring native and growth pms you need to bring native and growth pms you need to bring native ai people like 17 years old teens ai people like 17 years old teens ai people like 17 years old teens who spend their time on tick tock like who spend their time on tick tock like who spend their time on tick tock like 24 hours a day who can bring you real 24 hours a day who can bring you real 24 hours a day who can bring you real leverage there and their native in leverage there and their native in leverage there and their native in idols i think this is like a natural idols i think this is like a natural idols i think this is like a natural addition to all ua addition to all ua addition to all ua teams i get you i've just got uh having teams i get you i've just got uh having teams i get you i've just got uh having worked with many of them the challenge worked with many of them the challenge worked with many of them the challenge that they bring is unreliable listen that they bring is unreliable listen that they bring is unreliable listen and that's a very general statement but and that's a very general statement but and that's a very general statement but like they are just inherently very like they are just inherently very like they are just inherently very unreliable they have very little respect unreliable they have very little respect unreliable they have very little respect for things like time like you'll say for things like time like you'll say for things like time like you'll say i'll meet we'll meet at 10. they'll rock i'll meet we'll meet at 10. they'll rock up at 10 45 and text 10 minutes before up at 10 45 and text 10 minutes before up at 10 45 and text 10 minutes before being like where are we meeting again being like where are we meeting again being like where are we meeting again it's a completely different world and it's a completely different world and it's a completely different world and it's like oh i didn't do that we agreed it's like oh i didn't do that we agreed it's like oh i didn't do that we agreed that you'd do that i would probably run that you'd do that i would probably run that you'd do that i would probably run 100 interviews find the best one who can 100 interviews find the best one who can 100 interviews find the best one who can do it on time and hire one my hard thing do it on time and hire one my hard thing do it on time and hire one my hard thing with these guys also in girls is the with these guys also in girls is the with these guys also in girls is the really good ones know they can make a really good ones know they can make a really good ones know they can make a ton of money doing it for themselves i ton of money doing it for themselves i ton of money doing it for themselves i know so why would you i think you bring know so why would you i think you bring know so why would you i think you bring them for a very short time period them for a very short time period them for a very short time period they enjoy they enjoy they enjoy working for like a big brand working for like a big brand working for like a big brand getting some recognitions from their getting some recognitions from their getting some recognitions from their friends whatever and after that they're friends whatever and after that they're friends whatever and after that they're gonna leave you but it's fine actually gonna leave you but it's fine actually gonna leave you but it's fine actually yeah one who's amazing is zaria parvez yeah one who's amazing is zaria parvez yeah one who's amazing is zaria parvez who's um duolingo i actually had her on who's um duolingo i actually had her on who's um duolingo i actually had her on the show because she's done all of the show because she's done all of the show because she's done all of duosocials duosocials duosocials and she stayed for five six years which and she stayed for five six years which and she stayed for five six years which is amazing 16 years is like incredible is amazing 16 years is like incredible is amazing 16 years is like incredible for any growth person in any company for any growth person in any company for any growth person in any company insane are 90 of growth leaders full of insane are 90 of growth leaders full of insane are 90 of growth leaders full of yeah what's the biggest bs that you hear yeah what's the biggest bs that you hear yeah what's the biggest bs that you hear from growth leaders on podcasts from growth leaders on podcasts from growth leaders on podcasts what best frameworks look like and how what best frameworks look like and how what best frameworks look like and how a different heads of growth a different heads of growth a different heads of growth share their best failures share their best failures share their best failures and best wins which is commonly fluff and best wins which is commonly fluff and best wins which is commonly fluff and very very very far from the reality and very very very far from the reality and very very very far from the reality and how you run the operation in the and how you run the operation in the and how you run the operation in the reality reality reality the metrics and the benchmarks which are the metrics and the benchmarks which are the metrics and the benchmarks which are average and average commonly absolutely average and average commonly absolutely average and average commonly absolutely misleading misleading misleading or the the benchmarks that you need to or the the benchmarks that you need to or the the benchmarks that you need to at least be looking at but then like at least be looking at but then like at least be looking at but then like your product is different how you even your product is different how you even your product is different how you even look at those metrics and benchmarks you look at those metrics and benchmarks you look at those metrics and benchmarks you know that flow doesn't have real know that flow doesn't have real know that flow doesn't have real competition so we look at duolingo competition so we look at duolingo competition so we look at duolingo metrics metrics metrics and benchmarks is it really comparable and benchmarks is it really comparable and benchmarks is it really comparable not really and when others compare to not really and when others compare to not really and when others compare to floor or to best players in the market floor or to best players in the market floor or to best players in the market are playing the same game you don't have are playing the same game you don't have are playing the same game you don't have your own value and leverage but then your own value and leverage but then your own value and leverage but then like how you're comparing to that median like how you're comparing to that median like how you're comparing to that median help me out growth advisors i always help me out growth advisors i always help me out growth advisors i always want internal i just think it's better want internal i just think it's better want internal i just think it's better to have things internally done how to have things internally done how to have things internally done how should founders think about the decision should founders think about the decision should founders think about the decision to have a growth advisor versus i should to have a growth advisor versus i should to have a growth advisor versus i should hire a growth team that's internal and hire a growth team that's internal and hire a growth team that's internal and 24 hours a day i think the way you think 24 hours a day i think the way you think 24 hours a day i think the way you think about it is you need an internal growth about it is you need an internal growth about it is you need an internal growth team fully integrated team fully integrated team fully integrated and then you hire like a couple of hands and then you hire like a couple of hands and then you hire like a couple of hands on people who can run the show on people who can run the show on people who can run the show and as we said like young hungry head and as we said like young hungry head and as we said like young hungry head of growth who can manage the show by of growth who can manage the show by of growth who can manage the show by still being hands-on still being hands-on still being hands-on you will miss that big strategic you will miss that big strategic you will miss that big strategic perspective and experience and knowledge perspective and experience and knowledge perspective and experience and knowledge from the industry that the growth from the industry that the growth from the industry that the growth advisor can bring but then it's advisor can bring but then it's advisor can bring but then it's probably a matter of how you integrate probably a matter of how you integrate probably a matter of how you integrate them and how you work with them them and how you work with them them and how you work with them they can give you everything and that can they can give you everything and that can they can give you everything and that can give you nothing it really depends on give you nothing it really depends on give you nothing it really depends on what does you ask how do you work what does you ask how do you work what does you ask how do you work together together together how do you set the whole processing and how do you set the whole processing and how do you set the whole processing and mentoring thing around that for example mentoring thing around that for example mentoring thing around that for example the way i work the way i work the way i work i commonly mentor the head of growth we i commonly mentor the head of growth we i commonly mentor the head of growth we solve weekly challenges i'm being fully solve weekly challenges i'm being fully solve weekly challenges i'm being fully integrated into the team integrated into the team integrated into the team i learn the founder i learn how they i learn the founder i learn how they think how they create think how they create think how they create so i can like be fully integrated and so i can like be fully integrated and so i can like be fully integrated and support the growth and understand their support the growth and understand their support the growth and understand their product and their challenges as opposed product and their challenges as opposed product and their challenges as opposed to you know just come in tell a couple to you know just come in tell a couple to you know just come in tell a couple of fluffy of fluffy of fluffy smart sentences let them break their smart sentences let them break their smart sentences let them break their heads around the execution and go away heads around the execution and go away heads around the execution and go away yeah six months well paid thanks guys yeah six months well paid thanks guys yeah six months well paid thanks guys yeah and then by the time by the time yeah and then by the time by the time yeah and then by the time by the time it's done it's like ah i don't know if it's done it's like ah i don't know if it's done it's like ah i don't know if it was really successful absolutely what it was really successful absolutely what it was really successful absolutely what do you know now that you wish you'd do you know now that you wish you'd do you know now that you wish you'd known when you started in growth known when you started in growth known when you started in growth that 100 experiments at the end that 100 experiments at the end that 100 experiments at the end compound into something very big and compound into something very big and compound into something very big and even if 95 of them gonna be a feather even if 95 of them gonna be a feather even if 95 of them gonna be a feather you're gonna win big after getting all you're gonna win big after getting all you're gonna win big after getting all those insights and knowledge which those insights and knowledge which those insights and knowledge which company would you most like to be head company would you most like to be head company would you most like to be head of growth for of growth for of growth for that you haven't worked with but not that you haven't worked with but not that you haven't worked with but not because it's like a sexy company or a because it's like a sexy company or a because it's like a sexy company or a cool company cool company cool company but because you think they would most but because you think they would most but because you think they would most benefit from a growth playbook that you benefit from a growth playbook that you benefit from a growth playbook that you would bring open ai is huge they have a would bring open ai is huge they have a would bring open ai is huge they have a very interesting playbook of consumer very interesting playbook of consumer very interesting playbook of consumer virality going up market currently to virality going up market currently to virality going up market currently to end price end price end price i would love to lead grow there and see i would love to lead grow there and see i would love to lead grow there and see how we can complement the playbook that how we can complement the playbook that how we can complement the playbook that they are currently running they are currently running they are currently running katie listen i love this i i guess you katie listen i love this i i guess you katie listen i love this i i guess you saw with my tweets i hate it when people are fluffy or bring first principles or just bluntly don't answer the question. You've been fantastic. I've so enjoyed this. So thank you so much for doing this with me. And it's been awesome. Thank you. That means a lot. It was a pleasure. But before we leave you today, SecureFrame empowers businesses to build trust with customers by simplifying information security and compliance through AI and automation. Thousands of fast growing businesses, including NASDAQ, AngelList, Doodle and Coda trust SecureFrame to expedite their compliance journey for global security and privacy standards, such as SOC 2 and ISO 27001, CMMC, NIST standards and more. Backed by top tier investors and corporations like Google and Kleiner Perkins, the company is among Forbes's list of the top 100 startup employers for 2024. G2's best software awards for higher satisfaction products and a recipient of the 2024 Cyber Security Excellence Awards, something I definitely never got in school myself. Whether it's the software you build or the software you buy your tech stack should be creating results not creating roadblocks well pendo's no code software experience management platform makes your software better with tools to see where users get stuck guide them with in-app messaging and constantly improving your ui it's so easy that over 14 000 businesses use pendo to increase revenue lower costs and reduce risks businesses love the control engineers love the freedom everyone wins start for free today at pando.io forward slash 20 product.