How to get a business off the ground. Literally.

Let me paint a picture for you. 🌌 It was January of 2014 — and I was running an 8-person company with absolutely zero traction.


Just a few months prior, we had just started Breather — it was just me and a few other guys and girls — and I literally had no idea what I was doing.

We had just raised $1.5 million, and I had hired a small, tight-knit group: a designer, an iOS developer, a senior engineer, an NYC operator, and a few others. I could feel all of their eyes on me as we waited for precious “traction,” “users,” and “product/market fit” to appear.

There was just one problem — I had no idea how to create any of these things. I had never even run a startup. Hell, I didn’t even know if my service would work.

What a difference this is from today.

It took two more years, but now, Breather is the most funded of all of the “networks of office space” startups — those seeking to do for commercial space what Airbnb or Uber did for homes and cars. Since then, we’ve raised over $30 million. We have amazing, supportive investors, we are growing rapidly, and we’re building our network and user base more rapidly each month than last. And of course, the market is huge. Should we execute, an impressive outcome is pretty much a certainty. 🌠

But back then, I knew none of the above. In December of 2013, I had literally zero actual customers and a dwindling bank account.

Turning it around literally made my whole company possible. And this is the story of how that happened.


Step 1. 🌡🌡🌡🌡🌡

It all started with the focus that can only come from desperation.

Already, a big part of our runway was gone. Breather barely had any spaces in our network — maybe 3 rooms, total. No one was using them. We didn’t even have the locks we needed, after our lock partners failed to materialize. 🗝

In the fall, our CTO finally found locks we could use, with an API we could plug into. For the first time, our rooms could now be opened with the app we had built. It felt like we were off to the races.

But by December, we still had nothing. I had asked (begged) friends on Facebook to try the service, but they almost never returned. A few of them had given me money for a few hours of office space, but I was pretty sure that, if I hadn’t pressed them for a “donation,” I never would have gotten anywhere.

I remember sitting in a bar, on December 26th over a beer, with another CEO giving me advice. “Just pick a number and grow it 8% a week,” he said.

“Really? But then how do I actually grow the number?” I asked.

“You’ll figure it out.”🍻

I came back from the holidays on Jan 2, probably half my runway gone by then. I put a number on the wall—hours booked per week. I told my team we were going to grow it. 8%, every week. “How?” They asked.

“I don’t know. But we’re going to figure it out.”

Step 2. 😅😅😅😅😅

I was trying everything. Asking more friends. Giving it to them for free. Telling people to meet me themselves, in my own spaces… but having them book it. Whatever it took! I was going to get 8% more hours per week.

And it was working! But I was running out of ideas. Each idea was both exhausting, and losing its value — fast.

Around this time, we launched New York — a market we knew that we had to win, if we were going to be anything at all. This aggressive launch turned out to be prescient; it was a milestone that later helped us raise our Series A.

But of course, that didn’t matter until we actually got traction.

Step 3. 🔑🔑🔑🔑🔑

It first started to turn right as my sense of desperation deepened. I started offering free hours on Twitter — to literally anyone. I didn’t care if I was desperate. I was going to get 8% hours a week, and because of my prior successful weeks, it was getting harder and harder to succeed at doing this. I hadn’t failed once yet, and I wasn’t going to start now!

Miraculously, the idea of actually giving hours away on Twitter actually started really working. People were like, “Wow! What’s this new service?” I gave them directions to download the app, which they did. I gave them free hours. Then, some of them booked. All of these people were strangers, and even crazier, some of them actually liked it! Some even came back!

At this point, with evidence of people liking our service and the method working, I started hitting this angle hard. The flowchart you see to the left is what I built. I started from scratch, but over time, it became a machine.

Step 4. 💰💰💰💰💰

Soon, it began to hit the next level. My original Twitter audience was ~40,000 people (I was in the first 10,000 Twitter users, so my follower count started growing early). But I was running out of audience. So I took the next step: I made an ad out of my offer using Twitter’s early ad tools.

Here is one of the ads I wrote (they were all basically the same idea):

Do you see above how many retweets and favourites this ad had? Well, it got so successful that I got a call from Twitter HQ.

Twitter: “What are you doing with your ad? We’ve never seen anything like this before.”

Me: “What do you mean? Just trying to get people’s attention.”

Twitter: “It doesn’t look like an ad at all — and you’re asking for replies? Why? Most people want retweets or follows, don’t you want followers?”

Me: “Nope. I want replies. I want to start a conversation.”

I just looked back a moment ago. During those months, in my inbox, there is evidence of over 1,000 conversations that came from this method alone.

Yup, 1000.


Well, it turns out that this trick got me all the way to a $6M Series A… and beyond. It worked in New York, SF, Canada, and more. Eventually, we found better tactics, and we evolved our team so it had actual marketing people in it (we’re hiring btw), not just some schmuck that had written a few marketing books. 😜

I remember standing in the RRE partner meeting, where Schlaf led our Series A. Tom Loverro (now at IVP) was questioning this tactic. Naively, I told him this would work forever. It didn’t, of course. But it did work long enough for it to matter.

In case you’re interested, here is the graph that the tactic above created (from our Series A deck). It kind of still blows my mind that this is what got me here.

Looking back, I was very, very lucky.

  1. I had a service people actually liked (i.e. product market fit).

  2. There was nobody else really competing in this space.

  3. We had a lot of money from the get go, so we could try crazy shit.

  4. We had a great team of people that made sure everything worked along the way.

I don’t know if I can think of any particular lessons here, other than the usual cliches:

  • Starting up is like building a plane while falling off a cliff. You never know ahead of time what’s going to work.

  • Some experiments will be intelligent. Others will be dumb. It doesn’t actually matter. You just need a lot of experiments and tons of dedication.

  • Creating a hockey stick (of basically any kind that isn’t bullshit) is what’s really needed to raise a Series A.

And finally, I hope you know — it doesn’t actually stop there. Every week at Breather, even now, is filled with experiments like this. And we’re never going to stop. It’s one reason what we’re doing is working.

Anyway, thanks for reading. I hope you learned something, or were inspired to try some stuff at your own startup. 😁