Grouper: How To Build An Online Business When Your Users Are Offline – with Michael Waxman

How do you build an online business when 95% of your users experience happens offline?

Michael Waxman is the founder of Grouper, a social club that sets up drinks between 2 groups of friends.

So you and two of your buddies go to Grouper, if you’re accepted, they’ll introduce you to three other people, and you can all go out for drinks and get to know each other.

Watch the FULL program


About Michael Waxman

Michael Waxman is the founder of Grouper, a social club that sets up drinks between two groups of friends.

Raw transcript


Mixergy’s audio transcription is done by Speechpad

Andrew: Hey, there, freedom fighters. I am Andrew Warner. I’m the founder of Mixergy.com, home of the ambitious upstart. And before I even introduce today’s guest, I want to introduce you to a revolution in online advertising. I am right here revolutionizing, one person at a time. This is the new ad model right here. Forget about banner ads. Forget about text ads. Mug-based advertising, I think, is the future. You tell me what you think.

This is for my sponsor, Scott Edward Walker of Walker Corporate Law. He is the start-up’s lawyer. I’m going to hold this, and we’ll see how effective holding a mug with a logo is. So instead of doing an ad, I will just tell you right now, this is a sponsored mug. And onto today’s guest, who’s smiling there. In this interview I want to. . .excuse me. In this. . .let me take a sip. All right.

All kidding aside here, in this interview, I want to find out how to build an online business when 95 percent of user experience happens offline. That’s the way today’s guest’s business works. He is Michael Waxman. He is the founder of Grouper, a social club that sets up drinks between two groups of friends. So you and two of your buddies go to Grouper. If you’re accepted, they’ll introduce you to three other people, and you can all go out for drinks and get to know each other. Welcome, Michael.

Michael: Thanks for having me, Andrew. It’s awesome to be here. I watched your videos before starting Grouper, so it’s nice for it to come full circle and to be chatting today.

Andrew: Well, thank you. It means a lot to me that you were and that I’m reaching people like you who are actually building real businesses and using my interviews to learn from others. I want to get to know you before we get to know your business, and there’s so much that I want to ask you about your business. But I’m curious, were you social growing up? Were you popular?

Michael: That’s a pretty hilarious question, and I guess the short answer is yes, as much as I can say that with a straight face. I think that the reason why I started Grouper was, I think, for more of a general problem that a lot of people have started to increasingly face regardless of how extroverted you are, or how social you are, which is that our modern digital technology can be really isolating.

Andrew: I’m sorry, but before we even get into that, talk to me about you and what you were like before this business, before you were an entrepreneur. Where did you grow up?

Michael: Outside of Cleveland, Ohio, in the Midwest.

Andrew: Outside of Cleveland, Ohio. When you were in school, were there more than 20 kids in your class?

Michael: Yeah, I went to a small school. There were about 100 kids in my grade.

Andrew: A hundred kids in your whole grade? And of the 100 kids, were you the guy that was going out to parties, who was dating, who was introducing people to each other, or were you shyer?

Michael: I think for the most part, I wouldn’t say I’m and extreme extrovert, but I played high school sports and ran with that crowd and went to the parties and stuff like that.

Andrew: High school sports. What sport?

Michael: Football.

Andrew: Okay. And when you went to the parties, did you know how to talk to people? Because God knows, when I was growing up, I didn’t know what to say at a party.

Michael: Yeah. I think that those are the moments that really make us human, interacting in the offline world and having a face-to-face conversation. I think that growing up. . .I’m in my mid-20s, and so when I started to date and talk to girls or whatever, it was still before cell phones [??] muster the confidence to call on a land line and have the dad pick up. . .

Andrew: That was awful.

Michael: . . .and sort of explaining to the dad why you were talking to his daughter. And then meet each other offline. And of course there was AOL Instant Messenger and that stuff, but most of it happened face to face [??] . . .the skills of how to talk to people and navigate social situations like that.

Andrew: What about computers and businesses? Did you used to start businesses back when you were in school or program?

Michael: So I didn’t. . .start businesses, yes. I don’t remember. . .I think I was 11 or 12 or something when we hatched the idea to start our own ISP in Cleveland back during the dot com boom. That didn’t get particularly far. But I definitely did sort of small things like that growing up. And I didn’t. . .I always like math and science, but I never knew how to code until after my freshman year of college. I dropped out, [??] my first start- up. I moved to Silicon Valley.

It was a language-learning product. I raised money and worked with a lot of smart, talented people. And after about two years we went out of business. I went back to school with my tail between my legs. But of the many big lessons I took from that experience, maybe the biggest was that if I wanted to make a dent in the tech world, that I needed to know how to code. Because you know, you can’t be the editor of a newspaper if you don’t know how to read or write.

And I just needed that basically literacy, and so I got back to school and took a few CS classes but mainly started teaching myself as an extracurricular. And by the time I graduated, I was first of all probably spending 40 hours or more a week just writing code and working on side projects and doing all sorts of stuff, and had the skills of a full-fledged developer. By the time I left school, I kind of had that skill.

Andrew: That start-up that you launched when you quit school, it was called. . .am I pronouncing it right? Batiks [SP]?

Michael: Batik [SP]. Yeah. [??] the end, but that is the correct pronunciation of that short-lived [??] in start-up history.

Andrew: I read about it in Yale Daily News. The idea was that you were going to match people overseas with college students and help them learn from each other, specifically I guess it was learn language. Right?

Michael: Yeah, exactly. And the insight was after my freshman year, I ended up in Taiwan that summer studying. . .[??] teaching English. And I had no credentials, no background whatsoever other than that I was and English- speaking American college student, and I was being paid ten times more than my college job during the school year to teach this thing that I had no credentials in.

I was like, “Wow. I wonder if there’s a business opportunity here to. . .maybe we can pay people two three hours at a campus job, enough to leave.” They can do it over video conferencing, and people trying to learn English, particularly in east Asia can get a native speaker to teach them. And I think it’s a hug market. I still think it’s a big market where the leading players don’t offer very high quality products. I think the big issue we ran into was building this two-sided marketplace. We were really good at getting college students on board, [??] them.

In terms of convincing east Asian students, and particularly their parents, to buy a product was something that we really fell short at. Because none of the founding team had learned English as a second language, and we just lacked that sort of basic empathy with our users, because we weren’t users of our own product. And that just ultimately proved insurmountable.

Andrew: You couldn’t talk to your customers and learn that? You had to be them?

Michael: I think it’s one of those great questions of sort of founder- product fit. Can you do a really good, credible job? I think the answer is it depends on the situation, and for us, I think one of our big mistakes was that we started building the product almost immediately. And it sort of got feedback as we went along, versus, in retrospect. . .Korea was the first market we launched in, and ultimately the only one.

In retrospect, I wish before we had written a single line of code or even made a single wire frame, we would have just gone to Korea and spent weeks or even months talking to students and their parents about what exactly they were looking for and what were really the pain points in the current solutions and based on that feedback built a product. But we sort of had this thought we kind of knew everything and knew the product they wanted. And really, I also sort oversimplify and almost giving us too much credit.

The big problem we made every start-up mistake in the book. And you have to remember, this was 2006, and so there wasn’t a lot of this great start-up infrastructure that there is today, whether it’s stuff like Mixergy or Hacker News or even Twitter, [??] Stack Overflow or even [??] just getting going. And without that infrastructure, we just made rookie mistake after rookie mistake. And the overarching one was we were basically the opposite of a lean start-up.

Andrew: The opposite of a lean start-up. You just built and then showed it to customers, as opposed to learning.

Michael: Exactly. We were an obese start-up, or however you describe it, where it was a sort of waterfall, we ramped up this big team, we raised money to build a big product that really had never, you know, really … [SS] …

Andrew: Let me break this down, though. I want to really understand. Because you didn’t talk to your customers, what was it about the product that they didn’t like or didn’t understand?

Michael: So, one thing was around the difference between learning English in person versus video conference. We basically knew that regardless of what people told us or what we felt, that that was a big fundamental risk of the business, and that we felt that long-term video conferencing made a lot of sense as a way to connect language speakers and for this particular purpose. But we weren’t quite sure if the technology had sort of caught up to that promise, and more importantly if the perception of it had also been there.

People just felt like even if we could show them a demo that worked really well, there was still this skepticism of look, show me the data that says that my kids can learn as well based on this little screen as they will with someone in the room. Ultimately, for a lot of parents that meant preferring a room with hundreds of students in it and an English teacher who is a non-native speaker versus a trained [??] college student one-on- one, but over video conferencing … [SS] …

Andrew: Just because of the medium, they didn’t trust it.

Vivian: Yeah, and I think if we would have approached it more proactively, I think we may have been able to make it that. You know, maybe there are things we could have done. Brought more of the college students over to Korea and have the parents be able to meet them or something, and sort of show that there were real people behind them. But even with that, we may have just been too [??] with it, and that’s one of the mistakes.

Andrew: You said all the rookie mistakes in the book. How about another one? What’s another mistake looking back that you can learn from and we can learn from?

Vivian: Yeah. So, another mistake was we had a very high-risk, high-reward approach to hiring where, rather than a [??] no-fault [??] approach that I think you find at really good companies and really good start-ups where you sort of recognize that the pain of saying yes to the wrong candidate severely outweighs the pain of saying no to the right one who you potentially had some doubts about.

So I think we just basically had a very fast and loose interview process, and just weren’t really rigorous about, again, if there was any doubt or there was no doubt this person should be on the team. Among all those other problems we were dealing with of trying to find product market fit after we’d ramped up this big team, we also had I think disproportionately more interpersonal headaches, because we brought some people on the team who were a really bad fit and it was really painful.

Andrew: Why did you decide so quickly?

Vivian: I think we treated urgency in the wrong way. I think in a start- up, you always want to be moving quick and breaking things, and you know.

Andrew: I see.

Michael: You do kind of have this metaphor of you are this little sinking raft and you’re trying to figure it out before you’re profitable and you just sink into the abyss. So, there’s this urgency, but I think what I’ve learned since is there are some things you want to be really fast about and be 80/20 about, and then there are a few things that you don’t want to be 80/20 about.

Hiring is one of those things. Fundraising is probably one of those things where you don’t want to take an investor who you have some questions about because you just want to get money in the door [??]. So, I think it’s easy to misplace urgency at start-up when you do really need to move quickly, because you will run out of money if you don’t. You just have to be careful about the really important things.

Andrew: This article at Yale Daily News ends with, “Waxman said ‘Returning to Yale was bittersweet,’ as he had to hand over his company, almost like he was giving up a child for adoption. He said he had always planned to return to Yale, and even though he is currently in the class of 2011, he will accelerate and graduate with the class of 2010. As for his future with the company, after graduation, Waxman is undecided.” I’ll cross that bridge when I get to it. Was there any doubt at that point that you were done because it didn’t work?

Michael: Sorry, done with Boutique or with . . .

Interviewer: Yeah, with Boutique. It seemed like you’re telling them, I might get back, but you knew internally, this is not working, this business. I need to step away gracefully.

Michael: Yeah, at that point the writing was on the wall, and we were really just in a tailspin that was totally unrecoverable. What I really kind of owe that to a little bit is the friends I’d . . . [recording cuts out] . . . freshman year were going to be graduating. And one of our angel investors and mentor who actually stepped in as a sort of an interim CEO role to try to see if he could salvage things, really took me aside and, as almost a father figure, said, look, Michael, if you were my kid, I think that this business is very unlikely to succeed. It’s pretty much [??] have a soft landing or [??] we can do.

And I think you’re missing out a lot at college and you should go back. I’m really glad that that happened because it really was just within . . . Well, initially my plan was that if we could figure stuff out, I would still continue to do some work remotely, between classes. And over the summer, I’d sort of go back as almost like a summer internship and work on some product stuff or something. But [??] actually finals of that year, which was my first finals back in two years, was, of course, the time when we officially went under and liquidated the company, and [??] had to.

I think one of the hardest things that I thought to go through, which two years earlier, was recruiting people, some of whom had families and were moving from other states, and joining this mission and hop on board this great cause. And there’s a day when you have to fire everyone. And I think . . .

Andrew: Do you remember that day?

Michael: The [??] was I actually wasn’t as present because I was taking my finals. I think the [??] being remote, which I regret, certainly, just the emotional [??] was very difficult to bear because these are . . . While I wasn’t there to let everyone go, I pretty much had interviewed every person we brought on and in many ways recruited them, sold them on the vision, sort of drew them a picture of how we were going to be successful, and just couldn’t admit that ultimate defeat of a start-up, that we’re shutting down, it’s not going to work. Hopefully you learn some things, but even if it was in the camp of mostly what not to do, you’re kind of asked to start from scratch.

We had such a high burn rate. We were in such a tailspin that there was no even remote possibility of a soft landing or a [??]. It was just sort of burn straight down to the ground. It was painful, but start-ups are hard, and I think one of the hardest parts and one of the most important parts is getting the people [??] right, and hiring the right people and doing right by them, and those lessons definitely stick with me.

Andrew: What about this. Your dad was on the board of directors. He was involved in the business. How did it feel to tell him it’s not working or to have him know that it’s not working?

Michael: On my dad’s side, I come from four generations of entrepreneurs who have basically started their own businesses to varying degrees of success. I think one of the biggest lessons there was just while I have a ton of respect for my dad, and obviously there’s a special relationship there, that, sort of mixing, blurring those lines and becoming sort of a business partner with someone who has obviously totally your best interests in mind. On the other hand, just a lot of tangled up stuff with the relationship that started when you’re coming out of the rooms.

I think in your group of friends since.

Andrew: By the way, about this dad thing, Barry [??] told me the same thing happened to him and his dad at one library. Most people just know them as having this great, loving relationship, which they do, but he admitted on stage, I saw him once when I interviewed him that there was a lot of friction, father son friction. And that made it harder for them to work together and as a result I think they had to go and do different things.

You’re saying something similar.

Michael: That’s exactly what I found, which is funny because now thinking about sort of you know, going out Grouper’s upper management and whether it’s for directors or sort of [??] hires. I think one of the most important trade [??], especially from my perspective as seen, there needs to be someone who’s advice I really respect and will take. And I think just that added friction of the father-son dynamic, of like the same person who is, you know, telling me why I shouldn’t stay out past my curfew and why I should insert childhood-adolescent tenure resistant to, sometimes I think I threw away good advice because he was my dad and I think as just trying to run a business responsibly, it just really isn’t a [??].

For instance, he obviously continues to be, you know, a huge supporter or mine and informal adviser and mentor but he doesn’t have any formal ties to Grouper and I think it’s only for the best that way. But I also feel an incredible sense of gratitude and appreciation for him to spend the time that he really did and the critique. But I do think it’s really hard to make that work.

Andrew: So obviously the story has a happy ending. That’s one of the reasons why we’re here doing this interview. But after that, it seems like you’re just prolific. I was researching you and I came across, I mean, right on your home page I could see a bunch of your projects.

There’s Grouper, there’s Biography, there’s Comments, there’s PaperBuff, there’s Q&A, cora for your website, which by the way has such a cool intro on the landing page. I dare you to click it. I clicked the button and [??] the process.

Was that you trying to find what that big idea would be or was there something else going on?

Michael: That’s exactly what it is. I think, you know, one of the things that that [??] thought me was basically I graduated school, was an engineer, and definitely was so incredibly anxious to apply the lessons of Query to my next start up. I was like, look, you know, I was so young and immature, I didn’t know how to code, I learned all of these hard lessons but, like, now I’m just ready to crush it and I just like can’t wait to sink my teeth into the next thing. But there were things I found, it really [??] occurred to me what the next thing was.

And I think that in my view and in my own experience, I think it’s really hard, if not impossible to start a business just for the sake of starting a business. I think you really need to have a genuine credible passion for solving a problem and, you know, it’s sort of one of those things where people ask me how you know. It’s one of those classic. You’ll know when there is that feeling and for me with Grouper was trying all these little ideas.

Kind of like a funny story too of the Genesys Group that I saw on their project. [?] was this news reader, haven’t really talked about this with anyone but it’s kind of a funny story. So I pulled this news reader and basically had a little experiment, it was like a week-long project, going to start sharing it with friends and sort of like trying to be better and better like Reddit. I don’t really know what it was, it was just sort of the direction I was working on and decided to, noticed that one of the requests for [??] was…

Michael: [??] something that disrupted journalism, and so I was like, “You know what? On a whim, I’ll just kind of grab two random people and throw them on the application and send it in.” The worst case is nothing happens, and the best case is we get an interview or whatever. And so of course, after six weeks I haven’t touched Paper Buff or thought about during that whole period, because I had moved on to some of these other projects.

Of course, we got the email, “Oh, you guys are invited for an interview.” And it was like a joke. One of the people I put on the application was a friend of a friend and I had never even talked to him before. Right? And his name was on the application. The order of stuff to do to prepare for an interview, well, Michael, you should probably meet Amir [SP], because you guys have never even talked before. Meanwhile, as I later learned, there’s a good reason why [??] values the chemistry of the founding team so much. They had talked to thousands of start-ups and they called our bullshit immediately. They were like, “You guys clearly haven’t worked together. What are you doing?”

There are some funny anecdotes about that whole thing. I remember we had a pre-interview at the [??] and it was the first time that we chatted. I hadn’t thought about what our business model was, or anything. He starts going through these questions they’re going to ask in the interview, and he goes “Just so you guys know, we’re going to ask you these things like ‘what’s your business model?'”. I just didn’t have an answer for any of it. There’s this sort of silence on the other end of the phone. Lexus just goes, “You know… I don’t think that there’s really a business here.

In fairness, if Reddit applied to [??] today I don’t think they would get accepted. I think you can’t really get away with that anymore. One nice thing I can say is I really like your logo, it’s really cute. Did you draw that?” I was like, “No, we bought it on iStockphoto.” That’s like the one thing I didn’t even do. Then we basically got out to [??] and a day before the interview, I decided that our best chance was if I just wrote another app from scratch that would have maybe a modicum of a business model.

And I built Qomments in a Redbull fueled 36 hour hack session by myself, just plowed through it, finished it at 5:30 AM or so the morning of our 9:30 AM interview. [??] Ultimately they said no because Qomments also didn’t have a business model and they noticed, very acutely, the lack of chemistry of the founding team.

Andrew: Unreal. How did they know that there was no cohesion, that you guys didn’t know each other well enough? What gave it away?

Michael: Look, I think they have an incredibly acute sense of founder chemistry having seen it so many times. But you didn’t need to be a behavioral expert to notice that we were not on the same page.

Andrew: [laughs]

Michael: I hadn’t met this guy until days before the interview, one of the two people. And the other person, I hadn’t really worked with at all. I built the whole thing by myself and just filled their names on the application. We were totally not on the same page and it was evident to anybody.

Andrew: Alright. So you go home, back to New York. Is this the point when you broke up with your girlfriend?

Michael: So… [laughs] It’s funny, my girlfriend and I were actually still together at this point. It’s crazy, too, because only about five months elapsed between this [??] interview and my next [??] interview where I ended up going by myself to Mountain View with Grouper and we had already made, I don’t know… 30 or 40 thousand dollars of revenue at that point and had some traction. I wouldn’t say the story has quite a happy ending because I don’t think it has an ending yet. We’re just getting started and we’re 1% of the way finished with our mission at Grouper.

But one thing I am pretty proud of is …had 5 months. Really kind of a long wait at being rejected from YC, what are we doing?, what am I doing with my life? Unrelated, well maybe related frankly, a month and a half later my girlfriend brake up and it really is like sort of like all right, what am I going to do? I’m still itching to. And then it turns that the idea notebook, the really broad strokes for Grouper kind of kept circulating and [??] after YC I kind of ran it by my then girlfriend and she just like hated the idea.

And in retrospect, it was probably more because we were together and it was sort of this dating-esque app and she didn’t want me researching the competition. But then, I was living with her in Connecticut and basically packed up my stuff, drove back to Ohio for a few days so I kind of regrouped and get my life together with my parents. And then what basically, you know, essentially no money, no plan whatsoever, built the prototype for Grouper in week and moved to New York and launched on July 6th, 2011.

And it just started like a social experiment sort of like this is an idea that I think might work and it has a lot of resonance for me personally because now I’m this newly single, post-college kid in the city. I can’t just go to the college bar or the college party where you know everyone and [??]. Now I’m like wow, it’s like real world [??] the Manhattan bar scene, how weird is that? I don’t want to use an online dating site. That’s like weird and creepy. Like, why aren’t there better ways to meet people and to navigate the real world in your city.

So it just started like a social experiment.

Andrew: I heard though that at first it was a landing page that took credit card information, everything else was done manually. So you weren’t building out the full infrastructure. This time you decided I’m going to test the lean approach.

Michael: Exactly. So yes, at this point it was like, you know, I had probably like a handful of principals that in addition to all the sort of more qualitative things and lessons that I wanted to apply from [??]. There were like a few things that I sat down and thought about it like I was going to make sure we did 100% no matter what.

So one of them was this lean approach of like only building the bare minimum, buy the product, and you know, for me it was a lesson that was inspiring for Grouper was I rarely talked to Max [?? who’s one of the co- founders of Aardvark. And his story, when they built aardvark they did a team of Google algorithm experts and engineers and they were trying to build a search engine that connected experts. And they had this room of engineers but they decided to use an Excel spreadsheet to do the search for the first 9 months.

And the reason why is because the need to know what algorithm to build . And that is really the approach that I took to Grouper. First it was like in terms of matching people together and setting up these [?] drinks, I don’t even know what’s going to be hard about that, what’s going to be easy about that. So, you know, let’s like totally do everything by hand and just, you know, learn from what the bottlenecks are and just start to solve bottlenecks one by one with technology. And that’s what happened.

Andrew: But you did have Register with Facebook. So people would click and you’d get their Facebook profile and you did charge. And I heard also, I think this actually comes from the pre-interview, you called a couple of your friends and you said here is my idea, do you want it? They said yes, you said pay me. And they did .

Michael: Yes, and that’s exactly what happened. [??] that short list of principals, like one of them was to be lean, to go [??] and another one was certainly charge customers on day one and it was not to make money, it was not to get [??] although that would soon become a goal. But the real goal was something I was really fortunate to get some desk space at Betaworks for the summer. And what I was working on there never amounted to anything, but I did, throughout the process, pick up a lot of really amazing lessons. I think that Betaworks are some of the most talented, creative, all things considered, under the radar, sort of groups for what they’re doing.

One of the things that struck me there was, they had this lesson of, “You need to charge your first customers.” And the reason is because, what are you looking for out of your first customers? You want feedback. You don’t want their money, you want their feedback. But free customers don’t give you feedback. They just stop showing up to your website. Whereas, paying customers certainly give you an earful of everything’s that’s wrong, everything they hate, every reason they want a refund. But that’s what you want. You want that candid passion.

And so, our group was determined to do the same thing and it worked magically, just as that approach would [??]. You would think, where, literally it was [??]. First of all, [??] something people paid for, at all. [??] pay for it. An then we just got all kinds of crazy, passionate, adamant feedback of what was obnoxious or didn’t work.

Andrew: What didn’t work in the beginning, that you heard from people? It seems like it’s a straight up easy shot. If you’re picking out who they’re going out to drinks with, what’s tough? What didn’t work?

Michael: One way to think about it is, we all know from anecdotal experience of doing stuff with other people, getting six people to do something at the same time and place is just a fundamentally hard problem. Once you get into that [??] space, there are just so many things that can go wrong.

Andrew: For example?

Michael: Scheduling issues: one of them needs to cancel last minute, or one of them doesn’t show up, we’re [??] six people, only five show up, or they can’t find each other. And it scaled. The sorts of problems we had, it scaled. Fortunately, we minimized a lot of it.

This example, too, I shouldn’t call it funny because it’s not. But it’s just absurd. The types of problems [??]. We schedule, and we pick the bar, too. So, we had scheduled a group, we’re at a bar. I think it was in San Francisco on a Tuesday, and they show up to a Wednesday. And that previous night the bar had burned down.

Andrew: Oh.

Michael: It had burned down in a fire. And they literally showed up, and one of the reasons we invest in really [??] customer service, is they literally show up and they’re like, “Hey, we have a problem.” “What is it?” “Well we met each other, but the bar we were going to is burned down.” I think of that as an example of the sorts of crazy, unpredictable things you have to account for.

And lesser versions of those things were happening all the time. It was like, okay, this is the way we need to schedule, this is the way we need to do contingencies . . .

Andrew: How do you deal with that? What do you do as a contingency if the bar is closed, either because of a fire or maybe something less dramatic like the bar owner decided he was going to host a party there because he got a last minute booking?

Michael: [??] That happens, right? The short of it is, we need an on-call team who are manning the phones and text messages. We actually, at this point, boast some pretty sophisticated software that processes all of our incoming communications — [??], emails, calls — and prioritizes it in the queue, routes it to the right person [??] time.

Especially as we started to scale, we would get these things, too, where it’s like, it’d be a Thursday night and we’d have all these Groupers going on and there are also people calling us and emailing us about their Grouper in two weeks, or if they want to join the service. We have things again, where it’s like, people are standing outside the burnt out bar and we’re on the phone with the mom of someone who noticed a charge for Grouper on her son’s credit card.

Andrew: Right.

Michael: How do you prioritize [??? Everything we do comes down to using technology as a means, not an end, and combining it with a human element. So for us, everything we do, whether its [??] customer service, there’s this blend of technology and trying to use it in a really clever way and then humans, because after all that’s what our whole frickin’ thing is about. Right? It’s about being more human and getting off on interacting with people. And so. . .

Andrew: You know. . .I’m sorry. I have so many questions to ask and so little time to get it in. So let me just follow up on something that you said earlier, a small detail, and then I want to go bigger picture. What happens if a group with three people show up, and the other group that’s supposed to have three people only has two. What do you do with that?

Michael: So the short of it is we try to prevent it so that it happens pretty rarely, which I think we do a pretty good job of doing. And the short of it is, because it’s an online-offline business, our peer companies like AirBnB and Uber, etc. have to deal with this as well. And I think our approach, which is similar to those companies, is you have to have sort of really good, no-questions-asked customer service. Right? Stuff’s going to happen. Stuff’s going to go wrong in an online-offline business.

Andrew: So you just tell them how to deal with it, but the event continues?

Michael: Yeah. I mean again, it’s just about being really and customer- focused and calling them up, and again being a real human about it and talking so someone and saying, “Look, you know what, if you guys are uncomfortable because you want three and you want us to cover the bill and help get you guys something else that’s fun that night going on in your city, we’ll totally do that.” If you want to. . .

Andrew: You just make it right, and. . .

Michael: Yeah. Exactly, we can just make it right.

Andrew: Cancellation was an issue.

Michael: Yeah.

Andrew: What was the issue, and how did you solve it? Because I know that was a big success story for you.

Michael: Yeah. So the short of it was that as we grew, as we started to scale in the spring of 2012, our cancellations seemed to spike sort of disproportionately to our growth, and it was ultimately really frustrating, because it was just for a lot of reasons. It was a customer service nightmare, dealing with all these cancellations. It was a revenue nightmare. We had all these groupers that were going generating money, and then they cancelled [??] to make up for it and make it right. So it was a cost center.

And the short of it was we started talking with our users. And this was during Y Combinator, by the way, where appropriately the two things you’re told to do are write code and talk to users. And this was like, “All right. Let’s talk to our users and figure out what’s going on.” Anecdotally we found that there are some category reasons to cancel that are totally valid and that you really can’t do anything about. So sometimes you legitimately. . .one of the three people in your group gets food poisoning, and you have to cancel. Or there’s a death in the family, and you have to cancel. It happens.

I don’t know if it’s one in a hundred or one in fifty. It happens. You just kind of have to suck it up if it happens. But we noticed just in talking to people, asking them why they cancelled, there was this pretty big cohort of people who basically didn’t seem to have a good reason. So it was sort of like, “Yeah, I’m a little tired. I have a big day at work tomorrow.” It’s sort of like you could go. There’s really not a great reason why you’re not going.

Andrew: It’s like last minute jitters.

Michael: Yeah, or just laziness or whatever you want to call it. But basically just we didn’t feel it was a good reason, especially because from our view we realized one of the things that we had done sort of incidentally by trying to create this amazingly seamless, really convenient, really easy experience. Something we’d accidentally done was sort of shielded people from the reality that these were three other people who’d rearranged their plans and probably did their makeup or went out to dinner or did this whole thing.

And so you’re not. . .we don’t really care, because people are going to cancel or whatever. It’s part of our business. We factor it in. But these poor three people that changed their whole plans to hang out with you guys, and you better have a really good reason to cancel them. So we basically implemented this policy where you have to call the other group to apologize.

Andrew: They have to call the other group.

Michael: Exactly. I’m sorry.

Andrew: So you’re basically exchanging phone numbers ahead of time?

Michael: So I mean we wanted to be very careful about it, because you shouldn’t get to have someone’s phone number because you’re cancelling on them and being a jerk. If anything, we should. . .that’s the last thing we should get. So we built a little [??] app where basically it shielded the actual exchanging of numbers. But yeah, but I think just doing that a few times and sort of word spreading in the community [??] explained to people [??] that was the cancellation policy, we saw a really sharp drop in cancellations. Because of that, it was …

Andrew: That’s cool. So, because they had that little bit of social pressure of needing to call the people they were cancelling on, they wouldn’t do it and go on their date.

Michael: Exactly. And whether it’s social pressure or it’s just the realization that you’re not cancelling on some faceless, nameless website. You’re cancelling on like three other people who move their plans around to hang out with you, and don’t apologize to us, apologize to them if you still want to cancel.

Andrew: Michael, I called it a date there for a moment. Isn’t it a date? Some websites I see refer to it as Grouper the dating site. Others say it’s a way of meeting people socially, but if I’m going there it’s a date. Isn’t it?

Michael: No, it’s not. I think really we haven’t done a good job of telling our story in the way that we think about Groupers, and more so than some kind of BS marketing angle, just really genuinely what do we view as the value prop of Groupers, and why you would go onto Groupers or why we exist as a company. The short of it is we think of ourselves as, again, being the technology that saves you from technology and helps bring people together offline.

For us, for instance, when we go through our user flows and building out a product or feature, we think of ourselves much more in terms of just more broadly improving on the user experience of meeting people. Stuff like bars are kind of more where we think rather than something like a dating site or a dating app. I think genuinely when you look at what Grouper is and the product we’ve put together, it is much more similar to a bar than it is to a dating app in the sense that, whether it’s the oldest dating sites in the book or the newest apps, generally there’s sort of this paradigm of profiles and messages.

Our take is that that’s just kind of really weird and dehumanizing, and it’s like shopping for people, and there are all sorts of kind of weird social and behavioral problems with it. Grouper doesn’t have any profiles or any … there are no messages or profiles … [SS] …

Andrew: You guys just pick it for us. It’s not that I’m going on and shopping for a date.

Michael: Exactly.

Andrew: Let me ask you one other thing, speaking of connecting. You went to Y Combinator with Grouper on your own and they paired you up with someone else there, is that how it worked?

Michael: So, what had happened was basically I applied to Y Combinator as a solo founder and had two early employees who were holding down the fort in New York and sort of helping out with some logistics. But I went out there on my own and interviewed by myself and started. The long and short of it is, my now co-founder [??] Tom Brown [SP] had started YC working on a different project with a co-founder.

The short of it was they didn’t see eye to eye in the direction of the start-up, and some ridiculously high percentage of start-ups do [??] they split-up. Basically Tom and I had been good friends, and he had expressed for a while an interest and passion for what I was working on, and we sort of started chatting. One of the great ironies of that experience was, partly because of the urge of YC, but probably because I [??], I really wanted to find a co-founder.

Andrew: You were going on founder dates essentially, trying to …

Michael: So, that was the irony. Where was Grouper for finding a co- founder? I was going on these horrible, awkward, one-on-one dates that were just awful in much the same way that I think one-on-one blind dates [??] awful. It was just a huge time commitment, and you could often tell really early on that there just wasn’t going to be chemistry. People were just obnoxious, too. Grouper was already generating a decent about of revenue and a decent amount of [??].

The cool thing about Tom that was awesome and really the breath of fresh air, I think one of the reasons we ended up working so well together is, everyone else that I talked to would immediately just jump to their own interests and their own needs. Before we even talked about what the hell Grouper was, they would be like, what’s my [??] going to be?

Andrew: Oh, yeah. Yeah.

Michael: What’s my title going to be? How are you going to split up the technical duties, like what stack are we working on. I want to just jump, like, right into it, like, you know, because I want this percent. I don’t even want to work in [??]. And the awesome thing about Tom was he just came in totally from this, like, other-centered point of view where he literally came in. It was, like, hey, Michael, you know it sounds like Grouper is going pretty well.

And, you know, it must be like really hard work trying to, like, do everything for this site and, like, build it and be the CEO and, like, eat everything. And, you know, what can I, like, help you out with? Like make it easier. And, like, that was the presentation. And, like, that literally was, like, wow, yeah. Like, yeah, there’s a lot of stuff you can help out with. There literally was, like, no pretenses.

And we really didn’t [??] our conversation about it up front. It just was, like, let’s kind of work together on a trial basis. He ended up sleeping on an airbed and, you know, is [??] in the apartment with me and this other Solar YC founder. And, you know, sort of like about a month later, like, a little bit before demo day, we’re sort of, like, yeah, I guess we’re co- founders now. And it’s, like, this, you know, I guess should be the deal. And it was, like, yeah, that’s essentially fair.

Andrew: What is the deal? What kind of equity did he get?

Michael: So, you know, I don’t feel it’s my place to discuss that. I think, you know, even if we’re totally transparent about everything except for other people’s comp. I feel it’s not my place. But, yeah, but . . .

Andrew: How about this?

Michael: I think we had a really fair deal.

Andrew: What is your revenue today? How much revenue are you guys pulling in?

Michael: So just to continue the theme of not being able to answer questions. This is another one we don’t disclose. But again, the short of it is . . .

Andrew: Is it over half a million a year?

Michael: I will say that yes, yes, it is.

Andrew: Okay.

Michael: And I feel comfortable giving you that. Yeah, I mean, I think for us, again, the purpose of generating revenue really early on was not to make money. It was to get feedback, but very shortly thereafter it became about being profitable, and being able to get this sort of infinite runway. And, you know, sort of spend little and make enough to cover our costs.

Now we’re, you know, ramping up and, you know, actively trying to make money. Because we want to put a dent in our universe, which is our company mission of ending loneliness and doing that by bringing people together offline and using technology to . . .

Andrew: I can’t believe you would go offline. The whole business of going offline is so messy. Even as an individual, so many things can happen when I go offline. I prefer to stay online sometimes. But as a company . . .

Michael: Yeah . . .

Andrew: . . . you have to deal with all these issues that you’ve said before. Someone doesn’t show up. Someone goes and has a great time, but then at the last minute, a beer gets spilled on her or him, and then they’re upset. What a messy process.

Michael: You know, I think Paul Graham has this essay where he calls it schlep blindness, where there are problems where there’s like a lot of value intrinsic to solving them. But people kind of ignore them because they’re messy. I think this is like a classic example of that . . .

Andrew: Yes.

Michael: . . . where, you know, look, we’re a technology company. And for us, technology is just a tool. It can be used to isolate you, or it could be used to bring people together. And it’s kind of just all about how you use it. For us, it’s really about this overarching approach as, you know, a means, not an end. But, you know, for us, it really is, you know, there’s just tremendous value to create it. Because people still spend 90% of their time and money offline. And so it’s really hard. It’s really, really hard.

I would even say, like, much harder than it looks. Like, our whole problem is, like, try not to make it seem seamless and convenient and easy. Sometimes we’re not even always able to do that. But, like, [??]. Really difficult. But whether it’s from a financial perspective, or from an impact perspective, and improving people’s lives and put in that universe. You know, just meaning and value happens at a greater scale offline and people are face to face. And so, you know, sucking up the difficulties in that. Being able to leverage that, I think, really lets us make just a much greater impact on the world, and build a much bigger business.

Andrew: There’s one more thing that I want to make sure to ask you. Most of the people who I interview seem to grow using SEO or using tell a friend type mechanisms. For you, you actually buy ads. Display ads. Right? I think our video might have just frozen.

Michael: So, yeah, . . .

Andrew: There we go. Yeah.

Michael: The short of it is, yes, we do. We’re currently running some ads as an experiment to see what exactly the ROI is of them.

It has not been the way we’ve grown to date. Our growth has been fueled because we have a naturally viral product. You have to bring two friends along to use it. The word of mouth comes from creating an experience that’s hopefully amazing and sort of naturally shareable. Because you did this crazy thing with six people and you took Instagram photos of it. You came into work hung over the next day and had to explain to your coworkers why. That’s really how we’ve grown to date.

But, again, I think for we have big plans. We really want to put this dent in the universe. As we try to grow the company from this level to the next we think it’s going to probably require a wider assortment of different tactics for different markets or different reasons.

We’re in 25 cities now. Word of mouth is an awesome thing if you can harness it. But, oftentimes it can be ephemeral and hard to pin down. We need to be able to supplement that, whether it’s to get a new market off the ground or correct a gender imbalance in a market that’s already going. We need to see if we can tactically use more predictable strategies like ads, even if we have to pay for them – whereas word of mouth we don’t.

We’re using that as an experiment now, but I wouldn’t call it a core part.

Andrew: It’s not working yet, but you’re experimenting with it. You know that there’s a need for it. Or, it would help if you could get it.

Michael: Yeah, even that’s sort of a question. Because, ideally, if we could solve our solutions with ways to get users that are for free then we would do that. But, currently there are moments and markets and times, and we haven’t cracked that nut yet. It’s about seeing a wider assortment of initiatives, whether it’s ads, events, or PR to supplement that.

Andrew: Here’s another thing that you do for marketing. You ask people to apply. It’s always been apply to be a part of it.

I remember talking to the founder of JackThreads. He said that people aren’t going to come to the site and they’re not appreciate the deals that I have. But, if I have an application process then they’re going to value the product more. And, if they meet me, or if they see an ad somewhere, or if they get a blogger that writes about them to give out a coupon code that skips the line, then they’re really going to want to head over quickly.

It feels to me like the application’s important. It makes sense with your product. But, it’s also good for marketing. True?

Michael: Yeah, absolutely. I think one thing we haven’t done a great job of that we’re working on right now is just being a little bit more clear and consistent around what membership actually entails. I think we have this high level idea from day one that it’s always been group or social club. We’ve always fashioned ourselves as we’re a social club where people can come together offline, have adventures, meet other people, and spend time with their friends.

But, at various times and various A/B tests I feel like we’ve been, like, you’re applying, but what does acceptance look like. Or, when are you approved? Or, what are the privileges of being a member? And, if your friend goes along with you, are they a guest, member, or what?

So, I think in general it’s definitely been a key strategy of ours. We’re working on taking it a step further and being very consistent about what exactly membership is and what exactly are you applying for…

Andrew: …What does it cost?

Michael: Right now it’s $20 a person per grouper.

Andrew: Per week.

Michael: Per grouper.

Andrew: Oh, per group outing.

Michael: Yeah. Right now we’re mostly a pay per use model. You go on a grouper. You pay 20 bucks per person, and that’s that.

Although, we’re working on basically other ways of paying for your grouper. Again, when you think of this membership idea, it probably makes more sense for a social club for there to be membership dues, and being able to go on a certain amount of groupers in a certain time period for a flat fee. And so we’re currently experimenting with some of that, but nothing to fully announce just yet.

Andrew: Alright. Let me do a couple of Mixergy related things here, and then I’m going to end on a high note with something cool that I know happened. Two cool things actually that happened. But first, for a Mixergy related, one of my premium members, his name is Martin, he’s got an educational app that he wants to get to special needs students. And in order to get this thing going, he would like some help getting people to vote.

People from the Mixergy audience, to help support his app, to get some votes, so that he can get some more support in building his app. All he needs is 250 votes. I feel like we could do that in a heartbeat. And so, I hope that you guys will join me in voting and helping him out. What I’m going to do to make it really easy, since his product is called ‘Tommy Teaches’, and his site is Tommy Teaches dot com, but, he created a url for us to explain to the Mixergy audience what it’s about. I’m just going to create Mixergy dot com, slash Tommy Teaches, and I’m going to ask everyone who’s listening to at least check out what this member is building. So go to Mixergy dot com, slash Tommy Teaches, and on there you’ll see what he’s building. It’s an app that he wants to get out there for special needs students. Mixergy dot com slash Tommy Teaches.

And, one more thing. I’ve been holding this mug, and you guys know I’ve been trying to find out a new way to promote my advertisers, something other than those pre-rolls that you probably remember, Michael, from up until this last month. I’m curious, has it worked? Does it make sense? Maybe in the comments, so that everyone can see what you’re thinking, and that I’m not pruning just the good ones, or sorry, pruning, and just keeping the good ones for my sponsor. Maybe in the comments you can say what you think of it. If you don’t mind. If you think this is a good idea, tell me in the comments. If you think it stinks, tell me in the comments. I want an unbiased, open feedback from people. What do you think Michael? Is it distracting to you as an interviewee, to see me hold up this mug the whole time?

Michael: Well, I think it’s a cool experiment honestly, in like integrating advertising more deeply in to experience. I think it’s probably the future of ads, so, I’m all for it…

Andrew: Maybe we’ve revolutionized. There you go. You, me, and Scott Edward Walker maybe have taken one step forward into a whole new way of doing ads. Maybe YouTube will cancel all those pre-rolls, and start sending mugs for its sponsors to all the videographers. Maybe not. But it’s an interesting experiment, and I do think that maybe we’ve got something here. A way of not interrupting the flow of the conversation, except what I’m doing right now. But also making sure that people get to see the sponsor, Walker, his logo right there. Alright.

Thank you Scott, by the way, Scott Edward Walker for letting me experiment like this. You truly are the entrepreneur’s lawyer, if you allow me to experiment and see what works like this. Two things. One, we talked about Alexis earlier going; I don’t know that your idea is good. He heard about Grouper, what did he do?

Michael: Well, so yeah. We haven’t really publicly or with any of our investors, but Alexis [SP] is an investor in Grouper, and we’re really excited to have him.

Andrew: Yeah, and I know he’s excited to be a backer of the business. And the final cool that happened is, sometimes you go to bars and restaurants, and how do they treat you when you’ve been sending all those business to them, all these groups of people?

Michael: Well, so it’s fun, like I think something that we’re most fortunate with is, it’s a really fun business. It’s a fun problem. We help our members have a lot of fun. And one of the perks is that in the hundreds and thousands of bars we’re working with around the country, we can scoot by, and get food and drinks et cetera, and it’s a tough job, but somebody has to do it.

Andrew: It does sound like a really fun job with a lot of meaning. I can see how you’re helping people get together. The website is Join Grouper dot com, Join Grouper dot com. Maybe I should have made it clear at the top of the interview. Join Grouper dot com, where you can see photos that people have taken from their events, and from their groupers. Is that what it’s called?

Michael: Yep. Absolutely.

Andrew: And I just love your design. Can you just tell me who designed this, what’s the name of your designer here? Who are you wearing? What is your website wearing?

Michael: Yeah. Also the design of my t-shirt. Kyle Miller [SP] actually is the name of him, and you can look him up on Dribble [SP]. And Kyle’s a really very talented designer. Kind of nailed that mid-century, modern style. Where it’s sort of a back to the future vibe…

Andrew: Yeah.

Michael: …where like we’re getting back to our roots, but kind of pushing things forward with technology, so we designed something that we take very seriously.

Andrew: Back to the future in the sense that, almost the way letter press is. Old fashioned technology used to create something that feels a lot more elegant and modern than I usually see. Really, your website is just beautiful, way to go Kyle. And your t-shirt. You pointed at it, just because you did, we can’t see it.

Michael: I can, let’s see if you can see it.

Andrew: Ah, there you go. [??] all over that thing.

Michael: Yeah we rip out the tags, and have a custom label printed on the back, and stuff. We take our shirts very seriously, with the attitude that people are either going to wear them or not, it’s kind of a binary thing, so may as well go all out on it and make sure that people give you free advertising.

Andrew: You guys really do have a great sense of style. Check out Join Grouper dot com, if you’re listening to the MP3 and you don’t know what I’m talking about. Thank you so much Michael for doing this interview. Thank you all for being a part of it! Bye!

Sponsors I mentioned

Walker Corporate Law – Scott Edward Walker is the lawyer entrepreneurs turn to when they want to raise money or sell their companies, but if you’re just getting started, his firm will help you launch properly. Watch this video to learn about him.

  • http://www.pixelhappy.com/ Steve Young

    Andrew, can’t find the tommy teaches URL. I’ve tried http://mixergy.com/tommyteaches and http://mixergy.com/TommyTeaches. Would love to show some support.

  • Kyle Patrick McCrary

    A good story to learn from, Michael. Thanks for sharing!

    $1.) Sometimes you have to empathize with your audience. If you haven’t walked in their shoes, you may not be able to solve their problem. If you haven’t yet, go ask what they want.
    $2.) Hiring: “The pain of saying ‘yes’ to the wrong candidate severely outweighs the pain of saying ‘no’ to the right one, who you potentially had some doubts about.”
    $3.) You don’t always want to move quickly. Some things, you want to be particular about.
    $4.) It may not be the best idea to work with the same person who is telling you why you shouldn’t stay out past your curfew.
    $5.) It’s difficult to start a business just for the sake of starting a business. What you’re going to do well is starting a credible business that solves a problem.
    $6.) Be lean. Charge customers from day one. Not for the money, but for their feedback.

  • http://mixergy.com Andrew Warner

    Sorry. I thought I included that in my notes for Arie:

    https://www.missionmainstreetgrants.com/business/detail/75530