Wed · Aug 10

Employee #1: Tumblr

A conversation with Marc LaFountain, Tumblr’s first employee.

Employee #1 is a series of interviews focused on sharing the often untold stories of early employees at tech companies.

Marc LaFountain was the first employee at Tumblr. He currently works on Uber’s EMEA support tools in Amsterdam.

Discussed: Being a User First and an Employee Second, Cold Emailing David Karp, Customer Support as First Employee, Scaling a Support Team, Building an Office in Richmond, VA, Developer/Support Dynamics, and Working in Startups Overseas.

You can subscribe to The Macro newsletter to receive future conversations.

Craig : You live in Switzerland now, right?

Marc : I'm actually in Amsterdam, though I live in Switzerland. While we were in the U.S. my wife got a job offer in Switzerland and I wanted to support her in that so yeah, that’s how I ended up overseas. Then I got an opportunity to work with Uber in Amsterdam, so that’s where I work. I fly from Geneva to Amsterdam every Monday morning and then I do the reverse every Friday afternoon.

Craig : Oh my god.

Marc : [Laughter] Yeah, it's pretty crazy. I'm gonna get platinum status on KLM in another few months, which is gonna be very exciting.

Craig : [Laughter] Congrats man. So what are you doing at Uber?

Marc : I do tools and system stuff. We've got a worldwide organization that does tech support for riders and drivers and eaters. I'm one of the folks that makes the tools and systems for that support organization–making the help desk run, all of that kind of stuff. We run Europe, Middle East, and Africa out of Amsterdam.

Craig : Very cool. Ok, let’s talk about Tumblr. How did you get involved?

Marc : Very serendipitously, in terms of luck and timing. Tumblr started in 2007 and I was among the earliest users. I probably started using Tumblr within three months of it launching. I was in the first few hundred thousand users. At the time it felt like a community in the truest sense of the word. If you were a really active Tumblr blogger then you kind of felt like you knew everybody, in a way. Not literally everybody, but you know what I mean.

So I felt like a part of the community and at the time it was just two people running it. David Karp and Marco Arment were doing everything. They were designing everything, they were coding everything, they were testing everything, they were doing any tech support, any abuse response, and any PR. They were a two man shop and Tumblr was just growing at just a ridiculous hockey stick kind of pace.

It got to this point where they did all their own support through and it sort of felt like anything you sent was going into a black hole. And I didn't blame them, Tumblr was a free service, being run by two guys, and it was growing like crazy. They couldn't respond to every email. They couldn't even read everything. It was just way too much.

I didn’t know David or Marco and I lived in Richmond, Virginia, but I just sent an email and said, "Hey I'm one of your earlier and more passionate users. I have a tech support background. I love Tumblr. I love support. I think you guys probably need somebody to start working on support and I would love to be that somebody."

I’ll admit I didn't really expect anything. I thought they probably wouldn’t even read the message and then even if they did, they probably already knew so many people in New York that could help. It was a total shot in the dark. David, luckily, read that email and I think appreciated that I made it very clear that I loved the platform. I really spoke to what it meant to me, how innovative it was, how I had met people through it, and just how special it was to me. I think that's what most caught his eye. Not just that I was looking for a job but that I was very much looking for a job at Tumblr. So he emailed me back.

We ended up speaking by phone. Not even something like FaceTime or Skype. Phone. And he made me an offer to be a part-time contractor. I think I did like 10 or 15 hours a week initially doing support for them. I used my own computer in my own home in Richmond. I did it on nights and weekends and kept my day job. It was small enough then, there were a few thousand e-mails that had kind of stacked up but once I got that down, I was able to manage. That got us about six months.

Craig : Ok. I have a couple questions at this point. What were you doing before?

Marc : Before Tumblr I changed jobs like every one to two years. My wife was so tired of it, she just rolled her eyes every time. When I started using Tumblr I was working for a government agency in Virginia that does emergency preparedness, doing PR for them. Then like right when David hired me I worked very briefly for a city government in Richmond. Basically I have a little bit of tech background and a little bit of a PR background. What was your other question?

Craig : What was your Tumblr?

Marc : Oh, my Tumblr is just It's just personal stuff. I've never had a blog that was overly political or edgy or anything like that. I would blog news and geekery that was interesting to me. I would blog about travels I was taking. I would blog pretty photos I had taken, but basically it was just about me as a person. There wasn't really an agenda per se.

Craig : Was there a particular community that you were following on Tumblr?

Marc : I followed a lot of people that I viewed as prominent bloggers or prominent geeks. Essentially because I wanted to kind of learn more about that world and that space. In 2007 it was clear to me that social media and the influence of these people was becoming real, which almost seems passé to say now, but in early 2007 it wasn’t really. I've forgotten when Facebook let non .edu people in, but in 2007, social media actually wasn't that big of a deal yet. It was early days. I think Twitter had just started in '06. Everything that we now think of as social media with the possible exception of Facebook was in its infancy.

Craig : Okay, so to jump back, you make it through the first couple thousand emails and then you're contracting 10 to 15 hours a week. What happens next?

Marc : I think it was December of 2007. I started and the volume just kept going up, so finally, I think it was probably in April of 2008, I was just like, “There's too much to be done in 10 to 15 hours a week. So either we’re gonna have to start auto-responding, not responding, or I'm gonna have to go full-time.” It wasn’t like I could just tack on another five hours. I needed to quit my day job and really like commit to it if I was gonna do it. So we had another couple phone calls about that. At the time I still hadn’t met David or Marco in person.

Craig : [Laughter] That's so good. Richmond really isn’t that far from New York. Anyway, keep going.

Marc : We did a few more phone calls and then they sent me a full-time contract via email. I think I worked for them for like another four to six weeks then finally, I think it was June of '08, I flew to New York and met them in person, which was bizarre.

This was at a weird time where I had more than 15 hours of support work but less than 40+ full-time startup hours. I was sort of wondering how to spend that time so they created a portion of the site to feature Tumblr blogs and Tumblr posts. I think it was /browse at the time. I would spend a lot of time finding new posts to feature, new blog posts to feature–basically trying to help with the engagement aspects of things, which was maybe one of the most stressful things that I ever did there.

I would be browsing around Tumblr and there were some people who would really love the posts that I featured. Then I would see these posts that would be like, "Who the hell is running Tumblr browse? This is so boring." And I’d think to myself, "Their opinion also counts", so I would try to mix it up.

Anyway, I did that kind of stuff for a while. That got us another six months, maybe eight months. Then I hired a second person at Richmond to help with support. Unfortunately by the year mark, we were starting to get more into abuse issues. Once you get big enough you have people doing all sorts of harassment, libel, just all kinds of nasty stuff. So we brought on a second person to help. Not only did we need the bandwidth, we needed 24 hour coverage.

The thing is, once abuse gets reported, if you jump on it right then and you nip it in the bud, it's probably not gonna become that big of a deal. But if you finish for the day at 7 PM and abuse gets reported at 7:15 and you don't take action on it until 9 AM or whenever you start work, it will just fester all night. It'll get re-blogged, it will get screenshotted. Before you know it, you wake up and you're like, “shit.”

Craig : Yeah, that makes sense. Were you also doing technical support? At what point could people customize their Tumblrs?

Marc : That was actually a very early feature. You had the ability to do custom HTML and CSS. When I signed up, maybe three months in, you could do that. People were incorporating FeedBurner and doing all kinds of stuff. It was tricky for me in that I got progressively out of my depth, so people would write in to me about complicated JavaScript issues and I was like, “Well… here’s the thing.”

Craig : [Laughter]

Marc : [Laughter] Yeah, it was an issue. When we released a formal API it got really hard. Marco was our only developer and while he certainly could answer most questions, he didn't have the time. Fortunately password reset issues, not JS stuff, was the number one issue.

So yeah, that's the way it worked at the two person level. I was at Tumblr a little under five years and by the end, we were 30 people in Richmond doing support and safety. It got pretty big. We were supporting, I think, 12 languages by the end. Now, when I say 12 languages, we would literally have one person who could speak a lot of the languages, so it was not like it was 24/7 or even seven days a week support in all the languages. But it became a pretty big operation by the end of my time there.

Craig : Is that office still there in Richmond?

Marc : The office is still there. I think one of the reasons that David was interested in hiring me in Richmond and letting me grow out Richmond is that compared to somewhere like Manhattan or Silicon Valley or San Francisco, Richmond is pretty cost-effective. It's not like outsourcing to a foreign country cost-effective, but if you want to have full-time, benefitted employees in the U.S. doing support and trust and safety work, Richmond is a cost-effective place to get that.

You can pay them a fair, reasonable wage without breaking the bank. You can get some pretty cool office space without breaking the bank. So far they've maintained that office and I think they largely maintained the staffing levels in that office. I think economically it works out pretty well. As Yahoo's future unfolds we’ll see if that continues to be the case, but that seems to be what's going on right now.

Craig : I wonder if it's had a ripple effect. Have other support offices bubbled up in Richmond because Tumblr is there?

Marc : Definitely. One of the guys that I worked with at Tumblr, Jim Coe, has opened The Yeomen and they are a support outsourcing consultancy. I worked for Jim at The Yeomen for a year. And there's a company called Mobelux that does mobile application and website development. They actually did the early version of the Tumblr apps, and they're based in Richmond. I do think some good things kind of came out of Tumblr being there.

Craig : So you said you'd done some tech stuff before you were at Tumblr. Did you guys talk about equity and shares when you signed on?

Marc : Not really. To be clear, David was not only fair to me, but very kind to me from an equity standpoint. But, we didn't talk much about equity at the time. They sent me a sheet of paper, it had some shares on it, I signed it. I basically knew nothing about equity. I knew nothing about investing. I knew nothing about different types of options. I knew nothing about any of it.

The other thing is Tumblr ended up getting acquired for a large sum of money, but it in sounds weird to say this, equity just wasn't even on my radar. The idea that six years later Tumblr was gonna get acquired for major money and that that might matter to me just wasn't even something I was consciously thinking about.

I'm not much of a negotiator when it comes to job stuff. My biggest thing was making sure I would make as much as I was making at my prior job so my wife wouldn’t kill me. That was literally my biggest concern.

Craig : [Laughter] Then you negotiated effectively.

Marc : Yeah. It's not something that we went into terribly in detail. What's funny is I've been at several companies since then, and I know more about equity now, and I think it's wild how differently different companies do it. There are some companies that will tell you, "This is exactly how many shares overall there are in the company. This is how many you'll be getting. This is the percentage it represents. Therefore based on our most recent valuation, this is how much we think it represents."

And then there are others where it just feels like a black box. It's like you know that you're getting X number of shares but what percentage of the company that represents, what the company's valuation is may not even be something that you are aware of. You know what I mean? I don't know that one way is necessarily better or worse but it's just interesting to me how differently companies handle it.

I definitely understood more as my time at Tumblr went on and Tumblr did new rounds of financing. I would read articles about the financing and whatnot. I had a growing awareness of what the company was worth but it still wasn't very concrete for me.

Craig : Yeah, it's tricky even if you know the mechanics because ultimately it's impossible to know how it will play out. So were you there through the acquisition?

Marc : I was actually not there. I left in, I think it was October of 2012, and I think it got acquired in May or June of 2013, or something in that ballpark. I definitely stayed in touch with them and came back to Manhattan once during that period and visited the office. I'm still in touch with a lot of them on Tumblr and other networks, but I was not there for the acquisition.

Craig : I'm interested in how you think about your first employee experience given you weren’t a developer or designer. You’re the first support person we’ve had in this series.

Marc : Yeah, so one thing I would say, I don't know if David and Marco would consider Marco the first employee. I've seen articles where Marco is considered a cofounder and I've seen articles which refer to David as the founder. Marco was definitely present at the creation. He was definitely present at the creation of Tumblr. I definitely was not present at the creation of Tumblr.

So, I feel just really lucky that David read my email, that it connected with him, that he was willing to take a chance on me even though I was remote, and he let me build up the office. I feel very lucky and blessed.

But in general, when you get in early you kind of get to wear more hats. You have a broader role. In fact, you're gonna be actively encouraged to take on as much stuff as possible and get as broad as possible because there's just so much to be done.

That can really grow on you and stretch you personally and professionally. To a degree you can define your own job. There's so much that needs to be done. You can just kind of look around and say, “Well, that needs to be done, or I'm interested in that and that and that, too.” You can just kind of like cobble together a job that you like.

I do think all companies have a lifecycle and get to the point where things are a bit complex, where you do have a more tightly defined job, where you have to stay in your lane more. But I think it's nice to get there in the early days because it is broad and so flexible.

Craig : Were there ever any tensions between you and later employees who were developers or designers?

Marc : Having worked at several startups now I think there's an inherent tension between people who work in support and people who work in the product and engineering side of the house. They want to build beautiful, shiny, new things, you know I mean? They want to sign up more users, they want to increase engagement for additional users. They have things that they're focused on and fixing a bug isn't that. Building tools for support that users don't even see isn’t at the top of their list. So we were constantly like, "Hey, quit building that shiny, new thing. We need this tool or we need this bug fixed." There's an inherent tension there, all the time.

And I've seen the same dynamic at, gosh, three different companies that I've worked with now. I think it's just kind of the nature of the relationship. It is more fun to build things than to fix things. But when I say tensions, it's not like people yelling and screaming, and throwing things, but it just takes some arm twisting to get support stuff handled.

When you think about it, Tumblr is free to use. And support is a cost center not a revenue center. Do I think that providing good support could help you to retain and engage users? Definitely. But so can launching new features. So it was kind of an ongoing discussion.

And to answer the seniority part of the question. I think in most organizations people tend to respect tenure. Given that I was the third one, even an engineer who might have viewed me as fairly non-technical would kind of feel like, "Well you're a non-technical support guy but you're also one of the three people that helped build this up in the early years and I respect that."

Craig : Say I'm a founder and I'm going to hire my first support person, how can I set up a system so that the developers respect the support people and the support people don't feel like they’re always pushing a rock uphill to get stuff done?

Marc : Well, I'm not saying that I have lived this personally, but when I was running support at Tumblr I would go to conferences for support people and the way that Automattic [Wordpress] handles that I thought was really fascinating. I have no idea, by the way, if this is still the case today. I heard this story probably seven or eight years ago, but at the time they had two full-time engineers who did nothing but fix bugs and build support. I heard that one of them was a core engineer on WordPress–one of the people who built it initially.

So, to think that they took it seriously enough that they threw two engineers at it full-time, with one of them having that level of seniority and experience, I think really spoke to how seriously they took support. At the end of the day, resourcing is how you show that something matters.

Really it always depends on the size of your operation, but if you get to the scale of a Tumblr or a WordPress, you probably need at least one good engineer who feels like it's their mission in life to be killing these bugs and building support tools. It may not be perceived as sexy as building the new features that the company is working on but it's important. So I think that you will show the support people that you care if you give them the engineering and design resources to make their jobs easier.

Craig : I think that’s great advice. Ok, so if I’m going to be a first employee, what should I know?

Marc : So I think having passion for the product is really everything. Maybe there are people out there who can be hired guns who can just parachute into any job - whether they actually use the product or not - and do a good job at it. I am definitely not that person. I think having a deep passion for the product is everything.

I put in some brutal hours. I dealt with some significant problems and fires. I don't think I could have gotten through it if I didn't love and use the product as much as I did. So find a product and/or a company that you truly believe in and have a passion for would be the biggest thing.

I would also say that you have to, at some point, set limits. And this isn't just for a startup hire. It's any startup, it's any company. Nobody at Tumblr was gonna turn to me and be like, "Dude, you're working too many hours, chill, take time, sleep." No one ever turned to me and said that, ever. And it's not like everybody else wasn't working as hard or harder as I was. I'm not playing my little violin, but I think being at a successful startup means you literally could work 100 hours a week, indefinitely.

There's just an infinite amount of work that could be done and, at some point, you have to prioritize and just say, "I'm gonna set reasonable limits for myself. I'm gonna focus on what matters most."

Craig : What about vetting founders?

Marc : I look for great personal rapport. David is such a lovely person and I'm not just saying that to be nice. David is really a lovely person and one of the brightest people that I’ve ever met. What I maybe love most about David is that he can make you feel like you alone are responsible for this amazing success even though intellectually you're like, "Well I just do support. I wasn't actually here when it was created, and I'm probably not responsible for the success." Great founders can give you that kind of warm glow that makes you feel like you're special and you're doing amazing work. And it's very, very motivating.

So I think personal rapport with the founder is key. If you are the first hire or even among the first hires, you're going to be working relatively closely with that person. If you don't like them, if you don't respect them, if you don't like their work style, things will go south very, very quickly. And so I think personal connection is super important.

Craig : I’ve been wondering, what are some of the more insane or hilarious support requests you've had to field?

Marc : Oh gosh. There are so many stories. I'm not gonna go into who because I still to this day think it's important to like protect confidentiality but Tumblr got big enough that celebrities would use it and would write in for tech support.

You would be having like a really regular Monday, kind of scrolling through the Help Desk, and be like, "Wow, that name looks really familiar. Oh, wow.” [Laughter]

Richard Branson ended up coming to speak in Richmond at some event and I knew the organizers of it and because of that they asked David to come and speak. And so David and Richard ended up hitting it off and he ended up investing in Tumblr and this is all in Richmond, Virginia. I just remember standing there and watching the two interact thinking, "I am watching David Karp and Richard Branson strike up a friendship in Richmond, Virginia." It's so surreal and weird and so there are a lot of very cool stories like that.

Craig : Man, that's great. Do have any favorite Tumblrs?

Marc : Oh gosh, I mean, obviously, is amazing. I love a lot of the Tumblr people's blogs. Amanda Lynn Ferri is hysterical. She used to work at Bustedtees and Connected Ventures and never fails to crack me up. Katherine Barna who runs PR there, she has I love Katherine's blog. She has such an awesome New York life that I enjoy following. I love following the blogs of all the folks that I used to work with in Richmond because not only do I love following their lives, they post lots of stuff from work so there are times when I almost kind of feel like I'm still there. That's a familiar scene.

I found it was a really weird dichotomy, I tended to either follow the blogs of people that I knew in person or the blogs of people who were really prominent in something that I was interested in. So, there is a venture capitalist named Fred Wilson [Tumblr][Blog]. He's one of the partners at Union Square Venture and was a Tumblr investor. Fred doesn't even probably know who I am, by the way. But I follow Fred to this day because I find him so interesting.

Craig : How do you think about your work at Uber in relation to your work at Tumblr?

Marc : It's in some ways so similar and in some ways so different. Uber has always charged for the service that it provides. Tumblr, when I was there, sold a few premium features but for the most part did not charge for the service that it was providing. And I think the rules and regulations change when you're actually dealing with money. There’s a level of service that you're kind of obligated to provide. And support is a literally 24/7 here at Uber. At Tumblr, we did all the support in Richmond, Virginia and we would do some extended hours type stuff but we didn't do 24/7 support.

Tumblr is interesting in that Tumblr is a community and you've got people in that community who are interacting with each other in a very long-term, ongoing way. Uber is much more transactional. You take a ride with a driver and those two people may never interact again in life, you know what I mean? That is very different and it can be both positive and negative. People in the Tumblr community can get really close, but the people in the Tumblr community can also get a little mean. You had to balance that kind of togetherness and, sometimes, rudeness.

Craig : Do you notice any differences between the European and American startup environments?

Marc : I'm trying to think of the best way to answer that question. I've spent a little time in Berlin. I've visited SoundCloud and whatnot. I definitely think that there's a big and growing startup scene in Europe.

I do think that Europeans can bring a different mindset to the table. Something that I have found so fascinating while living here is that in a lot of Europe, nothing is open Sunday and nothing is open late at night. Nothing is open on holidays and the 24/7 nature of life in the U.S., the life that startups in the U.S. kind of exacerbate, just isn't here as much.

There really are people who take their six weeks of vacation. Even at Uber, there are people who take their six weeks of vacation, which I think is a good thing, by the way. Like, that's the European standard and I think in a lot of the countries they even require it. I think work-life balance may be a bit healthier here, but then again, certain attitudes are also just part of European culture.

Craig : Do you have any closing words? Anything we haven't talked about?

Marc : I guess the only thing that I would say for anybody who ends up reading this, is that I wanna make sure that I come across with the proper degree of humility here. I feel tremendously lucky. My email to David was a shot in the dark that he did not have to read or respond to. And then I worked hard at Tumblr. I think I did a pretty good job at Tumblr, but I don't doubt that there are many, many other people who could have done as good or better at the job. Life is really about trying to make the most of the luck that you have. And I think startups are that way, too.

Thanks to Matt Hackett for introducing us to Marc.

Sign up for weekly recaps of The Macro.