Optimizing Funnels, Pricing, and Retention at Zumba — Nicole Page & Lucy Levy, Zumba

0:00Hello, I'm your host, David Bernard. My guests today are Lucy Levy, chief consumer officer, and Nicole Paige, senior product manager at Zumba, the largest branded fitness company in the world. On the podcast, I talked with Lucy and Nicole about how customer-driven iteration led Zumba from VHS tapes in 2001 to launching an app in

0:242024. their app to web experiments that boosted LTV by 17%. And how they are able to charge for content when countless Zumba classes are available for free on YouTube. Hey Lucy, thanks so much for joining me on the podcast today. Thank you for inviting us. And Nicole, nice to have you as well. Always fun to have two folks uh from the same company to kind of have different

0:48perspectives. Thank you, David. I'm looking forward to the conversation. So, I wanted to kick things off with the story of Zumba. And we don't always dive into the story of apps, but Zumba's really interesting. You know, it's been around for 24 years and has kind of become synonymous with dance-based fitness. It's kind of the Kleenex of fitness in that a lot of people just say Zumba to refer to any kind of dance-based fitness. And and that's a remarkable achievement as a company. And you know, the the app, as we'll get to, is is very new to the story of the Zumba brand. Uh, but let's go all the way back

1:28and talk through the founding of the Zumba app. And I think it'll be especially interesting to folks like a indie developer looking for that next new app to build or a, you know, somebody at a bigger app company thinking about secondary product market fit or even, you know, somebody a big app company building a side project. Uh, I think there's a lot of nuggets in the story of Zumba of how to think about, you know, what kind of product to build, whether or not you're feeling that kind of pull of product market fit and things like that. So, Lucy, take it away. Tell us the story of how Zumba was founded 24

2:01years ago. Yes. Yes. Uh, I'm happy to share the story and being a big fan of the podcast. So, thank you so much again for inviting us. Funny that you say that Zumba Clinics that that example is on point. It happens very often that people just think that we are fitness discipline or a fitness category, not necessarily a private brand. But I'll

2:22tell you a little bit about the story. So Zumba has been around for 24 years, which in the fitness world is unheard of and a lot of people had little faith that we're going to survive that many years, but there is a lot to it and a lot of power into the community. So Zumba started and it's a funny funny story. It's three Albertos, so same name, different backgrounds, but the three of them are Colombians. So big coincidence on that front. And it actually started with Alberto Perman, who is our current CEO and one of the co-founders. Uh who had lost his job. He had started a business, internet

2:56business 24, 25 years ago that obviously burst with the internet uh when the internet bubble burst. And he was out of a job, out of path. He didn't know what to do. He was having dinner at his mom's house and the mom said, "Why don't you come downstairs to meet my fitness instructor? He teaches an amazing class that everyone wants to attend." Of course, he was like, "Why the hell am I going to have to meet your fitness

3:20instructor? Like, this makes no sense." But as every mom, uh, she was very persistent and he finally came and they went to meet Beetto, which is the other Alberto. He was a successful fitness instructor in Colombia and was just getting started in Miami. But successful meaning that he had a class that had a lot of people, but he hadn't created a brand or or or a business around it, right? So, he was teaching at the at the building and the class was packed. A lot of people were outside waiting to join the class. But the biggest insight is

3:51that everyone was laughing and enjoying. So, Alberto Perman, who is super creative and very smart, he saw that situation. He said, "You know what? there is something here. People are actually exercising without suffering which is not something that we see very commonly right especially 24 years ago. So then you know they invited the other Alberto who is our COO um the three of them are still super active and involved in the company and they decided to create a VHS which eventually went into a DVD to start selling online. They got inspired by Tao and Billy Blanks. So they started selling the DVDs through through TV, through a license and and a

4:29few other partners and they were getting some success with that. But the pivotal moment was when viewers or people that had purchased the product started calling the call center of the infomercial company saying that they wanted to become instructors or they actually wanted to join an inerson class. So as you know the infomercial products normally go through a cycle where they even if they succeed at some

4:51point they die down and they disappear. So they were already starting to feel that that could happen and they were already trying to think, okay, what else can we do? So they decided, let's do a training in Miami to license instructors. Let's see how many people show up. If more than 60 people show up, we keep going with that idea. If not, we all have to start thinking about what is

5:11going to be our, you know, career path. Uh because these DVDs were not necessarily very sustainable long term. So it ended up happening that about 120 people showed up to the training in Miami. super passionate, you know, very excited. And so they continued to replicate that situation. So they were training people all around the country and long story short, a lot of those people wanted to take the training multiple times because they needed support and they decided to create the Zumba instructor network, which is our

5:41uh core business and our biggest asset. It is our community, our mission as a company is to make our Zen members successful. So how it works is that normally you take an intro training which is the basic one Zumba training and then as you take that training you have the option to join the Zumba instructor network where we provide a lot of benefits including apps for them to teach uh music choreography marketing materials license to teach around the

6:07world. So a lot of them end up joining. And lastly, what happened? So this company was built based on, you know, a customer feedback or user feedback if you want to call it that way, just to find parados with the app. The last thing what happened is that people were actually tattooing the logo of the company and they were actually creating clothing. So they were using white t-shirt or whatever and they were putting the brand in their clothing. So the third revenue line is an apparel brand that's pretty successful. It's like the uniform of our instructors and uh we sell through e-commerce and to distributors internationally. So long

6:41story short, the instructor license certification is one uh part of the business. Then the Zumba instructor network is the second part and then apparel is the third part. Company has been around for 24 years. We're in 180 countries. We're super global. We have trainings happening and instructors teaching in 180 countries on an ongoing basis. and we have 15 million students

7:03taking Zumba classes on a weekly basis. So super proud of the accomplishments. There's one interesting data point. There are more than 200,000 locations uh having Zumba. There is less gyms around the world that exist. So there are more classes happening because they happen in schools, in churches, in parks. So Zumba is actually bigger than the than actually the gym locations. But you're all wondering how do we end up doing an app? Obviously pandemic came a lot of disruption specifically uh obviously in the fitness industry and we were first of all our initial reaction is how can we support our instructors we stay true to our mission so we decided that let's

7:43not do an app to compete with our instructors because it's not what we do and let's actually create a system for our instructors to to teach virtually while while the pandemic was happening. So, we did that. But in the meantime, of course, we've realized that people went back, you know, five years later, people went back to their in-person classes. A lot of gyms where some instructors were teaching had shut down. So, we were hearing from people that they wanted to take Zumba again, but they didn't have a class near them. Um, so that's when we decided to do the app. We launched the app about a year and a half ago. It's

8:15been fascinating, a lot of learnings. We have a class locator in the app. So the app is mostly especially the subscription part to access high quality content but we all we actually have a class locator where we can match you with a local instructor. So eventually once you feel comfortable you can try an

8:30inerson class. It's been quite a ride. We're still around. We're strong. Uh people do Zoom around the world every single week. So happy to be part of the company. Such a cool story 24 years later to now be like one of the hottest fastest growing fitness apps on the app store uh 24 years later and again you know launching an app in 2024 as a fitness brand feels late you should have had an app in 2010 or 2012 you know you've been around that whole time uh but I think it is really fascinating how um missiondriven you are and how the instructors and the classes are so core

9:10to the brand. Uh so Nicole, I want to ask you, you know, as Zumba started thinking through what could an app look like? How do we do that and not kind of kill what's been so successful for us? How do we not alienate our instructor? What was the thinking behind all of that? How the product came to be after the pandemic and as you started thinking that you really did need to build an

9:34app? Yeah, definitely. So we actually at Zumba we have a lot of experience with apps because we have a number of apps that we've built for our instructors. So we have a lot of apps that our instructors use to put together their playlist to teach their classes to organize their social media. But during the pandemic when we realized that we needed a way to bring Zumba to people around the world who weren't able to participate in the classes during the pandemic we started doing a lot of research and we went through many iterations for the app and we tested a lot of hypotheses during this research

10:07but the one that we found was that a lot of people wanted to do Zumba but they were a little shy or they had been out of fitness for a while and they didn't feel comfortable going to the classes. So the niche that we found ourselves in that would also help our instructors was get beginners, get people who are coming back to fitness, get them into the Zumba app, get them comfortable with doing classes, and then eventually they would

10:32go and they would find an inerson class. So that was really our sweet spot. And we continued to interview, we talked to a lot of instructors, we talked to a lot of potential students. And everyone was telling us that they want to get comfortable with the classes first. they want to feel like they're confident um so they can eventually go and find a

10:50class. I'd love to dig a little deeper there. I think myself and so many app developers, especially, you know, indies don't necessarily think they have the resources or whatever, are so bad at user research and talking to people. What did that look like for Zumba? Did did you already have a lot of email? Did you do phone calls? Did you do Zoom? Was it a survey? How how did you actually do that research and kind of start to uncover the insights that then became

11:18the app? We're so fortunate as Zumba because we have so many people that love Zumba like Lucy mentioned, you know, tattoo getting level of love for Zumba. So, we've always had tons and tons of our instructors who have been willing to talk to us and we talked to them regularly. when it came time to just finding students, we did go to classes in person and we talked to students in classes, but we also just have a lot of people in our database that are were really willing to talk to us. So, we talked to hundreds of students and instructors over the course of the app development, also way more than that in

11:58surveys. So, luckily, people are really willing to talk to us. They want us to make Zumba better for them. So, it was always really easy for us to find people to talk to and we got them on Zoom, we got them on the phone. We did a lot of different kinds of interviews where we would do structured interviews, but we also did things like a study where we tried to get them to do more workouts throughout the week and we kind of just treated it like a check-in and we would let them talk to us and not ask them as many questions. So, we did a lot of

12:27different iterations and got a lot of great feedback that helped us shape the app. Very cool. As you were doing that research, how did you think about, especially as you kind of honed in on, you know, beginners and people returning to fitness being where you needed to focus? How how did you think about not overindexing on the dieards, not overindexing on the instructors who've been with you 15 years, not overindexing on the participants who have Zumba tattooed on their arm? Uh, how did you balance, you know, kind of filtering the feedback of, okay, here's here's a segment of like die hard and then, you know, but we really do need to still

13:06find those kind of new users to really get that fresh perspective in addition to the dieards. Yeah, that was so important. We spend a lot of time looking at our data and pulling out specifically people that downloaded the app and then churned or took one class and then churned. And we spend a lot of time talking to those people as well because obviously we want to build the app for the people that love Zumba, but we also want to make sure that we're bringing people in and letting them engage with the product enough to find the love for it. So, we're trying to get people that churn

13:37out early and talk to them and figure out what is it that made you leave? What is it that did you find another product that works for you or did you just give up fitness altogether? We're trying to we spend a lot of time talking to them as well. I would want to add um yeah definitely cohort research. So for sure we we're very mindful of that. But what's interesting is that in the findings a lot of the people that end up turnurning is because they rather go to an in-person class which is also a good situation for us. So it's interesting

14:04the findings that that we're getting. Obviously not great for the app but great for the business overall and it really stays with our mission. Yeah. And one of the the big challenges that that y'all were sharing with me in preparing for this uh podcast is that there are so many free Zumba lessons on YouTube at this point. 24 years in. I love that it went all the way back to VHS that the company started by sending out VHS tapes. So you have 24 years or I guess at the point where you start you know started working on the app you know 21 22 23 years of content that's been

14:41built up and not just by Zumba the brand but by instructors by influencers maybe even some people who don't even realize Zumba was a brand and they just created a YouTube video that says hey look at my Zumba you know dance fitness how did you think about not only balancing kind of competing against the the instructors in the live classes is, but then actually building something that would differentiate the app from just being able to get free and and charging for something when you could just go to

15:11YouTube and get it for free. Yeah, we actually specifically to the YouTube question, we ask ourselves all the time when building our road map, how is this better than YouTube? How what would make someone want to pay for this when they can go to YouTube and get it for free? And one thing that we learned as we built the app, as we experimented with different features was people were looking for and were willing to pay for a guided experience, not just a content repository. So one thing that we built

15:40and and tested last year was programs. So instead of just dropping them into the app and giving them a bunch of content to choose from, we started developing goal-based programs. So, as they came in through onboarding, we asked them about their goal, whether it was weight loss, getting started, just having fun or reducing stress. And then we would build them uh content packages or programs based on their goal. And it gave people a lot less decision paralysis. They were able to come in, get to a program right away, get guided through the classes that were the perfect fit for them rather than having to sort through content, which is what

16:14they have to do on YouTube. So, once we built programs, we saw a huge improvement in our numbers. people that joined programs um were watching two times as many videos as people that weren't enrolled in the programs. Our month-to- retention went up. It about doubled after for people that joined programs. So, we saw that this was really a differentiator for us and we continued to build from there, releasing more programs and building more features that reduced this choice that users had

16:42to make when they got into the app. Yeah, it is such a fascinating thing. I mean, so many apps are competing with other apps, but you're also competing with YouTube, you're also competing with yourself, but also really cool that that it's a broader business that you can find success as a business if somebody uses the app a few times and then starts going to the live classes that still benefit the business. But that differentiation is key. Like, what is it about the app that differentiates it from other fitness apps, from free YouTube content? any other kind of like key insights or features or or things that were kind of aha moments for the

17:19Zumba team as you were building this out that really became those kind of core valuable pillars that people were willing to pay for. A lot of it for us was quality and making sure that our content was super high quality. The feedback that we got from users was that when they went to YouTube there was such a wide wide range because you anybody can post a video and like you mentioned they don't even have to be doing Zumba. they just call it Zumba and it's so hard to find what you're looking for and people would tell us that they would go through 5 10 15 videos um and they just don't have the

17:50time or patience to do that. So we really focused on highquality content and we did a lot of research to figure out what content it is that they wanted. Um beginner content in the in the beginning we had a lot of really varied content. Um we still have a lot of varied content but we learned from users that a lot of people are looking for beginner content because they're just getting started. They're getting back into fitness. So, we started really filming a lot of high-quality beginner content that can get them started. So, just getting the really high-quality content that users want, getting it in front of them and guiding users to the

18:20content that they are looking for as soon as possible so we can get them to that aha moment, which we've learned is three classes. Um, so we see that once users get to three classes, they're much more likely to stay with us. So, they really have to experience a little bit of it and then they they stay and they love it. I would want to add that uh of course we've seen great improvements over the year and a half, a lot of learnings uh through features and also acquisition and growth strategies, but I think one thing that we're still looking to unlock is how to build a a community

18:51through the app. The magic of Zumba, it's really in the power of the community. We really don't think that it can be replaced with a virtual experience. I think the in-person experience is is always going to win. But uh we've tried with some basic feature of community. we we didn't succeed, so we turned it off. But we this year and hopefully next year, we're going to have the we're going to dedicate time into thinking how can we make and build communities online through the Zumba app. Uh so we can actually enable what happens a little bit, you know, replicate a little bit

19:21what happens into the in-person class. Yeah. And that's such an interesting kind of growth factor too in that the magic of community in person is incredible. Like you said, you can't replicate that for I mean probably hundreds of millions of people around the world. Even though you're in 180 countries, there might not be a local Zumba class that they can join. So finding ways to to replicate some of that magic, get a little hint of the magic inside the app could be a really powerful unlock for that kind of next

19:51phase of growth. Uh so yeah, really fascinating that you try any kind of lessons from from having failed? uh like what do you think didn't work about the your first attempt at that? So our first attempt at community was we had done some research and tested a few hypotheses for where we wanted to start and our hypothesis was that people want to talk about the classes after they take them. Um so we ended up building a bunch of different chat threads that you could join and talk about the classes or talk about a health journey or talk about um just starting out Zumba. And we kind of were thinking, you know, if if

20:27you build it, they will come. And we built these communities. And what we found was that people really didn't want to engage, not in the context of the class. They weren't looking to take the class and then go into a community and talk about it. So, in the next iteration, we're thinking about how we can do community a lot more contextually. And this kind of goes back to being in a live class. You know, you're in the class, you're engaging, you're high-fiving people in the class, you're talking to your instructor after, your instructor's asking you for feedback. I'm smiling at Lucy. Lucy's smiling back at me. There's all of these

20:58different interactions in a class that I think are a lot more powerful than this approach of people wanting to discuss the class um before or after. So, that was what I believe didn't work for us. And we're going to try again with a more contextual approach. Community is so tough. So hopefully people can can learn some lessons from from that and and keep an eye on what Zumba does because I'm I'm sure as you continue iterating uh you're going to find something that does work and that a

21:25lot of other apps will will learn from. The next uh kind of section I wanted to talk about is iteration. And I've really appreciated kind of the growth mindset I've heard from both of you in preparing for the podcast is that there there does seem to just be a real growth mindset that even though you know you had a big launch with the initial launch of the app in 2024, there's just been hundreds of iterations and a constant like small tweak to this, small tweak to that and a constant kind of growth mindset. Um before we dive into like specific experiments and some of the things

22:00you've been working on lately, any any kind of higher level thoughts about how you approach and how you instill this growth mindset with the team and and at a 24 year old brand, you know, that is kind of unique to have such a growth mindset. So any thoughts there? Yeah, it's such an interesting point. So it it was a challenge in the beginning for sure because Zumba we have established departments we have established pricing we have I mean we always are looking for our next lever or what else can we do to continue to improve the business and continue to grow but at a much different pace than

22:32what we're doing with the app. So the app is a small team. It's a very small team. Super agile. And we got into this growth uh iteration like test, fail, move on, test, you know, succeed, move on. And even when we're having good results, we never settled. We're always thinking, okay, what else can we do? Should we do another pay wall test to continue to improve the results? So it's a it's a small team. And I think in this case, it helps because they they really allow us to move fast. And it's just a culture that we've built in the past year and a half. But it hasn't been

23:02easy. It's been a process, but uh I don't know, Nicole, how you feel, but I'm super proud of the progress that we've been able to make both on the on the results, but also on the on the team culture. We do a lot of great research before going into our next feature um building out our road map. But we treat every new launch as a experiment and as a hypothesis that we're testing. So when we launch something, we are constantly checking the metrics. We're constantly checking the performance. um when something isn't working, we'll do that analysis and we'll say, "Hey, it's not working. Let's try something else." So,

23:37we're not, you know, married to any of our ideas, no matter where it came from. We just want, we all want the app to be better. We all want to see continued growth. So, we treat everything like a continued experiment. And even features that we launched last year, we check in on them and say, "Is this still working for our users? Is this still delivering

23:55the results that we need?" What does your stack look like? Are you using like amplitude or mix panel and are you using like a AB testing tool specifically or using that inside a product analytics like what does the stack look like and how do you like dive into the data to better understand how

24:14those things are performing. Um so we use mix panel heavily and we also use an internal tool that we built for AB testing. So the combination of that with Mix Panel, we're constantly looking at AB test data, retention, churn, acquisition data, a lot of different performance metrics uh that we've built into the app, everything. We're looking at it every day. Lucy and I usually by 8 a.m. we're pinging each other about what we're

24:38seeing in the data in Mix Panel. Revenue Cat has been a great tool for us to to test pricing and pay walls. Um so super happy and and also look at the metrics. So we have like three sources of of data. So, one of the big things I want to dive into because I think a lot of folks are really curious about this right now is appto. I don't know who came up with the web two app, the web number two app, but now because that

25:04became the the standard way to say it. I've started saying app to web with the number two. But with the recent injunction against Apple that does allow linking out, I know a lot of people are starting to experiment with that and seeing mixed results, but y'all seem to be doing pretty well with it. So tell me, you know, what you've been experimenting with and and how that's

25:26gone. Nicole can explain better what we've been experimenting, the actual product changes, and I can talk about the results. So really high level uh we've done two iterations uh of testing and what we've seen is that there is definitely an initial drop in conversion. So we lose about 35% of the people converting either to a trial or a paid. So our pricing is we have annual at 99 with a 7-day trial and we also have the 1499 monthly price with no trial. So we've seen that that initial conversion uh we saw a drop initially of about 35%. With the second iteration we managed to improve that to about 25%

26:02drop instead of 30 35 but because conversion from trial to paid on the annuals is much better on web. we recoup some of that loss and on top of that uh obviously through web we have much better retention and the lifetime value therefore it's better and we avoid the Apple Apple fees so when we look at everything together we see a 17% lift on lifetime value that more than offsets the drop in in the initial conversion we use the revenue cat tools to build this so that initially we tested before the package selection feature was available with revenue cat we initially tested just letting people check out um click

26:44the app payw wall, go to the web, and we compared that against just the inapp payw wall. And then once Revenue Cat released the package selection, we developed it so that users could actually choose which one they wanted monthly or annually, which package. And once they chose that, we would send them to the web and we gave them the most stripped down view on the web where we just showed them what they selected. We tried to replicate how Apple has that view, has that window on the app when you do an inapp purchase so that it looks very familiar and that actually helped us with the conversion a lot as

27:16Lucy mentioned. So with the payw wall, are you showing inapp and web or are you showing only web? And then in these experiments, were you AB testing specifically one payw wall that was only inapp purchase and one payw wall that was only web or tell me what how you've been setting up those payw wall tests. More specifically, the tests that we've done so far, we were testing just only inapp versus only web. Lucy mentioned we're still continuing to improve on this. So even though we've seen a big jump in our lifetime value, we know that there's still room to improve the conversion that we see drop when we send them to

27:55the web. So in the future, we're talking about testing some other approaches. For example, a payw wall that allows them to check out either they can choose between checking out on the web or the app. And we're also thinking about some other experiments to try to improve the conversion. Gotcha. So, so then the way the payw wall currently works now with the package selection would be that it looks like a standard like inapp payw wall, right? So, you have a monthly selector, annual selector, and then you have the call to action button of, you know, check out, continue, subscribe, whatever that actually I'm curious what what's

28:28the call to action on the button that you've landed on. We've landed on start today. um tested a few different ones and expert today seems to be motivating for people and also hopefully encourages them to actually go and try their first class. So that's where we are right now. Gotcha. Okay. So then package selection and then a a big button that says start today as your call to action. And so when they tap start today, it bounces out to Safari, but then that page on Safari is super stripped down. No more package selection, no other options, just a checkout button. And then what are you using on the I I guess Zumba

29:04having been around 24 years you already had uh payment system and other things that you had built out on the web but are you offering Apple Pay or what is that web checkout experience like? We use Stripe for the web checkout experience but they can also pay with Apple Pay or Google Pay and one of the optimizations that we've made that we've seen um help us a little bit with the conversion is we default to what we see that the user has available. So, if they have Apple Pay, we'll default to that one selected first. If they have Google

29:33Pay, we'll default to that one first. And if they don't, then they'll just uh see the credit card option. Gotcha. Yeah. One of the feedback we got on Revenue Cats test with our app dipsy was that when you go to the checkout page, it was like kind of a big form and so even though it said like Apple Pay, you know, at the bottom, it still looked like you had a lot of work to do. So I imagine like what you're saying is you detect in the web browser that they have Apple Pay available and then you don't even show a credit card form, right? You

30:04just show like check out with Apple Pay, right? We just we make it as simple as possible if they have Apple Pay. Nice. Yeah, I I can see why that's increasing conversion. That's the next test we should probably do with the with the dipsy payw wall on the web. One debate that we had internally was whether we give an extra incentive to get people to subscribe through web but we decided to test it without it because obviously we wanted the help on the on the lifetime values. We didn't want to give that extra incentive and and I don't know that people know so much the difference between you know with this

30:34seamless experience I don't think there is uh there is so much hesitation and we're seeing it in the numbers right we still have room for improvement in that initial conversion but I think we've done we've done well this is one of the most successful tests that we've had yeah I mean a 17% increase in lifetime value is fantastic one of my theories about how this is all going to play out in the long run is that apps will eventually start to get a little more sophisticated during onboarding to kind of understand user intent and kind of find those higher intent users and send those to the web to help offset some of

31:10that decline in conversion. You know, it was a big drop 35% drop in conversion on that first test and then great to see that trial to paid did help offset that and then further tests have even further reduced that drop in conversion. But how are y'all thinking about potentially adding another layer in there? Potentially sending some of those lower intent users and just make keeping them in the app to make sure it's like ultra seamless. What are you thinking on that

31:36front? One thing that we're thinking about doing is optimizing the payw wall for where they are in the app. So changing the message, changing the payw wall, potentially changing which option we give them based on where they are. So maybe if they're in one of the places in the app where the payw wall doesn't convert as well like potentially some of the ones that they see sooner in the app or one of the ones that they see in the settings potentially using some different messaging or or just using the inapp payw wall for that versus some of the apps that we see converting really well. For example, we have a welcome

32:05class that we give them now. So when they come in uh we give them one class that they don't need to make a lot of decisions and we say try this one first and it's a fantastic beginner's class and we see that payw wall doing pretty well um because right after it we give them a payw wall. So those seem to be higher intent users. So maybe those are

32:20the ones that we can send to app to web. So we haven't done that yet, but it is one of the things that we're thinking about. Yeah, it's a it's an interesting point. We're just sending half of the people to one experience versus the other. We're not being very cautious about it, but it's a it's a great point to look at the data, understand which are higher intent users, and based on that, maybe present them with the the correct option. And then just to kind of fully close the loop, you've made the checkout experience as seamless as possible with the Apple Pay button and everything else. And I imagine as soon as they've

32:49checked out, there's a button. Does it auto redirect right back to the app or there's a button they press and they get right back into the app? And kind of like it's almost like a seamless you bounce to the web, make your payment, and bounce right back to the app. Exactly. We tried to when we were building out the experience, our north star was let's make this look as much like the inapp purchase as we possibly can. And so we auto redirect them right back and we try to make it as close to

33:14the inapp purchase as possible. Gotcha. Yeah. Yeah, that makes a lot of sense. Yeah. Again, some of the feedback we got from folks on our Dipsy experiment and our experiment didn't do too bad. I mean, we only saw a 6% drop in realized LTV on the initial conversions. Uh, which is not that bad. And then to to your point on LTV, what we see in the data is that far fewer people on the web have turned off autorenew. And so, you know, we're not, you know, calculating the exact predicted LTV of of what that's going to be because they're all annual plans, but it's looking like the LTV is going to be

33:52higher on the web over the long haul, even though it was a 6% loss in that initial revenue conversion. So yeah, this app to web is is going to be interesting as people start to discover the best practices and iterate their way toward toward making it work. Yeah, and it it can be very scary because you can see CPA climb and and you're like, oh, like you know, let me turn this off because if you don't focus on the big picture, you always get caught up on the CPA increasing, which is something that we saw. But then when you do the lifetime value calculation, I think that's when

34:24you see the value, right? Well, on that uh topic of retention, I know you have done a lot of experimentation and iteration on better retaining users. So, tell me about some of your work there. One of the things we've done with retention is trying to celebrate their wins and make them feel good about taking classes and and try to help them come back. So, we've done a lot of work here. One thing we did was a postclass experience. So after they take the class, instead of just letting them drop back into the app, we celebrate them. We tell them that they did a great job. We

34:57built a badge system, a gamification system, and they can reach different milestones. They can earn different badges. There's number of classes taken or there's number of steps taken. Um so we do a lot of we spend a lot of time celebrating them after the class. We also have started to ask them to rate the class after to improve our recommendation algorithm. So now when they take a class, we ask them if they liked it. We ask them if it was too hard, too easy, too intense, not intense enough. And right away, we recommend them the next class that they should take. And when they sign back into the

35:28app, we have it right there when they sign in at the top of the app. So, we're really trying to complete that loops for them so that they can come back in and they can take their next class. So, we've seen that really improve the amount of people coming back and taking their next class. Um, just trying to get them to the right class for them. We've also worked a lot on our communication journey. So, when is the right time to send them a push notification? When is

35:50the right time to send them an email? We're working on a few things right now to try to get their stats from the week sent to them so that they can really celebrate their wins and move on to their next week with confidence. So, we've done we've done a lot there. When is the right time to send a text

36:07message versus a push versus an email? I'm sure people are like you you kind of breezed over that. What are what are some of the lessons there? We're definitely still experimenting a lot with that. we're we're checking the performance of our emails and our push notifications and making changes um when our notifications aren't getting opened or when they are leading to optouts, we'll change it. But one of the things we're doing right now that we're testing is we're testing sending them a notification at the time that they usually go into the app. We say if a user typically works out on Mondays, we know that they're probably higher intent

36:39there. Let's send them a notification and remind them um that they can take their class. or one thing that we're trying is we see that usually going into the weekend, Fridays, we see a little bit of an activity dip. Um people are getting ready for the weekend. So, we're trying a notification there kind of on the flip side to see if we can get them excited about maybe take a shorter class, maybe take a fun class um and trying to suggest some things where they might not otherwise um work out. But timing wise, we're trying to send them the notification at the time that they're usually active also. So, you

37:09know, if someone is typically active at 9:00 a.m., we're trying to time it so that we send it at the time that they're active versus, you know, if they're never in the app in the middle of the day, maybe they're working. We don't want to send them a notification at 3 p.m., for example, if there's someone that's not available then. So, we're

37:24we're still testing this a lot. Yeah, that's great. Some great lessons there, though. Yeah. And just to add to the previous point uh regarding retention, I think one of the biggest shifts which Nicole talked a little bit about it in the beginning was really focusing on that audience, the beginner audience. We saw that about 70 even more% of the people coming through on boarding were saying that they were beginners. We did research to validate that. So now really we're really focusing the product on on that beginner experience, the intensity of the classes, the complexity of the classes to try to cater for that audience to make them feel like they can

37:57accomplish, you know, the workout and that they can feel as winners as they can be. And I know you've experimented on onboarding with longer onboarding and shorter onboarding and adding friction, removing friction. uh what are some of the lessons there in onboarding that you've learned and then how have you continued to optimize that towards that those downfunnel metrics and that longer

38:19time horizon? Yeah, definitely. I think this was a lesson in just challenge all of your assumptions that you have because for a while in the beginning um we were operating as if it was just best to get the user to the payw wall as quickly as possible, get them into the app and to

38:35the payw wall as quickly as possible. But um we had a hypothesis about longer onboarding. And we started experimenting with guiding the user through an onboarding experience that kind of got to know them a little bit. Also gave them some interstitials during onboarding to show the value of the app. And we saw that that onboarding experience actually led to a much higher conversion than without having the longer onboarding. And it also gave us the ability to personalize what they saw right when they got into the app. So, we asked them a little bit about what they were interested in, about what their level was. And we were able to recommend

39:08them a program before they even saw the app. And doing this just helped them see that we're listening to them, we know who they are, and we have what's right for them in the app. That greatly increased our conversion. And I think it was a surprise to a lot of us that we introduced friction, but it really did help us. And one thing we did after introducing the long onboarding was we actually introduced a payw wall before the users even got into the app which was another assumption that we challenged was that oh no they definitely have to see the product before we showed them the payw wall. But

39:39we were able to build enough value perception into the onboarding experience where we showed them a payw wall right after onboarding before they dropped into the product and that really helped with overall conversion as well. I heard one of you mention that that you now just give away one free class. Where did that come into the mix and how do you introduce that in the onboarding funnel and and whether they start a free trial or not? I guess just to better understand as well is the is the app kind of hard paywalled where you really only get that one class unless you start

40:09the free trial. No. So you can have that free class without necessarily having to start the free trial. We do get a lot of feedback still about users again because Zumba is so popular saying that they want more free classes. So we we always have this internal debate of like should we just keep it to one class or should we can can we give them access and I'm always pushing back because obviously I want the LTV to ratio to always work and I'm always pushing back saying like let's just keep that one class focus that class on beginner and try to get them to start their trial because they also have

40:40an option to navigate the entire app and access all the classes if they want it with their with their free trial. So that's why we landed on on that one class. Normally when they go through on boarding they see the paywell a big percentage of them convert but then the other big percentage that doesn't convert has direct access access to that free class if they choose to take that class whenever they cut it off even if they don't go through the entire class we present a pay wall. So that's a payable that as Nicole was explaining before, maybe we can contextualize it a little bit more because we get hundreds

41:12of thousands of views in that class and we can do a better job at converting people from that experience. Is that one class the same for every user or are you actually using the personalization to personalize that one free class? The one free class right now is the same for every user. The reason we did it like that is because the class is a Zumba 101. So no matter who they are, even if they take Zumba, it's a really good introduction to what you're going to see in the app. So it shows the basic steps. It kind of gets them familiar with the app format content. So for

41:47right now, we're showing everyone the same. But I think it's definitely a possibility in the future that we personalize that welcome class as well. Yeah, I just wanted to add as we are trying to prioritize the road map and what else do we do next? As we're seeing more than 75% of the people coming saying that they're beginners, the Zumba 101 seems like a good fit for for the majority. But eventually, I think there could be more more room for

42:07personalization. Yeah. Or even five different Zumba 101 classes where it is still a Zumba 101, but maybe it's just a little more focused on on the different aspects that that people are looking for. But I do love that, you know, as you were talking about introducing that friction. I mean, so many apps now will do the long onboarding, ask all these questions, but then not actually do any level of personalization. But it sounds like once they do start the free trial, you take into account a lot of that to then recommend those best classes that are going to be more retentive and a better experience based on that

42:42personalization. Yeah, absolutely. We start influencing their recommendation algorithm as soon as they start the onboarding, whether they start on the web or the app. So, as soon as they drop in, they already have a curated class experience. What do you consider your aha moment or or what are you kind of optimizing toward one class toward some other experience or what do

43:04you optimize toward? Uh we found that three classes seems to be the aha moment. After three classes, they really seem to be much more likely to be a long-term user. So we're trying to get them to the three classes where a lot of our communication early on is focused on getting them to the three classes. We have a class counter on the homepage encouraging them to get to three. In the onboarding, we actually have them set a goal for how many classes they want to commit to for each week. We default that goal to three. And we also have a little blurb about how just three classes a week can really

43:38help you see progress toward your goal. So, we encourage them to choose that three because we've seen that it helps that others in the app have succeeded with the three. And then throughout the app, we have some small reminders to get them to the three as well as the communication journey. Gotcha. I wanted to talk a little bit about price testing. So, you know, Lucy, you had mentioned early on that, you know, at the corporate level, you know, you've had prices and iterated on price over a very long 24 year history. How did you approach price testing in the app and what are some of the lessons you

44:13you learned there? Yeah, for the core business, we were super careful with pricing. So, we hire research companies, pricing study companies, and we take years before we make any change because we already have such a large community. For the app, we started with uh with research on what we thought people were willing to pay. But as you know, once you're live, it is completely different than the than the research. You try your best, but things can change. And that's when the culture of like testing and iterating really matters. So, we started with the monthly and the annual plans, but both of them were free trials. We

44:47believe in the beginning our monthly price price was $19.99 and the yearly price was $129 and we were seeing like really not great conversion. So, we started to iterate. We use revenue cut to AB test uh different pricing packages and after a lot of different testing which I still think that we have room for improvement uh we are now at monthly $14.99 no free trial which led to great results so I highly recommend it for anyone that is hesitating definitely try it and then contrary to many fitness apps we just give a 7-day trial on the annual plan a lot of apps give 30-day free trials or

45:24even 14-day trials but we saw that you getting them to a shorter trial led to higher commitment, higher engagement, them rushing into giving it a try and and we actually didn't see a drop in conversion. We saw a huge improvement in conversion from trial to paid. So when we started with uh with the app about 30% of our users even a little bit less were choosing the annual plan and the conversion was in the 30s from trial to paid. At this point we are getting about 60% of the users choosing the annual plan and about 56% conversion from annual conversion from trial to paid on

46:00the annual. So nice. Yeah. And it's a combination right? It's not only the pricing or the or the packages is the app to web and it is the first time user experience and the beginner content and it's it's exciting to see how everything combined really can lead to good results. Yeah. Yeah. And the nice thing there too is without the free trial on the monthly, then you are getting $15 upfront from the users who are choosing that, which I do think is fair in that, you know, $15 you'd pay that to go do your first Zumba class in person or whatever. Um, and so 25% of the users

46:35are paying you upfront without even starting that free trial. And then now 75% are doing the free trial and and landing on annual. So, it seems like a really great outcome for the business and and a fairer, you know, trade-off for the users. Yeah, for sure. All right. Well, I think that's a really great place to wrap up. It's so fun talking through all these iterations, all the learnings, and and just fun talking to a couple folks with a real growth mindset at a company that's been around 24 years like we keep talking about. So, thank you for for sharing the insights and it sounds like you've got

47:07hundreds of more tests to do over the next couple years and I think we'll be seeing a lot more uh about Zumba in the app community in the coming years. So, thanks for joining me and thanks for sharing so many insights. Thank you for having us. Thank you, David. Thanks so much for listening. If you have a minute, please leave a review in your favorite podcast player. You can

47:31also stop by chat.subclub.com. subclub.com to join our private community. [Music]