Airbnb’s rocky path: Learning from hard failures and building strong company culture to scale rapidly – Part 2

Note: Since this is a long session, we have divided it into two blog posts. Link to the first part here.

A group of entrepreneurs taught a class at Stanford called “Technology-enabled Blitzscaling” and this is Freshteam’s attempt to present the key learnings/takeaways based on the lectures. 

In this class, Reid Hoffman, Executive Chairman and co-founder of LinkedIn, interviews Brian Chesky, co-founder and CEO of peer-to-peer lodging service of Airbnb, on how the idea came about (it’s quite a story!) and how Chesky and his team built a strong company on the bedrock of a solid culture.

The story is written in the first person.

Living the product

Reid: I think you spent nearly a year staying at Airbnb homes. Why did you do that and what did you learn from it?

Brian: We started Airbnb in our 3-bedroom apartment. Unlike most companies who find an office space, we decided to work out of our apartment indefinitely (originally, we were inspired by Craig’s list, which works in a house). We hired people quickly and we couldn’t find an office in time. At one point, we were around 15-17 people working out of that 3-bedroom apartment where we had meetings in the bathroom and stairwell.


By the way, never interview someone in the bathroom. It’s apparently a huge violation.

I decided to give my bedroom to the company to be converted into a meeting room. While figuring out where to live (such as Craig’s list), I thought we have thousands of Airbnb homes in San Francisco and I could just rent one for a couple of months. Then I thought it would be fun to stay in Airbnb homes for a few nights and test them. Weeks turned into months (almost a year) and I would stay in a different home every 3-5 nights.

This sent a huge message to the team that this is not a job, not even a career but a calling. I think a strong culture is when people believe in what you’re doing. The idea is not that you’re financial systems or websites or not designing screens and mock-ups but that you’re building a mission. You’re creating this kind of world. And, that’s only possible if you’re constantly using the product.

I feel like everyone of us has to be a product person and not a business person. We need to be deeply passionate about everything we’re doing with the product. It really told the team that they should use the product while travelling, use it when they’re home, should become a host and become an expert in every part of the product. I see a number of people who work in the big companies are so disconnected from the end customer that their companies get disrupted.

Designing Airbnb’s strong culture

Reid: The design of Airbnb’s culture was one of the things that I see you put more energy into than any young entrepreneur…

Brian: I define culture as a shared way of doing things. That’s it. I don’t think there’s necessarily good and bad cultures. I think there are weak and strong cultures. What I consider to be a bad culture, others might think it’s good. It’s just the way they do things. I wanted to have a strong culture where everyone was on a mission. We had a series of beliefs and a few governing ideas.

We toured Zappos. Alfred Lin, who’s on our board, ran Zappos with Tony Hsieh. We did a lot of research. One thing I learned is a strong culture is a place where the founders really impose this strong way of doing things that people, who are deeply passionate about being there, buy into.

The most important cultural event is when you hire somebody. The second most is deciding when people aren’t a fit to the organization and removing them. I decided to interview every single person. It’s not crazy to do so when there are five but I think I interviewed the first few hundred employees and people were begging to stop. Today, you have to go through culture interviews along with functional interviews. We have two people testing for six core values.

One of our core values is to be a host to check if they are passionate about the notion of giving and hospitality. We created a core values council, which has about a dozen people that are experts in the values  and the culture. They act as an adviser to everybody in the company.

Culture, like beliefs, should never change, under any technological conditions.

Scaling Airbnb’s culture

Reid: There are certain things that you can’t do any longer when you scale…

Brian: You have to be relentless in doing cultural things every week. I do as many cultural things as before but they’re no longer person-to-person but I have people doing things on behalf of me. In the early days, I did all the interviews. Then, I hand-picked and trained people to conduct interviews. I built an inner circle to train the interviewers.

Earlier, I used to meet every single employee and give them a personal orientation. Later, I started doing weekly orientations. Now, we have everything recorded because we hire people from all over the world. We have institutionalized this whole on-boarding with videos and stories. I write these Sunday night email series, which are fairly thought provoking and cultural, to the employees.

Culture is about repetition. It’s about repeating over and over and over again the things that really matter at a company.

And, then it’s about trying to design as many things, such as the office space. When you check in, there’s a key card, on which it says “Champion the Mission” – it reminds you everyday to do that.

I wrote a blog post based on Peter Thiel’s advice who said “Don’t f**k up the culture”. You don’t f**k up a culture by focusing on it and designing it. The counterpoint is cultures are organic and that if they are designed, they’re fake and sloppy. I don’t believe that because most people who have created cultures organically wake up with a culture they don’t usually love.

That doesn’t mean you control everything but it means you control only a few things. I’m not going to tell the host how to act but I’m going to tell you that everyone has to be hospitable. I’m going to impose that on you. 

Culture is the (set of) norms that define what is the team we’re playing on and what is the winning condition. And how do we hold each other accountable – not through a hierarchy but through a community.

Scaling challenges

Reid: What were the things that started changing for you, which you had to learn, when you were scaling?

Brian: That’s a long list.

Hiring and management: Before you achieve product-market-fit, you and your co-founders could presumably do most of the work. After achieving product-market-fit, you do very little except manage, create a vision, hire people. Suddenly, you shift from building a product to building a company that builds the product.

I had to figure out a way to find great people, attract and recruit them and once you hire these people, you’re like – What do I do now? I have to manage these people. Oh God! I don’t really know what management is!

You learn, partially through trial and error, by making tons of mistakes. A classic management mistake is when people complain, you immediately appease them, which means you reward people who complain. And the people who don’t complain get disenfranchised and since they are not extroverted, they just leave the company.

Moving from intuition to data-informed decisions: When you are starting and building a product, I think data is not the most important thing. I know people do a lot of A/B tests to figure out a product but I really feel like the pre-product-market-fit data is not the most important thing. It wasn’t for us. We were depending on person-to-person interactions.

Later, you have to move on towards data information. A/B testing and cohorting and other things were completely foreign to me.

Moving from survival to strategies: When you’re a startup before achieving product-market fit, you do not think long-term because there may be no long-term. When you’re dying, you don’t think of what you want to do when you grow up. You think how do I not die and how do I plug this wound.

You don’t have plans and road maps and strategies when you are a startup.

The road gets tougher beyond first stage

Reid: What got you here, does not get you there. You move from having short-term plans to annual plans and hiring executives…

Brian: People asking about Airbnb’s founding days and it’s as if you have this idea and it takes off. This glosses over all the other stages. There are five stages where the first stage is fairly straightforward. Solve your own problem, do things that don’t scale, find 100 people that love it, have some really great co-founders that you trust, make sure you are full stack (designers and engineers in your founding team, ideally). 

Even if the first stage is complicated, it’s well-written about. There are not a lot of books on how and what to do once you’re past the first stage and everyone gives you wrong advice. In hindsight, I can say every six months, you have a totally different job.

It’s not like this metaphor where you’re a tennis player, you play small tournaments and you work your way up. You start playing tennis, then you’re like a bowler, then you’re like a football player and then you’re playing hockey. The job changes so much, you’re almost playing a different sport every job.

I don’t know if it’s rare but it’s hard to be the right person to start a company and be the right person to manage it when it’s a 1,000-strong company because it means you have to be different types of people. 

The skill sets start changing and the most important thing is being adaptable because nobody’s an expert at everything. There are two traits that people have who have been able to scale:

  1. General intelligence and talent
  2. Curiousness and adaptability

We have to be kind of kids at heart in startups where you’re curious, open-minded, welcoming, adventurous and not a know-it-all (they never scale). You should be shameless about getting feedback. I’ve had to surround myself with people much smarter than me and those with much more experience than me.

Competitive challenges lead to Blitzscaling

Reid: What’s your perspective on how competition changed as you began to scale?

Brian: There are different stages of a startup: 

  1. Survival: everybody calls your idea crazy and you can’t raise any money)
  2. Firefighting: Oh my God! We’re growing! I need to hire people and I don’t know what to do.
  3. Existential threats: There are people who decide to copy you and destroy you. One was government relations and two, competitors. 

We were warned about the Samwer Brothers and that they’re notorious. We were told that when they usually copy your website, they scale it and kill you. All of a sudden, we noticed suspicious activities where our users were being spammed.

I’ll tell you how scary it was. In 2011, Groupon was the hottest startup in the world where they went from $0 to $1 billion in revenue in a year or two, which was unheard of. Groupon grew that fast, not because of its US business but because of the European business called the CityDeal. This was the clone created by the Samwer Brothers that Groupon had to buy

We felt it wasn’t possible to beat them after having raised $7.2 million from you (Reid Hoffman of Greylock Partners) and Greg McAdoo at Sequoia Capital. In early 2011, they raised $90 million. We had 40 employees (it took us 2.5 years to hire them) while they hired 400 employees in 30 days. They opened 20 offices while I didn’t even know how to open the second one.

They were planning to build in Europe what Airbnb built in the US. For most companies, if they lose Europe, they’re just smaller companies. Airbnb would cease to exist if it wasn’t in Europe. We had a proposition to buy the company but it would have been very expensive, both financially and culturally.

I spoke to some people including Mark Zuckerberg, Andrew Mason and Paul Graham. Andrew warned that the Samwer Brothers would do anything to kill us. Mark told me not to buy their company and that whoever has the best product would win. He ended up being right.

Paul gave the best advice. He said we’re missionaries and that they are mercenaries and often, the former win. He said we should pretend they had a baby but that they didn’t want to raise it. So, I wanted my baby to grow into this wonderful company. My view was that I wanted to make the Samwer Brothers, who wanted to sell their business, run it for a long period. That was my way of revenge. I knew they could probably move faster than me for a year but they couldn’t keep doing it.

The ultimate way we won was we had a better community and a better product. It was a do-or-die time. 

We flew to Europe, hired a bunch of country managers, got them to San Francisco, trained them for a couple of months and asked them to go back, hire their teams and open their markets.

We opened 8-10 offices in three months. We hired hundreds of people and the whole speed of the company picked up at that point. It was insane. A couple of years later, the game was over. The Samwer Brothers were the principal triggers for Airbnb to Blitzscale. We went from being a US company to an international company in just a year.

Kindness in return for a (government) punch

Reid: The other hurdle you faced was government regulations. How did you deal with that?

Brian: In June 2010, we got word that the New York city was going to pass a law where they were going after a particular landlord (who wasn’t on Airbnb) and was converting apartments into hotels without getting licenses. A couple of years later, the New York State Attorney General told Airbnb/its users are violating the law. They wanted data of our users to enforce the law against thousands of people because they didn’t know if they were paying taxes and if they were breaking short-term rental laws.

We thought we cannot just handover specific, personal information of tens of thousands of users just so they could discover rules people might be breaking as that’s a fishing expedition. We challenged them in the court and we won. We eventually compromised by sharing a narrow set of user data, which all companies basically comply with.

It was a huge challenge to deal with government relations. We had a lot of trial and error. My first instinct was to fight. We staged a political rally in 2010 in the City Hall in New York City. Then, we realized that’s not the right approach. We’re a company where people are living together and we’re trying to teach that people are fundamentally good.

So, we decided to kill them with kindness. We wanted to show them we want to be partners. I wanted to avoid conflict. I have always had this instinct that if people don’t like you, you should never talk to them. Somebody told me this, “It’s hard to hate somebody close.”

I created this counterintuitive thing where I decided to meet everyone that hates me; my goal wasn’t for them to not hate me at the end but to hate me less after they got to know me. So, we went on a listening tour to meet with people who hated Airbnb and explained our story. After months of horrible meetings, there was a lot less vitriol against Airbnb. Now, we meet with senators, politicians and we were on the ballot in San Francisco.

Since we’re a business in the real world, the scope of things I’ve had to learn how to do is crazy. Over the summer (in 2015), we had a million people living in an Airbnb home every night. How do you keep them safe? They are staying in strangers’ homes in different countries.

Starting a company is like jumping off a cliff and slamming into an airplane on the way down. And, I guess, my point is, that never stops.

Airbnb’s tryst with terror

Reid: The Airbnb team was in Paris during the terrorist attacks in November 2015…

Brian: During that time, we had this annual conference called Airbnb Open (organized for the third year in a row) where hosts from all over the world fly in where we educate them on the new products and values. In this edition, we had 645 employees fly to Paris in addition to 5,000 hosts. Everyone from my life – my parents, sister and girlfriend were there.

We were done with two days of the conference and we were having a celebratory dinner for the first 40 employees who had completed five years. We got to know that there has been an attack in Paris. We thought it was sad but just continued. Again, our phone starts buzzing and we heard that there was a masaccre in a theatre and a suicide bombing outside a stadium.

At that point, 645 Airbnb employees and 5,000 hosts from 110 countries were going to dinners and events throughout the city of Paris. And, at that time, seven coordinated attacks happened within an hour in random places in the city, especially in the areas where our employees and hosts were at.

Fear struck over dinner. Some of our employees were at a restaurant near the theatre where the massacre happened. One of our employees was sitting in that restaurant, looking out the window, when he saw people screaming and running for their lives. We had a team at the stadium where there was a stampede.

I have never had to deal with something like this before. Our head of security created a remote command center. Since there were 50 of us in a 2-bedroom apartment, we had to make use of the available space and we created this command centre with our phones and laptops. We’re told we might lose the Internet any moment.

We created a list of all the employees and hosts and started calling and emailing them to check if they are safe. Next, we contacted the local and Federal authorities to help them with housing. We got our 300 hosts to open their homes for those that needed them.

We cancelled the conference the next day morning, spoke to the families of employees, flew all of them back and provided them with support. I’ve never had to deal with life and death like this before. 

And this experience reminded me of the first four years of Airbnb – the constant heart pounding and firefighting. It reminded me of the enormous responsibility I have – a couple of thousand employees, tens of millions of community members and that their livelihoods depend on you. It was a huge and humbling realization I had.

Learning how to learn the Brian way

Audience member: Could you talk more about learning how to learn?

Brain: I’m certainly not an expert on how to learn. There are a lot of tips but I’ll give you one. Let’s say, I want you to learn about the basics of UI design in a week. I want you to learn everything that’s important and become at least a temporary expert. You would probably read a ton of books, talk to a lot of people and go through this fairly exhaustive process. 

What if I told you in that week, I want you to learn about the basics of frontend development, accounting and also, understand how to incorporate a company? How do you do that? What you learn is you can’t learn everything about a topic and so, you have to be very good at short circuiting to learn from the definitive source about the topic. The skill is then about going to the right source to learn.

Somebody gave me the book ‘High Output Management’ by Andy S Grove and I had to just read that one book to know everything about management. I learnt to seek out the experts and the more successful you get, the more you have access to them. Even before you get really successful, you can certainly read about the best people.

You have to be shameless because I’ve found that most people will help you if you ask a question. The biggest reason people don’t help is because they don’t get asked. You have to have the courage to ask people and seek knowledge. 

I was shameless in asking questions, like over and over again, to the point where it was probably annoying, but I didn’t care.

How to find co-founders

If I wrote a book about how to start a company, I think the one part I wouldn’t be helpful with is how to pick co-founders because I got really lucky. I went to RSID with Joe and he’s one of my best friends. My other co-founder, Nate, was a happenstance. I had an intuition he was a great engineer and I could tell by the things he built that he is extraordinary.

Though I didn’t have to struggle with co-founders, I can give a few pointers. The mentality should be that your co-founders should be better than you. If they’re not better than you, you should at least feel they are equal. I see a lot of people who find co-founders who they consciously know aren’t equals and they become proxy early employees who don’t scale and get disenfranchised. They are not useful and are unable to contribute.

I think people do that either because they are lazy or insecure. They don’t want an equal in the company because it might be a power struggle. I think you should find people whom you admire, who are better than you and challenge you. 

The mentality is, if they’re better than you, then you will rise to the occasion and become better because you’re around them.

You have to have people you deeply trust and like. You are going to be around them for 12-16 hours a day. If you’re annoyed by them after four hours, you’re going to be annoyed with them after seven months. Ideally, you should have a long history with them. Generally, I find that complimentary skills work than overlapping skills.

Airbnb’s Growth challenges

There are many things but one of them is the challenge of transitioning from city to nation or village to city. At the end of 2015, Airbnb was at the phase where the core business was working well, the product was scaling well, we had hired a full executive team who were managing it. I could presumably go away for a couple of months and the company would basically run fine.

The product was working fine but the company needed day-to-day management. The two biggest things I focused on were: 

  1. Scaling the culture for the new size: A lot of the cultural norms and the things we institutionalized was for 500 people but by the end of 2015, we had grown to 2,000 employees. I wanted to make sure it still felt like a fast-growing startup, that it was still product-oriented and that everyone was mission-driven. This meant a lot of the methods and behaviors had to change.
  2. Shifting from single to dual-product focus: This is a bigger challenge where most companies that are really, really big have more than one product and multiple sources of revenue. For example, Apple has the Mac but they also have the iPhone, the iPod, the iPad and the Watch. Very few companies have a single product. Maybe Google is a notable exception where it has many products but one generates most of the revenue.

That was the challenge for me then. The core product could continue to grow but there were opportunities to start other products. I realized starting a product is so different from starting a product or a new business inside of an existing business that’s successful. You could pull up people internally and have access to unlimited funding but sometimes, people inside the company would inadvertently try to stop it. I was thinking a lot about this and all enduring companies must think about this shift.

Everyone said it’s the worst idea ever and it worked. So, maybe that’s a nice connection… the worst idea ever that worked!

So, what’s the worst idea you have?