Learn how Fitocracy, a fitness social network, went from 0 to 1.5 million users. Dick teaches us why gamification can only account for so much, why the community is more important than anything else, and the importance of team-market fit.
TOPIC DICK COVERS
- He is the chief growth officer at Fit Talk, a fitness social network that combines community, gamification, tracking, and content to help users reach their fitness goals
- Fit Talk was created to help people make the same transformations as the founders, who were both overweight and skinny kids growing up
- The platform is designed to help users level up their fitness, and it uses community and gamification to keep users engaged
- Fit Talk has analyzed both qualitative and quantitative data to understand what keeps users engaged on the platform
- They utilized a lot of growth hacking to get to 200,000 users without an app
- How big has photography grown over the years
- Used customer development to gather feedback and add features that users wanted
- And a whole lot more
LINKS & RESOURCES
WATCH THE INTERVIEW
READ THE TRANSCRIPTION
Bronson: Welcome to another episode of Growth Hacker TV. I’m Bronson Taylor and today I have Dick Talons with us. Dick, thanks for coming on the show.
Dick: Thanks for having me, man.
Bronson: Yeah, absolutely. Now, Dick, you are the chief growth officer at Fit Talk, Chrissy. So let’s start there. Tell us a little bit about photography first. What is it?
Dick: So photography is a fitness social network. It helps everyone get to the next level of fitness. And we do this through a combination of community. That’s our main focus on photography. But we also leverage gamification and tracking as well as content and knowledge. So we help users who have different goals and might be different parts of their fitness journey. And we tell them, Here’s what you need to do next. Ten people you should connect to, and here is how many points you run for what you do. And here’s like a little squat for you. And so the entire platform is created to help people level up their fitness.
Bronson: Gotcha. So it’s kind of community. It’s kind of gamification, which seems like both of those would be really strong in terms of helping people get in shape.
Dick: Yeah, that’s exactly right. As it turns out, gamification is it’s really good at bringing people in and it’s really good at getting them moving at the start. But the diminishing returns and gamification kick in really, really simple. So community is actually what keeps people for a long amount of time. And we’ve done a lot of analysis to figure this out, both qualitative and quantitative.
Bronson: Yeah. And here in a minute, I want to get into the gamification a little more. You can tell us about that because that’s interesting. Tell us first what was kind of the impetus for starting photography? Was there a story behind there or something?
Dick: Yeah. So, you know, growing up, I used to be really, really fat kid, £220 of my max. I was super chubby. I was like the the kid that everyone kind of made fun of at school, actually more of a class clown than anything else. But, you know, supermarket at some point because I’m a geek, I started geeking out about fitness in the same way you got it about video games. And so I learned everything I could about nutrition, exercise, both the physiological aspects of that and also the kind of psychology and following the protocol and, you know, became an avid bodybuilder. And so when me and my partner Brian decided that we wanted to start a company, Brian also had a very similar background, except coming from the other end of the spectrum. He was a super skinny kid who was like a beanpole growing up. You know, we decided to start a company that would help people make the same kind of transformations that we made. And given the fact that we were both video gamers growing up, we started photography with the initial concept of turning fitness into a game.
Bronson: Gotcha. Now, I’ve heard that Arnold Schwarzenegger actually uses your product. Is that right?
Dick: He he actually does. And how he got Arnold up, it is a little bit of a, I guess, growth hacking story. Yeah. But he uses photography. He runs the Arnold Challenge, the Arnold 1% challenge, which gets people to spend 1% of their day on fitness with 15 minutes of their day. And every single month we have a new exercise, ARNOLD or something from Arnold’s team or some fitness pro puts out, and people follow that challenge. If you do that exercise 15 times in a month, you get a badge. If you get three of those badges, you get a cooler badge. And it’s a really fun way to get people started on this.
Bronson: Gotcha. That’s cool. So tell us about the growth of photography, because that’s really what our audience is, you know, are concerned about. How big has photography grown over the years?
Dick: Yeah, so right now we’re approaching 1.5 million users or so. For context, we launched our our iOS app last year in March, so most used as a 12. When we launched our iOS app, we were at about 200,000 users and before that we were in closed beta and closed beta got us from 0 to 200000 users. And if you think about the fact that like, you know, you think about a fitness app, it should be mobile, right? Like you enter stuff in at the gym, otherwise you’re going to have to enter it in your computer after the gym. But we didn’t have an iOS app until March 2012, despite launching in November of 2010. So we had to utilize a lot of a lot of growth hacking in order to get to 200,000 users without an app.
Bronson: Yeah. And here in a second. Well, actually, tell me now, what did you do to get from 0 to 200000? Because it’s hard to imagine closed beta, no iPhone app. It’s about fitness and somehow you still have, you know, a fifth of a million users.
Dick: Yeah. So I guess it’s best broken up. I guess the strategies and the tactics that we use are best broken up into different stages. Right. So for context, because this is important, you know, we started by me and my co-founder Brian, and neither of us had any technical ability. We weren’t coders or designers. So the first iteration of Eatocracy was just absolute, absolute back. And so to get from 0 to 1000, 2000 users, we had to be extremely creative, right? And so what we did is we really leveraged our our. Founder starting, right? Because if you look at a lot of fitness products, there’s not a lot of authenticity behind them. So a lot of people trying to make money or who don’t understand the fitness space. Whereas with myself and Brian, you know, we’ve been living, breathing, eating, whatever fitness for the last decade. And so we really honed in on that, that founder story. Eventually, we ended up posting on Reddit about, you know, starting this and starting this, this website that turns fitness into a game. Brian Power said that, you know, both of us were geeks. Brian and I were Redditors. And that really resonated with people. And so people tried out photography, even if it wasn’t the best. It wasn’t even any good. They still use it. They still registered. And we got about 2000 users that way in the span of a week or so. After that, we kind of got to the next phase where it’s like, okay, now we actually these features are. So Brian and I started pointing out features that people on Reddit said they wanted because now we had people to tell us, you know, here’s what we’d like to see on his ocracy. I think that this is what would make me choose the product. We had no analytics or no metrics back then because, you know, we didn’t really have users using the product we had. Users aspire to use the product. And so all of our growth hacking was really qualitative. So lots of customer development. Let’s figure out what people wanted. So we took those features out. And then I guess the biggest key for getting to 200,000 users is that we made the autocracy private beta and we did that for a few reasons. One is we didn’t know how to scale. And so at some point if 100 people were on the site at the same time, the database would get ping like 10,000 times. And so a crash. But then the second, I guess there are multiple reasons. The second reason is because we really believed in community, right. And we wanted to make sure that not only did our technical stuff scale, but our community scale as well, and that we gave a lot of due attention to new users who tried to do it. And then the third reason is like, you know, obviously there’s the marketing aspect behind it. If you if you have a club with a nice velvet rope around and people waiting in line, then people are going to want to line. And so what we did is we gave everyone ten and my kids those ten codes. And I think the way that we did it is pretty important because it led to the most amount of what we call in by trends that would occur. So everyone would get ten invite codes and you’re going to be like EDF, GM. Right. Well, so those are used up. Obviously, no one could join for Toxie using that code. So what people would do is take their invite code back to their home forums, and so they take them back to like the most random forums on the internet, like poker forums, car forums, like even like, you know, forums for, you know, new mothers, for example. And they would say, you know, check out the site that actually I’ve been hit by code. It made them feel really special because they were the first person kind of on the scene. Mm hmm. They post their for their code on forums. All ten would get used up, and once all ten got used up, some just reposts. So you had these, like in by transitions like threads long and Long Island and you can go to like team liquid dot net, for example. And there’s going to be some sites that’s, I believe, a gamer or like a Starcraft forum. If you look there, you can actually see the original posts, something awful, the forum as well. You can see the original photography post and it’s just like people posting their codes. So that got us to about 200,000 users. And you know, we spent a lot of time really making sure that those users had a good experience, as good an experience as we can provide, given that, you know, we didn’t really have a development team.
Bronson: Yeah. So you got from 0 to 2000 through kind of Reddit in the community there. You got from 2000 to 200000 through kind of a closed drop invite, closed beta system.
Bronson: And then 200,000. That’s when you launched the iPhone iOS app, right?
Bronson: So did it just take off on its own until 1.5 at that point? Oh, there’s still things that you had to do to prime the pump and get from 200 to 1.5.
Dick: Yep. So. Most of our our distribution now comes from the App Store in order to, I guess, get a lot of leverage from the App Store. We did a few creative things around our keywords, and I can’t really give too many details, but let’s just say that, you know, if you think about the way Apple does its algorithms, you know, there are probably some keywords that will lead some users to come in more than others. Right? Like if somebody searches for fitness tracker, you want to be the first person on the scene there. So there’s some testing that you can do in between releases, right? Like you can’t AB tests, obviously, because it’s just not possible. But you can run certain tests and say like, you know, let’s say I put and this is just I’m making this up, pulling out of my ass. You could say like let’s say I put Tracker in the title and tracker in the description. Let’s see how that affects rankings. What if I take that tracker from the title but keep it in the description? All right, dragons fall out. What if I keep tracker in the keywords so you can see, like, how to maximize? Yeah. So that led to a significant bump in rankings. And unfortunately, what we realized later is that I’m so geeking out now. Yeah. You know, you’ve got you’ve got multiple types of metrics. You’ve got vanity metrics and you’ve got metrics that actually matter and are a gauge of how large your audience size is. What we found out is that when you do, I guess growth hacks like very, very high r y growth hacks like you were tracking, you get a lot of users who find out about your product and they’re unqualified. So they’re not like organic users. And depending on the keyword, they might expect one thing and get something else. And so. Despite the fact that doing that kind of stuff with keywords, you know, blew up distribution, you saw a decrease in activation and retention because those users were incredibly, you know, unqualified. And, you know, not thinking back, I’m not sure if while on the the vanity side, it looks great. I’m not sure if that actually made us a better product because I’m sure a lot of those users won’t come back. Right. And so might have actually been better to grow slower and focus our efforts on optimizing for engagement and retention, which we eventually did. Hmm. The other thing we did to get to, you know, 1.5 and this is where Arnold really kicks in. We spent a lot of time networking the finishing, right. So. If you go to any fitness probe is trying to build a reputation online and ask them like, what’s a top site on the Internet? Most of them will say photography. And we did this because, number one, we reached out to them. Right. And we just became their buddies. Right. We you know me and, you know, Bryant and and other people from photography and fitness pros our entire that go drinking have a fun time with them so there’s like a lot of the softer side of, like, networking and like there’s no intention there. I like legitimate love all the people in the fitness industry I hang out with. And so what that did is that allowed us to, you know, make friends in the fitness industry. We’ve actually let us let us to Arnold because Arnold is also very connected in the fitness industry. We also figured out how to I guess, how to help people best in the fitness industry and use I guess, use that in order to leverage network effects. So what I mean by that is if you think about content, write content creates network effects with your audience and people who create that content because if you have a large audience, then people who create content are going to want to reach that audience. Those people create content. That content, you know, gets your audience, it spread that grows your audience size. And that larger audience is going to attract more people who are willing to create content. And so we spend a lot of time doing that, like, how can we help these vendors for the best? Who are the best pros in the business? And you know, be my partner are also very, very firm believers that a lot of the fitness industry is broken. So what we tried to do is be, you know, the reputable, reputable brand that hold, you know, mainstream users or, you know, mainstream Internet, Internet for which fitness pros are worth following. And so we very much became, I guess, the gatekeepers of, you know, whose brand we should promote. Yeah. And so, yeah, we, you know, just networked with fitness industry. They started creating content for us. They started telling their followers about the toxicity because we had a good relationship that eventually led to Arnold. And, you know, when it comes to business, you only hear about fitness through a few channels like most mainstream people, right? Go see it on like Oprah, Dr. Oz, both are like bullshit sources they’ll hear about it from or they’ll Google. How do I lose weight? So you’ve got Google and then they’ll hear about it from celebrities like Kim Kardashian on Schwarzenegger, etc.. And so you think about the big, big wins in distribution. It’s one of those, you know, searches just like wages you saturated, like you’re not going to compete with fitness and search unless you’re like a demand media. You’re not going to compete with the mass media and search. And then, you know, getting on Oprah. Dr. Oz, it’s a little bit more difficult. So we very much thought about what celebrities we can we can leverage or who give us this big wins. And Arnold is one of them. Mm hmm.
Bronson: Yeah, that’s great. You know, it seems that one of the themes that I hear that kind of permeates so much of your story is something I’ve been thinking about lately, which is team market fit. You know, we hear so much about product market fit, which is important, but I think there’s a lot to be said for team market fit because you guys seem to have such an authentic story. It’s real. It’s who you are. You actually love the industry you’re serving. And that’s going to allow you, I think, to make so many little decisions, well, that you wouldn’t have if you actually weren’t the right people to serve this industry. Do you guys feel like it’s a big part of your success having that team market fit?
Dick: Yeah, I think I definitely think that that is that is the reason for our success. And I don’t say that like in any sort of egotistical way. I’m actually like very fortunate that I was lucky enough to be like a fat kid growing up and play video games. But I think I think people oftentimes attribute success to product. And product is definitely a pivotal part of success. But I know at least with photography, the reason our product is so good is because we were able to attract rock stars who honestly shouldn’t be working for us, like, for example, our chief product officer. You know, this guy is the guy who redesigned Web M.D to go to the number one number one health site on the internet. He designer hire entire, you know website and take it from like what me and Brian did to like something that is beautiful. Like, we were so very.
Bronson: Beautiful right now. I’ll say that I see a lot of stuff I analyze last year. Design has a good like quality to it. It feels like approachable yet still well-designed.
Dick: Thank you. Yeah. We were embarrassed to show our product before, like when Brian and I would demo because it was, like, blue. It was like, made by two, two dudes. But the guy that we hired as our chief product officer, he was an absolute rock star. And, like, there’s no reason that you should be working for photography except that he’s super passionate about our mission. Right? That’s that’s why he’s doing it. You know, our first CTO, he was originally a chemical engineer in Texas, and he quit his job in Texas to come. Upstate New York to be her CTO without pay. I mean, he did that because of the team, the T market as well. So I think that there’s a lot there that, you know, people usually attribute to products, but a lot of it has to do with the team.
Bronson: Yeah, well, the team builds the products of the team is exactly the product can’t be right and it can never be a solution to a market. So it really starts with people that know what they’re doing and the right people for the job to begin with. And I think that’s, you know, like you said, you are fortunate to be affected because it gave you team market fit as an adult. So that’s awesome. Yeah. Now, let’s talk about gamification a little bit. Give us a kind of a taste of it earlier. And you said something that was kind of surprising. It wasn’t. What I expected is that gamification gets them only so far, but then community kind of picks up from there. So walk me through that kind of trajectory. What what does gamification do? How is it being used? And then when does it stop? And what does the community do to take over?
Dick: Yeah. So I’m going to be really controversial here and say that most gamification, if not almost all gamification, is bullshit. Right?
Bronson: Tell me what you mean.
Dick: So. A lot of people will use gamification thinking that it will create long lasting habit. It won’t be any more than a fad, and they’ll often apply it to products that just aren’t concurrent to gamification. I see some people in health trying to apply it, you know, not fitness, but health. And it’s like the more info you give us, the more you’re fucking like I said that word. You can edit that. I know folders sometimes are the more, the more your levels will go up. And I mean, for some parts, it just doesn’t work because people don’t want to give more information. So that’s and that’s one way that gamification is used poorly. And the second thing is, you know, the assumption that it will actually go long lasting habit. People get very bored very easily. So we’ll talk more about like hardcore numbers in the talks with gamification. But whenever we look at users who retain for a long period of time and have made significant life changes, we ask them why? Like, why did you stay on photography? And they will frequently tell us for the community, not a single person mentioned gamification. So we started looking at the numbers on that and we started looking at activation. We started looking at like basically all of the data that we could get our hands on. It turns out that. Most people sign up to photography for photography because they wanted the trumpet and single game aspect. And then we saw that, you know, if you submitted, we saw a huge correlation between people who submitted a workout on day zero and people who are stuck around for the first month versus people who did submitted work. So there was just like a much higher percentage of people who submitted this. So we decided to say, okay, let’s call that activation. I’m sure most most people who are viewing this like are already kind of familiar with the funnel. So we call that activation. And then we said, okay, well, if that’s activation, why are those people sticking around? The first part was, okay, well, those users are sticking around because it must be correlation and not causation. And it’s like if you already know what you’re doing, probably going to submit a work on day zero and that type of person is going to say it’s actually because they start a business. And so we we assume that for a very long time. And then we started looking more at the numbers, right? And we started looking at the interactions that were happening on a deeper level. And we saw that if you had, I think it was like ten followers by the first week, you were a lot more likely to stay. Then we’re like, okay, well that isn’t necessarily cause it might be causation, it could also be correlation that there’s a specific type of persona who does very well on social networks, and they follow more people to start out and they stick around. So we started building these like mental models and started, you know, hypothesizing what of our hypotheses were was what if, what if it’s not correlation that’s getting people to say what if it’s causation in that if somebody submits a work out, they zero. When they sign up, they get props from the community. And those props prop is like a like on paper and those props lead people to check out who was propping them and follow that person because we noticed that at the tops of people just prop random strangers, which is kind of cool. It’s like, you know, you’re encouraging people to keep walking. And so we came up with a test and we said, What if we take the 50 most active people on photography, most active people in the community, and we got permission for them to auto prop. New users on day zero. Will there be a lift in retention and so on? Got their permission. We got like, you know, maybe half of the 50. And then we wrote a script that would randomize prepping to use to submit on these are we a B test of that? And there was a huge lift in retention. And so, you know, because we, I guess, assumed in the start that it was all correlation, normalization. We missed out a lot of those gains. But just by like thinking about the disparate parts of putting it all together, we realized what this what the overall story was. And that led to a lift.
Bronson: Yeah, no, I love the story you just gave us because it’s a it’s a great example of try and differentiate between correlation and causation, which is hard. We have all this data, we have all these analytics, but we don’t know what’s causing what. A lot of times it’s like, you know, the the book by Crawl Lean Analytics, you know, trying to figure out what’s causing what. And I like what you guys did. You came up with a test to then show you if there’s a causation there and then you realize there is. So now you can build things in the product to make that happen more so. So that’s awesome. Now I’ve heard that you guys have engagement numbers that are second in terms of social networks only to Facebook. Is that.
Dick: Right? That is absolutely right. Our average user, average active user spends 4 hours per month on Twitter. So that’s more than Twitter, Tumblr, Pinterest, like LinkedIn, everything except Facebook. Yeah. And the reason is that, well, there are a few reasons. One is that if somebody makes a huge change in their fitness, they are often addicted. Right. And so, you know, you look at other apps like RunKeeper or and like other utility apps like Endomondo, people only use those apps when they are going for a run, going for a bike ride. On photography, you just explore around like there are people coming back every single day to come back during work or come back at night on their iPad. They’ll no longer work an app. It’s actually they’re spending all day on photography and they’re doing things like posting cat pictures and interacting with other users. I think one person, like recorded something ridiculous like 20 plus now has been what it was like 20 to 30 hours on photography. We were looking at what this person was doing and he was just flirting back and forth with with women on photography. And there’s a lot of that that goes on, right? There’s like a lot of flirtatious ness which, you know, it’s good to be cognizant of because it’s always good to know that the basic human drivers that are keeping people engaged for your product, whether it’s like aspiring to, you know, build wealth or have sex or etc.. And we notice that the people who are sitting there spending a lot of time on stocks and we’re spending a lot of time flirting. But I mean, there are a lot of people who obviously don’t like it but still spend a lot of time. But it definitely has to do with the community aspect. Right. And I think that that is the competitive advantage of of a product like photography because anyone can build a utility and you can always build a better RunKeeper and the community creates, I guess, you know, value that can’t easily be replicated.
Bronson: Yeah. And that’s what I was going to ask is because, you know, when I look on Twitter and you do a search for photography, it seems like a users like Love You at a deep level. They love the product at a deep level. And we’ve covered a lot of things that kind of make up the whole of photography. There’s the team aspect, you and your coworkers that really are building things. There’s the product. It’s well-designed. It works really well right now. Even though the first iteration didn’t, there’s the gamification really seeing workouts as a video game, which is one of the things that interested you early on. And then there’s also this community aspect. Do you feel like the love is kind of all of those things combined, or is it just the community? I mean, what do you think?
Dick: It’s yeah, it’s just the community and also the team. So but one thing that we’ve spent a lot of time doing is actually using the product and being a part of the community and like being transparent and like a photographer. If you follow, like me or Brian or Jared, like we’re real people in photography, right? We’re not like a PR bot. So we spent are real people. I like talk about my day, I like to talk about like my dating life and like share embarrassing stories on there is just kind of funny and I think that people really identify with that, right? Just because we’re authentic. Because I feel like the reason we’re like that is because fitness needs to be relatable. Like you mentioned that, you know, I think you said some that relatability or like.
Bronson: Yeah, just the transparency of the team and all those things.
Dick: Yeah, yeah. Like with our design, it’s like, you know, we’ve got our cute robot, it’s purple and those are all intentional. And because you look at most fitness apps and they’re usually very much so, yeah. And so we just try to be really relatable because we are big believers in our in our mission and we just want to be as helpful as possible. And part of being helpful means like making sure no one is intimidated when they join photography.
Bronson: Yeah. Now that’s great. Now before photography, you actually worked at comScore. What is your role across comScore? What do you do there?
Dick: Yeah, so I was a product manager for the advertising effectiveness team and I spent a lot of time with. Not a B testing, but a lot of analysis around data tests, control things, meta analyzes on on lifts and add. So essentially my product was responsible for measuring whether or not there was a lift in sales or lifting brand or lifting conversion based on whether or not somebody had a saw an impression of an X and come to work and actually this is pretty cool can trace to see whether or not let’s say if Pampers is is advertising online and they say they spend hundreds of thousand millions of dollars on ad spend. ComScore can measure whether or not that spend translates to a lift in stores. And so I spent a lot of time immersed in that data, and that’s where I really figured out the difference between correlation and causation and what makes for good, good test methodology. It was a console.
Bronson: Yeah. How would you recommend people kind of learn that skill set? Because, you know, working there, obviously you’re going to pick it up. Is there any like things they could do to get a taste of it? Is there a book they can read? Is there something they should do to really get their head around data?
Dick: Yeah, I, I learned a lot just by AB testing like little things are actually I would say most of what I learned around, you know, that side of growth hacking, the really analytical side is just by doing. Because if you AB test things right you start to see how people react differently to product and you start to get a sense of like, is this an actual list or is it just a is it just numbers line to make a sample size is big enough, right? And so you start to get like a feeling of whether or not there was a live what kind of sample you need. Like what will cause a lift? You know, what might be just like an illusion. I feel like, at least for me, a lot of that. Is, you know, intuition at the start that you want to test. And I think the only way to get that is by it is by doing.
Bronson: Yeah. Now that’s awesome. And I actually agree is that, you know, you go to the numbers, the numbers are hard science and they’re awesome. But as you kind of hone your skill set, you have like this kind of sixth sense in terms of what needs to be done and what’s correlation and what’s causation. And I think you’re right. The only way to really get to that point is just by doing stuff. Yeah. You put out a little product, try to sell something, AB test, anything you can think of. Like just get around numbers and get around data because until you get your hands on with it, it’s always just someone else’s story. A book about someone else’s.
Bronson: It’s not your data that you can actually, you know, get things out of. So I like your answer a lot because that’s that’s the way I would answer it.
Dick: Yeah. I mean, it’s it’s mental masturbation unless you’re actually doing it right. And, you know, I mean, you kind of mention it. It’s a lot about doing an intuition I think that I would agree with. Like if I were to pick the qualities in a growth actor that would make somebody really good. It’s not about being like a rock star engineer, right? It’s about being somebody who’s, like, extremely empathetic so they can understand users and understand how they’re thinking. You know, someone who’s creative so they can create these mental models and say, like, people are doing this, why? Why doesn’t this align with data that we’re seeing? Oh, I wonder I wonder how you think about this. All right. Maybe maybe the mental model is like this and then somebody who, like, has a basic understanding of of numbers. Right. And, like, not an amazing understanding, because, I mean, you always have like, people who are rock stars in math that can always help you out. But being a good growth hacker is, I think about think a holistic view and having those three qualities.
Bronson: Yeah, no, that’s great. I totally agree. Now, let me ask you this. Tell me about the numbers that you’re tracking right now that mean a lot to you, because in every phase of a startup, different numbers are kind of at the forefront. You know, when you first launch, you have zero users, you’re signing anybody to sign up and you get to zero 2000. But now scientists may not mean as much as something else. So what are you currently kind of thinking about?
Dick: Yeah. So for X, the number one thing that we’re improving is retention and we measure a lot of things that go into it and I’ll tell you a little bit about that. But yeah, retention is the most important thing for us because fitness for our product fitness is heart rate. Like having a fitness app or an app is hard because you have to keep people on the app and you have to keep them going to the gym. And it’s not enough to just grow users. You actually have to keep them. And so we we measure and so we split retention up into three different things. Number one is what does a user do in the first week or first two weeks? So we measure activation. Mm hmm. We measure activation. You know, I said activation is submitting a workout, so we really track that in the funnel leading up to activation. And then we measure retention for, you know, the first five days, week one week to week three. And we we also measure and track a few metrics that we’ve seen correlate highly to retention. One of them is cross-platform usage service user uses photography across mobile and across the web. We’ve seen that that user is a lot more likely to stick around for a while, so we’ll track that as well and try to make that go up. The other thing that we measure with retention is kind of current audience size, right? Like how big is your user base? So if you think about like economics, like macroeconomics, right? People are economists have a lot of different tools and different numbers that they can use to, I guess, look at a country or GDP or like something different. Right. And so we think about that with the technocracy and it’s like different metrics are good for measuring audience sizes in different way, right? So you can look at active users, but let’s start at the very top. Like let’s go to like vanity to like, yeah, let’s say you sign up to sign up for visitors, right? That’s like the most vanity of vanity models. And then you have, you know, let’s say 30 day active, right? That’s not necessarily good at any metric. However, you know, if you have a large spike in users and you’re not retaining any of them, you’re still going to have a high 30 day active number. Right. So you can split that up. So what we did is we said, okay, this is in January this year, we had a deceptively high 30 day active just because so many people sign up. So we’re like, okay, how do we separate that into like true actives? So one of the things we did was say, I have 30 day active only counting users who were on the site for more than 30 days. Right? So 30 days or so they gave us one number, as you know, a decent measure of how active our user base was or how larger user base was. The other thing is we looked at active users in the last 30 days who have activated. And so we said, okay, well, this is activated then, you know, we’ll continue in the 38 and 30 day active. And so those were really useful. But the really, really useful thing that we found is that we said, okay, how do we truly, truly measure how big our user base is right now, like how big or active users are because we wanted one clear number just to optimize against. Yeah. And one of the things I was, I was looking into is do users actually keep using photography if they don’t track anything right? Like it’s a fitness social network. So technically you don’t really have to track. You can just like float around, you can use the social aspect in that track. And so I looked at users who weren’t tracking at all and they were like very, very like users who were coming back and not tracking. There’s this very, very, very small amount of them. And then I started thinking about different analysis. I could run and I said, What if you separated? What you took all the users in the last seven days? But compared to those who track the workout and those who didn’t track workout, what would their half life look like? Right. And just as a refresher, half life, you know, in chemistry, it’s if if a compound deteriorates, I might be butchering this. But if you combine materials, it’s like the amount of time that it takes to get to that.
Dick: And so I was like, okay, for these two different groups, how long will it take for users who track this week to get, you know, to have never used the product again? What about users? We didn’t track? And what we found is that users who did not track half life something because they small it was like four days and it was consistent over any period of time. After actually we measured it for January 2012, it was consistent like summer 2012 was consistent 2013 it was consistent.
Bronson: They stopped using after four days.
Dick: Yeah. If that if you took your user base of non trackers for that week, the half life is four days and then we looked at the trackers and it was something really long was like half life is like 60 days. So in 60 days half of those users will have turned right. And so what we found out is that like. You know, if you’re looking what we found is that the one metric that really tells us about the health of our user base and like the size of it is users who have tracked in the last seven days. Right. And just having that one very distilled, like sophisticated, yet simple metric that you can optimize against is just like really powerful. Because everything you do, if you focus all of your efforts on optimizing for that metric and it’s a really powerful metric like everything else will, will improve.
Bronson: Yeah. Do you recommend startups kind of imitate that to really work at finding the one metric that is not vanity that they know actually mean something? Because when you tell me in the last seven days they tracked something and that’s the size of your actual user base. Intuitively, it makes sense. Like, as said, I’m like, Yeah, obviously that’s the right metric. Like because then any point in time you’re seeing a real snapshot of activity, not of sign ups or past usage. Current usage, exactly. Last week. Here’s what it is. And it’s not 1.5, but it’s something that at least we know what it is.
Dick: That’s exactly right. So I do suggest everyone find out. I mean, you might not find it in the beginning and it might change over time. But I think once once you figure out your other shit, like there’s so many things as we’re about to start about once everything is kind of like coasting along, that’s I think, the right time to jump into analytics. Yeah, because if you do it too early, it’s like I spent all this time just kind of like neck burning over that one number when like, you know, you don’t scale your server is like, what does that number matter? If like and that’s not going.
Bronson: To be so small, the numbers are almost meaningless anyway.
Dick: Yeah. So I mean, it really depends on the timing, but I do think everyone should find the one number that determines the health of their their user base.
Bronson: Yeah, that’s right. Well, this has been an awesome interview. I have one last question for you here. Yeah, it’s kind of a high level question. You can take it any way you want, but what’s the best advice that you can give to any startup that’s watching this and they’re trying to grow? What would you tell them?
Dick: Startup that’s watching this and trying to grow. Hmm. So I would say so we did 500 startups, the accelerator. Right. And Dave McClure was I was like, the default is that you’re going to die. Like if you’re a startup, you’re probably going to die. So knowing that you’re going to die, like, what would you do differently? And it’s like what other people aren’t doing, like make some noise, like go against the grain. Because I think that in the tech scene you got this echo chamber and like lots of people who are following others and it’s like, Oh, should I do this? Well, let me see what this company is doing. And the people who are really growing are those who have like great products and are innovating because the growth happening is nothing if you have a crap product, right? Like you can only optimize for activation so much. And so people realize, Oh, this product sucks. It’s only short term wins. It’s like gamification that short term wins unless you actually make real change. Where was I? Go on this. Oh, right. So for startups trying to grow, just like go against the grain, like make some noise. I know that’s like, abstract, but just for a concrete example, look for a channel that other people are aren’t tapping. Right? Like, I’m just making this example up completely. It’s like if you have a startup with, you know, let’s say localize events, maybe look at something like OkCupid for distribution and see if there’s like any way to like, like leverage that. Right? Just like, do things that other people aren’t thinking about. And I think that that will get people to, to, to really when I guess.
Bronson: Yeah. Well, that’s awesome advice to end on. Dick again, thank you so much for coming on growth out of TV.
Dick: Thanks, man. Appreciate it.