clock menu more-arrow no yes
Photo Illustration by Grayson Blackmon / The Verge

Filed under:

How Australia became the test bed for tech regulation

Atlassian CEO Scott Farquhar on the challenges of governing online companies

A couple weeks ago, Facebook briefly turned off the ability for anyone in the world to post links from Australian news publishers. They just blocked ‘em with a little message. This is a real thing that happened. The company, along with Google, was locked into a fight with the Australian government over something called the Media Bargaining Code, which would have required social platforms and search engines to pay news publishers for linking to their work. Just linking to it. This was a big deal — it would change the way the web works. Both Google and Facebook threatened to leave Australia over the proposed law, until Google cut a deal with Australia’s biggest news organizations, and Facebook turned off their access to the news feed entirely. After a few days, the law was slightly changed, Facebook also paid the money, and access was restored.

What struck me about this is that we’d heard a lot from Google, from Facebook, and from Australian government officials, but we hadn’t really heard from anyone from the Australian tech industry. So I called up Scott Farquhar, the co-founder and co-CEO of Atlassian, Australia’s biggest tech company and one of Australia’s biggest companies, period. Atlassian has always been a global company — even when it was just Scott and his co-founder and co-CEO Mike Cannon-Brookes, their first sales were outside of Australia. That’s the power of the internet.

Scott and I talked about why Australia is the test bed for tech regulation around the world, due to its size, location, and how its government is structured. We talked about what happened with the Media Bargaining Code and how it will work in Australia now that it’s passed. And we talked a lot about how to run a global company in an increasingly fractured world, and whether new regulations help level the playing field or just cement incumbents.

Okay, Scott Farquhar, co-founder and co-CEO of Atlassian. Here we go.

This transcript has been lightly edited for clarity.

Scott Farquhar, you’re the co-founder and co-CEO of Atlassian. Welcome to Decoder.

Thanks for having me.

What time is it there for you?

It’s about 10 past 11 in the morning.

So, you’re in Sydney?

On a Thursday, I think ahead of where you are.

Right, so it’s 7PM Wednesday for me and it’s 11 in the morning for you.

Living in the future.

I’ve never talked to somebody in tomorrow. How is it?

[Laughs] It’s great. It’s really nice here. The weather’s fantastic.

I have a number of questions about Atlassian, how you are running the company and I want to talk about the Australian tech industry.

But, really basic question, you used to be a guy who managed a company from a plane, right? You used to fly to San Francisco, you’d fly around the world. How are you handling that in the pandemic?

It’s been great on a personal level. Not taking away that Covid has had many bad stories across a whole bunch of people and companies and individual lives disrupted. But from a personal perspective, I haven’t had to be on a plane every week or every month to go see people in the United States, or travel around the world and see our staff. And we had an executive team that was pretty distributed previously.

Some people would be in an office, where they would want to grab their colleagues and be in an office on a meeting. And that was great, but it meant we sort of had three or four meeting rooms dialed in, which is way less immersive than everyone dialed in Brady Bunch-style on Zoom. So that’s been really good as well, to have that sort of increased connection with staff around the world.

Is the time difference just impossible for you to manage? How are you solving that?

I think the world will move from where do you work in an office, to which time zone you work in. And that’s something we’ve been living with. We’ve had offices in a dozen countries or more. So we’ve had that for a very long time. And so the transition hasn’t been that difficult for us.

I feel like I could ask you “future of remote work” questions for the entire hour, just because it is a distributed company, you are in Australia.

But, let’s start at the start. Tell me about Atlassian. You are one of Australia’s biggest companies, you are Australia’s biggest tech company. Give me the story of Atlassian.

So, Atlassian’s mission is to unleash the potential of every team. And we do that by largely turning work into teamwork, getting your teams to collaborate better. And the teams we work most on are the teams involved in digital transformation. They’re the makers, the developers, those people that are working with software and beyond. We help them to be more productive and collaborative. And it started about 20 years ago in Australia with myself and my co-founder, Mike Cannon-Brookes; we still work together as co-CEOs almost 20 years later. We started off really building one product in Sydney called Jira — which is still our largest product — that started off as a bug tracker for software developers.

There were a couple unique things about that. One was we were in Australia, which is sort of an unusual place to start a company in 2001. And I think as a result of that, our go-to-market was very, very different. We didn’t sell enterprise software in the traditional way, which was to get a salesperson and sell to the CIO. Our experience was more consuming shareware, you know, games where it was “try before you buy.” And so we took that approach to enterprise software.

The way that Jira got marketed and sold was on the internet. It got sold at a low price, it got sold globally, you could easily download and install it. And that sort of started a flywheel of our business. And over time we’ve grown from that two employees at the turn of the century, to over 5,000 employees now around the world. And of course our products have grown from just one, in Jira, to dozens of products that handle all sorts of these cases around collaboration.

You recently acquired Trello.

Yeah, that was a couple years ago. And Trello has, I think it’s 50 million accounts or more that we disclose. And millions of users using it and it’s got incredible CSAT [customer satisfaction]. And people of all walks of life use it, from people using it in their personal life to manage information, all the way through to large companies. We’ve got hundreds of thousands of users in large companies using Trello to manage their work. So, we’ve got products like Trello, Jira, Confluence is a knowledge management product. We’ve got a whole range of products that help our collaboration.

For a long time, The Verge ran on Trello. And at one point, I realized, we were planning our wedding in Trello. And I was like, “This is too much, I’m spending too much time in this software.”

You’re not alone.

Why enterprise software, for a couple of 20-year-olds in Australia? In early 2000, enterprise software was not the thing, right? It was photo sharing apps or whatever. Why enterprise software?

Mike and I both did a scholarship co-op program. In that, part of what you do is work for companies, as well as do your university course, like a sandwich program. And in working for these large companies in Australia, both of us realized that we didn’t want to go work for a big corporation. We wanted to build something for ourselves and build something new. And so we wanted to do a startup, but we weren’t really sure what it would be.

The first incarnation was we provided support for third-party companies. So, there’s a company in Sweden, providing an application server, and they were great products but terrible support. And we thought, well, okay, we can provide the great support from Australia. Which was a terrible business because we were providing support to people in Europe, and people in the United States, and so you’re up at 3 in the morning. I remember my phone was set to the loudest ringtone to wake me up and you’d sort of stumble out of bed trying to sound lucid at that stage. And that product or, I guess, that service we provided was so bad that we decided we wanted to build something else.

At the turn of the century, there was just so much blue ocean in software. We built a ticketing system for managing those software customers, we built a website tracking system for tracking people moving across our website. And the thing that we were really excited about was that ticketing system, and turning it into a ticketing system that everyone could use. Because we felt that, back then, there was software that cost $100,000 to install, and there was free software, open-source, but there was really nothing in between. And we felt there was a huge market gap there.

What was the path to go from two people, “Okay, we’ve got a software product we like, now we have some customers,” to where you are now, which is Australia’s biggest tech company? Was that linear progression? Or was it fits and starts?

The product we ended up on, Jira, has been pretty successful from the start. In the early days, you’re doing lots of hustle. We didn’t have any venture capital behind us. It was just two of us bootstrapping. Coming out of university, we didn’t have money behind us that we’d saved to bootstrap. It was really put on credit cards at that stage. And so, we would do the usual hustle. We would go to conferences around the world. We couldn’t afford to sponsor a booth, so we would turn up at a conference with business cards and use those little beer tables that were usually set up around the place to open our laptop and do demos for people, just in the hallway.

In fact, at one of those conferences, we realized there was a podcast much like this [one] that was live-streaming from one of the sessions. So we stood at the entrance and we pasted or, you know, got stickers and put them on the beer. So, we went to the local beer shop, got cases of beer, then as people walked in, we put an Atlassian sticker on the beer can and gave it to every person in the audience. And so, that was sort of the guerilla marketing that we did in the early days. And then, over time, we sort of built the flywheel up, obviously hustling customers.

And then one day, we got a fax from American Airlines. And I asked Mike, had he been working with American Airlines, and he said no. And neither had I. And so, we’re like, “Wow, this really works.” They had faxed through the credit card number, and we’d never heard from them. And we’re like, this software business works when people send you money for effectively doing nothing. Obviously we’d spent a lot of time on the products, but we hadn’t had to hustle every individual customer. And that was probably a turning point for us.

So, that kind of brings me to the larger set of questions I want to ask, which is, as tech companies get bigger, their relationships to the countries they’re in and the countries they operate in get more complicated. There’s all this discussion of regulation. But American Airlines is an American company. You were in Australia and they just faxed you an order and you suddenly had a customer in America.

How quickly did you go global? Because the promise of the internet for a software company is that you have a global market from the first day, right?

Totally. I think our first sale was in the UK. We did a few more in Europe and Nordic countries. Then I think we had some sales in the US. And I don’t think we got an Australian company in our first 10 sales. And so you’re global from day one. And I think it’s interesting to see how that changes because, I think we’re seeing a new breed of companies. If you go back maybe 50 or 70 years, companies employed people locally.

Effectively, they sold their goods locally. They were governed by a government that was elected by people locally. If they polluted, they polluted locally. And of course, that was the place that their employees would be working in or put in the environment that they kind of existed in. And if you go back to the ‘70s, we moved to the sort of megacorporation or the global corporation that disconnected a lot of that local thing, and opened up supply chains and other areas and disconnected that link between a company and a physical location.

And I think, at Atlassian, we’ve got staff in dozens of countries around the world. We’re headquartered in the UK, we sort of feel like we’re Australian-backed. Most of our customers are in the US. So we’re a global company by default. And we benefit from that. We get to employ people in all these places and sell to people in all those places. But we’re also governed by laws in all these countries, both employment laws and, you know, how we sell in those countries. And so I think that’s just fascinating to see how the world evolves.

Before we came on, I was doing the research. I watched some of your other interviews. I noticed that when you are on the Australian media, they glow. They’re very proud of Atlassian as a big, Australian tech success story. Do you think of Atlassian as an Australian company? You just described it as a global company, but in the local coverage, there is a national pride around your company. Is that how you feel?

Atlassian’s the largest technology company in Australia. And the largest, I think, employer of people in technology, in Australia. And we’ve got Australian heritage and we’re really proud of that. I think that if you went to any of our offices around the world, you would feel like it has some Australian roots and Australian heritage, even if it was in many different cultures that we cater towards. So I’m proud of our Australian heritage.

We want to do as much as we can to support the Australian technology industry. And I think that, because we opened our office in Australia first, we got to tap into an Australian culture and community. We weren’t competing for talent with any technology companies in the US. And so we probably had more of a captive audience for our staff in the early days, which meant we had great tenure, and that helped us build amazing products. So I think we’ve benefited a lot from Australia. And selfishly, I live in Australia. I have three children here in Australia, and I want to make sure that Australia keeps up with the global technology arms race that we’re all in.

And Australia produces between 1 and 2 percent of the world’s GDP (gross domestic product). And if we want to continue our quality of life, we need to produce between 1 and 2 percent of the world’s software in order for us to keep up with that. And that’s kind of why I’m a bit of an evangelist for the technology industry in Australia.

But there’s a little bit of a tension there, right? The new breed of companies like Atlassian, like Facebook, like Google, you name it, are inherently global companies. They are conceived of that way, they begin that way. Atlassian began that way. And now we are seeing some of this nationalistic spirit, “Hey, our country has to do well.”

One of the reasons I wanted to talk to you is, my frame is the United States, which is a very complicated place to live most days lately. The clash between Facebook and Google and the Australian government, seemed asymmetric because those companies are really big. But even in just talking to you for 15 minutes, you’re saying, “I want Australia to succeed.” What is the dynamic there? What is the balance there? How should I think about that?

Well, I think it’s interesting for your audience to know that a lot of the things that are happening in Australia are a bit of a test bed for how things may end up in the world. And because they’re a regulatory regime, like we’re a parliamentary democracy, so we don’t really have a sort of third system of government, an executive branch. The legislature does, you know, both run the country and enact their laws. And so that means a lot of laws can happen faster and can be enacted a lot faster than they can in other jurisdictions. We’re also a relatively small country. You know, 25 million people is a lot smaller than many other countries. And so we can, I guess, move a little bit faster.

And so what you’ve seen in Australia is a bit of a test bed of legislation on a whole bunch of different areas. Whether that is encryption and privacy, antitrust involving media with big technology. Skilled migration is a big issue for us. There’s a lot of things that are being tested out here in Australia that I think are having global implications.

Let’s talk about the News Media Bargaining Code. This is kind of an inflection point, I think, in regulation around the world, particularly as it relates to Facebook and Google. The quick version, for people listening: The Australian legislature was going to pass a law saying Facebook and Google had to pay the media industry in Australia to post links from news. Both those companies were unhappy about that; they threatened to leave, full stop. Google made a deal with publishers; Facebook said, “We’re turning off links from those publishers.” Some time passed, there was a compromise, Facebook paid the publishers, and now they’re back in effect.

You’re sitting there as one of the co-CEOs of Australia’s biggest tech company. How are you watching this battle between the country you’re in and the American tech giants?

It’s been fascinating here in Australia. Atlassian doesn’t have, really, anything to gain or lose in this particular instance directly. But we are involved, because we want to make sure that we live in a country where the rules make sense, in a country where we can attract great people to come work in the technology industry, and where the laws don’t adversely impact the way that technology gets developed in Australia. Those jobs go elsewhere and so that’s where we come at it from, not because we directly benefit or are hurt by this.

And to further do what you said, in Australia and probably globally, we all recognize the need for a free press, a free media that is vibrant. But globally, media’s traditional method of funding, being advertising, has been eroded. Classifieds, you know. You don’t go into the newspaper to buy a car anymore. We have carsales.com.au in Australia, we have realestate.com.au. So those places have pulled a lot of the media dollars.

But also, Google and Facebook are now benefiting, I guess, in terms of just advertising dollars, not necessarily from news, but from advertising dollars. And so the government has said, “How do we have a vibrant media ecosystem that is free and fair, that we all benefit from, and how do we fund that as a country?”

They looked around and said, “Well, you know, Google and Facebook are making a lot of money. They also happen to be using news links in their products,” and so, in somewhat, I think, conflating two different issues, they said, “Well, they must be benefiting from the news and they should have to pay a lot for that benefit.” Then, what happened is, the Media Bargaining Code said, “Well, Google and Facebook are large. They’re maybe not monopolies, but close to it in terms of digital advertising.” So it would be unfair for any individual news company to try and negotiate with a monopoly, which I think is totally fair, right? Like, the mom-and-pop news company with 50 people that produce news isn’t the sort of person who’s going to be able to negotiate with Google.

On the flip side, Facebook and Google have said, actually, “News itself is one of many things on our platform and even if news disappeared, it wouldn’t make a huge difference to the engagement we have with our platforms,” and so that was where the stalemate ended up.

What happened is, Google realized that pulling out of the country, which is what they would have to do. The way this code is written is, you can’t just turn off news, you actually have to turn off news globally, to every person, which is very difficult for Google to do. And for Facebook, it was a little bit easier, because people actually post news in Facebook. Like, Facebook doesn’t scrape the web in any way.

And so we got to see two different reactions to the government, as to how that happened. So at the end of the day, what’s ended up as the compromise, is that, this law has gone through, which basically mandates named companies to have to go and negotiate. However, no company has been named, yet.

So even though Facebook and Google have done all these deals, they’ve really done those deals to prevent being named by, effectively, the minister for communications. The problem we’ve ended up with is that, at any stage, that minister for communications could threaten to name them and bring them back to the bargaining table again, and again, and again. Time will tell how that plays out, but if you came up to an election and the government wasn’t getting favorable news coverage, would they be able to, effectively, direct dollars from two old-school media companies, from technology, by threatening and saber-rattling?

And so we’ve sort of ended up with a very interesting stalemate at the moment, as a result of this bill.

Atlassian, basically, said, “This bill is not a great idea.” You have a statement of principles up on your website about how you think regulation should work. You’re in Australia, Facebook and Google are gigantic US companies, they do own, basically, the entire digital advertising market. How do you square all of this, as you think about, “Well, the next generation of companies are inherently global?”

There’s a couple things. One, as a global company based in Australia, our Australian reputation matters to our business and we want to bring more people here, so we spend a lot of time thinking about, “How do we have good laws that get made?” And, what we found is that politicians, until maybe five years ago, they didn’t really understand what technology was or that there’s a vibrant industry there. It’s amazing when you look through the newspapers, how many photo ops there are with politicians in a high-visibility vest, visiting some factory and, kind of, that’s their view of where jobs come from.

They’re still building things, creating things, but those things are software or they’re digital goods, and there’s less opportunity for photo ops in a high-visibility jacket, but that’s where the jobs are gonna come from.

You should just start getting people the jacket when they visit the office.

Walk through the office space with a high-visibility jacket. Watch out for that desk in the corner.

[Laughs]

But that’s what’s happening, all these jobs are becoming digital. Software is disrupting every industry, so our job as a company is to help educate our politicians on this. So we’re trying to find people that can speak both languages. We’ve got a head of policy, David Masters, who came from Microsoft, and he can speak both technology and politics. We’ve defined our principles, where, effectively, “How do we engage with policy makers?” and some of those principles are things like, “Treat the ailment, don’t kill the patient. Consult early, consult often,” those types of things, to make sure that we kind of have the principles for how we think laws should be made.

And if I could ask for one thing, it’s that both technology invest the time to understand the political landscape that politicians operate in, and politicians spend time to understand the technical landscape that technology companies operate within.

I’ll give you an example of the second. We had a different bill, which was, in Christchurch, we had a terrorist attack, maybe one or two years ago, and it was really bad, because one of the terrorists live-streamed the terrorist attack, whilst it was happening, on Facebook. And, horrible: no one should have to ever turn on Facebook and see something like that being live-streamed, but the way the law was written meant that, effectively, you had to prevent a live stream of a terrorist attack if it was being filmed by the terrorist. Of course, if it’s being filmed by someone else, it was actually fine, according to the law. And if you’re a person trying to determine, with machine learning algorithms, how to do this, it’s really difficult. Like, I don’t know how Facebook would do it, apart from banning all live streams.

This law was effectively conceived and passed within a week of this tragedy, and so you’re like, “I agree, we should do something, but it feels like this law was rushed in and I don’t know, actually, it’s gonna prevent anything happening in the future, because actually complying with it is near impossible.”

This is one of the balances that doesn’t exist in the United States, for example. Our government is very slow and often deadlocked. The Australian government seems very fast and often maybe too fast, but isn’t that better? Like, isn’t it better to have a government response and say, “Here’s how we think the tech industry should work.”

I mean, that law, eventually, was changed. I know there was a law about encryption that was eventually changed in Australia. Isn’t that a better cadence to be on? Speaking from product language, that seems like an iterative cadence where you’re trying to find the fit, as opposed to doing nothing, forever.

I do love the speed at which we can do things and engage, but it has to be in engagement with industry and so, when things are rushed through in a week, there’s clearly not any engagement happening. A good example is the encryption bill. There is a trade-off, as everyone in technology understands, between privacy and security. If you use the telephone in Australia, your telephone line can be tapped, according to a communications act that happened in 1970, that bound telecommunication providers to allow governments to listen in on phone calls, if they have a warrant and those types of things.

Now, people using end-to-end encrypted devices, like if you’re using Signal or WhatsApp, the government can’t listen in on those and so the government can’t prevent any terrorist attacks that might be coordinated using those tools. On the other side, consumers want encryption and privacy, to make sure that the government can’t listen in on their messages or even so those messages can’t be seen at a later date by a third party.

There’s a fundamental trade-off between those two, and Australia passed, effectively, an anti-encryption bill that said, “The government can designate any company, and they have to put in back doors for the government, in order for the government to be able to listen in.” And, there’s things in the bill that says, “Well, that should not introduce a systemic weakness into that particular platform,” but it’s not defined what systemic weakness means.

So that’s up to the eye of the beholder and again, most people would say, “Either it’s encrypted or it’s not, there’s not a halfway point where it’s half-encrypted or it’s encrypted with a back door or some keys there.” And so though that bill has had, subsequently to the bill being passed — so, it’s currently law — it’s had parliamentary inquiries to go look at how to fix it up. Politicians, once they’ve sort of fixed the thing the first time, there’s not as much energy in politics to go back and revisit something and make it right. There just really isn’t much benefit to politicians to make those laws be great, as opposed to just get them across the line.

So we’re still waiting on that law to be adjusted and improved. Even things like having an independent judicial review, under strong recommendations, haven’t come through yet. I think, we’ve seen in the US, potentially, with FISA warrants and other things out there, that independent eyes and independent review make a great difference to the public’s ability to have confidence in these areas. And so even though it was passed in 2018, it hasn’t been changed yet.

How does that affect how you think about building products, right? There is a regulatory apparatus that is doing things. They might pass laws. It might be a while before those laws are enacted, or improved or, even enforced. But, Atlassian is still building products in Australia. You’re shipping products in Australia. Does that feel like an unstable relationship? Does that feel like something you can count on? How does that affect what you actually make?

Well, as a global company, it’s not just Australia’s laws that affect us, it is the laws in every country that we operate in. If we just set Australia, the US, and Europe as three unique examples: Europe introduced GDPR, which is the General Data Protection Regulation, and that has been around privacy.

But the laws in the US and Australia don’t correspond with GDPR and so we’ve had to do specific things just for Europe. And in many cases, there are areas of conflict between those two jurisdictions. So there used to be a privacy shield which would go between the US and Europe to effectively say that one nation’s laws will apply in certain ways to the other nation, which allows companies to operate and work across both areas.

I think the privacy shield was recently struck down by the Supreme Court and so now, as a company, we really operate in an uncertain environment. That’s the thing that is probably the hardest overhead on companies, is that the uncertain environment across different jurisdictions, where everyone wants to make their laws, and in many cases they are conflicting.

In Europe, GDPR would basically say I must do end-to-end encryption, because I can’t share private data, and Australia’s law says I must share private data with the government, under certain conditions. It makes it very difficult to build one product, globally.

And like all regulation that provides sort of a tax on everything, it really benefits incumbents and large companies. Atlassian’s getting to that stage now, where we’re larger and we probably benefit, because it would hurt some of our upstart challengers to come have to have all this overhead that we can advertise over a large platform and a large number of products.

I just worry about what that means for innovation, globally. As an example, Clubhouse, which has come to the fore in the US, violates GDPR because it leverages your telephone address book to share things. It already can’t use Twitter’s network, it can’t use Facebook’s network, because those are closed, so it used your phone book’s network. But that means

Clubhouse could never have been created in Europe. I think that’s just an interesting state of the world you end up in, where the laws are so conflicting across different jurisdictions and what, potentially locking down which start-ups can exist.

Yeah, it’s funny, I was saying that we planned our wedding in Trello. Clubhouse, I wouldn’t have been so unhappy if they had been kept away from my phone book, because they sent me a notification that our wedding planner was talking about something on Clubhouse. I was like, “I haven’t talked to her in a decade.” Like, why would I? And, there’s a push and a pull there, right?

But, that’s an interesting part because they can only use your phone network, it’s the only thing available to them. If they’d used your Facebook friends or your Twitter network, they would have ended up with a much more high-quality network and you would have had a better customer experience, but the laws, or even those companies, prohibited that, and largely those companies prohibited it because of the laws that happened. You know, Facebook, with Cambridge Analytica, used to actually share your network with everyone, and people thought that was a great thing, until someone stole those data, used it in a nefarious way, and now they don’t share it with anyone.

We are sort of seeing the impact that the regulatory environment that everyone operates in, now we’re seeing that in, actually, how products get built.

So this brings me to the question that I ask every executive who comes on the show. The journey you’ve described is starting Atlassian, now being the co-CEO of a large company, there were two of you, you were excited to get a fax with a credit card number from American Airlines. Now we are here, talking about a globally fractured internet, and how to build and ship products in that internet. What is your decision-making process like? How do you evaluate all of these things? What has that development been like for you?

There are two things that I consider when making decisions. One is our mission as a company, and our mission is to unleash the potential of every team. That factors into how we think about things long-term, the areas we need to build for. And the other area that we take into account is our values as a company.

We have five distinct values that every one of our employees knows and lives every single day, including “open company, no bullshit,” which is one of our values to talk about transparency and working together. Another one is, “be the change you seek.” We expect our employees to take an active role in shaping Atlassian. So, between the mission that drives where we’re going and our values that kind of drive how we’re gonna get there, that sort of factors through every single decision.

Put that into practice for me. David Masters, your head of policy, who, by the way, the audience should know, he’s on the call on mute, he’s in the background here. David comes to you and says, “Rupert Murdoch, and Sundar Pichai, and Mark Zuckerberg are in a fight with the Australian legislature. We should do or say something.” How do you evaluate that and guide his course of action as your head of policy?

Well, if we want to unleash the potential of every team, one of the things we need to do is have, firstly, great staff, who are going to come and work for us at Atlassian. And we need to operate in a world where we have laws that allow us to do that. And so, for us, the derivation of that is that policy framework about how and where we engage.

And there are certain areas where we as a company engage proactively. You know, we have these principles that say, “Right, we’re going to proactively engage in things.” There’s certain things we’ll engage in reactively. So we will put it through the lens of “how do we help the world?” So we’ve engaged proactively on debates such as same-sex marriage in Australia.

Actually, Australia was behind the US in legislating that. Atlassian was a big proponent of helping us get there, because we believe those sorts of things really matter to our employees. We’ve had other areas where, you know, we don’t proactively engage, like we haven’t sat down and looked at the competition framework for Australia, but when something happens, we spend a lot of time educating people. And so, that all fits through that lens of how we as a company can make an impact in the world.

Let me put a cynical hypothetical decision in front of you. You’re a wealthy tech executive, you’re in Australia, you are part of a growing and vibrant Australian tech economy, Google says, “If you pass this law, we’re leaving.” Did it ever occur to you to think, “Well, I should just invest in a couple of founders, to build an Australian Google, to immediately replace that product.” Because that seemed like a very natural opportunity at that moment.

It’s interesting, in fact, that’s almost what happened, except Microsoft stepped into that frame. It’s actually interesting, Ben Thompson from Stratechery, and I’m sure he stole it from somewhere else previously, talks about strategy credits. You know, there’s areas where, because of your strategy, there’s a strategy tax, which I think impacts your business. Where your sort of local maxima aren’t reached, because you’re trying for a global maxima, and the strategy credit. And in this case, Microsoft and Microsoft’s Bing was a bit player, like sub-10 percent market share in Australia. I don’t know a single person that was using it.

I’m sure it could be an amazing search engine, just not quite as amazing as Google. And when Microsoft heard that Google was threatening to leave Australia, they called our prime minister and said, “We would happily live under any regulatory framework you want to come up with. You know, please make it as onerous as possible to make Google leave, so that we can come in and save the day, and be the white knight as the only search engine in Australia.”

And you know, that was very effective from Microsoft. I think that really got Google back to the bargaining table, and from Microsoft’s perspective, they don’t make much money from Bing, comparatively, so they would have been very happy to shut Google down in Australia and take that market share. So we have started to see the sort of geopolitics, you know, multi-3D chess game come out, where people and companies are actually using politics and legislation as competitive weapons.

I’m imagining Brad Smith at Microsoft being like, “Wait, we just brought Google back to the table? Like, “That’s not what we wanted to do here.”

Well, you see why Brad Smith is at the seat of the table at Microsoft, and you know, he’s on the executive team. Our head of policy, Erika Fisher, who runs all of our legal and HR, and compliance and policy, she sits on the executive team, because what you’re starting to see is that public policy is becoming a strategic business consideration. And of course, as Atlassian is the largest Australian technology company, our job is to help the local ecosystem, but it is something that is as strategic an asset as your products or your go-to-market these days.

Australia, in the world, is in a physically interesting position, right? You are close to China, you have these deep cultural connections to the UK, to America. Do you think you perceive that geopolitical fight that is brewing between all the big countries in the world differently than your average American tech company CEO?

Well, a couple of things on that. One is that there’s actually a book called The Tyranny of Distance that was written about Australia. If you go way back, Australia was a colonized country, and they’ve got to work out what to trade. Unlike India, we didn’t really have a spice trade, where we were trading stuff back. We couldn’t sell lumber, that didn’t make any sense, and there was really nothing to sell back to the UK, until when we imported sheep.

There’s a saying that Australia was built on the sheep’s back, and why is that? It’s wool. Wool was the first thing that was light enough, and valuable enough, to be worthwhile shipping all the way back to England, in order to pay for the spades and shovels and other things that the Australian economy needed to keep growing. And so, our geography has actually shaped the Australian psyche and the Australian economy over a very long period of time.

And it’s only maybe, I think in the last 10 years, we’ve shut down all the local car manufacturers in Australia. Because it didn’t make sense for us to build cars here and then ship them on ships around the world, to these geographically disparate and remote locations. So what is really exciting for us as a country, is now with these digital goods, we can build software — or Cochlear, which is an ear implant company, even though they don’t ship something virtually, they ship something very light and very easy to do. They can build and ship something around the world, and that’s how Atlassian came to be, is that, if we were trying to build physical goods from Australia, we wouldn’t have survived.

So that’s actually built into the Australian psyche, from very early on. Now, to get back to your question about how we see geopolitics as a result of this: I think Australia is unique, because our largest trading partner is China. Our largest, I think, cultural partner, would be the United States, and as a company, Atlassian sells to almost every single country in the world.

So we get to see a unique perspective of how that all plays out, and I would say from our vantage point, or my personal vantage point, the increasing parochialization of the web — or the splintering it into a Chinese web, a European web, a United States web, an Australian web — is not good for the world. These economies of scale that we used to get, where effectively 2,000 engineers could build a product that scales to the whole world, now we’re gonna need 20,000 engineers to build that same product 10 different times.

And that’s not a really good use of resources, and so that worries me. And you know, if governments lose trust in each other’s regulatory functions, because they have different approaches to data, security, privacy, then you’re gonna see these roadblocks emerge.

You mentioned the splintering internet, the phrase that we use at The Verge is “splinternet.” Do you see that actually happening? Is that a perceived danger, or is it a real and growing danger?

I think it’s a real and growing danger. I don’t think we have seen it getting implemented in companies just yet. But also, by the time it happens, it will be very difficult to undo, because the reason it will happen is because the different governments will have entrenched their particular viewpoint. So when the Australian government won this, in their mind, won a battle over Facebook and Google, and then Europe does in a different way, there’s no incentive for them to go back then and harmonize the two different approaches. And so then suddenly, Google, instead of having a lot of employees in one place that can build stuff globally, are just gonna have to start building it two different ways.

And like, you won’t notice it to start with. It’ll just be the number of employees dedicated to a country, or a particular jurisdiction, goes up. And that’s sort of a dead weight loss for society, and probably the biggest worry is how it entrenches the incumbents.

I hear you on the incumbents thing, and that’s a point well taken. Let me push you on the loss to society point. Over a year ago, pre-pandemic, in a different lifetime, I had an economist on The Vergecast, his name is Thomas Philippon, and he was saying, one of the big issues in the American economy in particular, is that software companies make more and more and more money, but they actually contribute less and less to the real economy. They hire fewer people.

So if GM wants to make 100,000 more cars, they have to hire X number more people, they have to purchase X number more car seats, their suppliers will have to hire more people. There’s just a net increase to the economy, to the productivity of a company like GM. Google wants to ship another billion web pages, they have to hire three more people and a new data center, and that’s it, and there’s no net gain to the economy.

Wouldn’t part of the argument here be, “Okay, well, if Google has to build a European Google, and an Australian Google, they might just hire more people in Australia. They might bring more senior software engineers to Australia, to build a version of Google that is local, and responsive to the needs of the people in Australia.”

I think every government wants more jobs in their jurisdiction. That doesn’t necessarily mean that it’s good overall, as a maximum. I think if you go back and say, “Well, we used to deliver a lot of CDs to music shops. And you know, there are a lot of jobs in people that would press the CD, pack the CD, put them in cars, and drive them around. A lot of people in retail, that were involved in selling CDs. And yet we buy all our music online. Should we go back to having people doing all that, just to create jobs?”

Like, no, actually, it’s way more convenient, and we actually sell way more music, and we’ve got more musicians producing music, than ever before. And so, it’s sort of what jobs do we want? I think there’s a big discussion to have around how the largesse is divided in the world. So you know, taxation policies and redistribution policies, whether that is universal basic income, whether that is taxation, whether it is a social safety net, whether it’s free health care.

There’s plenty of different ways in which governments can redistribute money from wealthy to less wealthy. But I don’t think creating jobs for jobs’ sake, the old Depression-era, “Let’s dig a hole and then fill it in, to create jobs,” is the right way to think about it. We’ve got to think about, what’s the best experience we want? And then, great, how do we redistribute that, or create new jobs in this new economy?

We have spent a lot of time talking on policy, and to be fair, I invited you here to talk about policy, but you said public policy is a strategic advantage for a company like yours. How much time do you spend on products versus policy?

As the CEO, I always wish I could spend more time on products. That’s my love and my passion is building stuff for our customers. Our customers don’t buy our policy, our customers don’t buy our packaging or our pricing. Our customers buy and use our products, and so that’s where I want to spend most of my time. Our senior executive team, I would say that — let’s say GDPR for example, it was named internally, something like the Generalized Destruction of Product Roadmaps.

That law came into place and every person in product had to stop whatever they were doing that they had on their roadmap to provide for customers globally, and instead had to pivot. We had hundreds of people building these data protection regimes, and for privacy it’s a great thing. But it wasn’t top of the list our customers were asking for.

And so sort of having a statement, or at least a voice in these discussions, to make sure that these laws are implemented in the right way that achieves the policy outcomes, achieves the way that politicians — and again, politicians are elected by people — so ultimately it’s the will of the people, what gets done.

And so how do we do that in a way that actually makes it the best way for technology firms to do this, and that’s, I guess, why I spend time on the public policy. I don’t pretend to know better than politicians what the world wants, what voters want — that’s their job. But in terms of translating that into laws that can be implemented to achieve the outcomes, I think that’s where we in technology can help.

You have a co-CEO, most companies don’t have co-CEOs. What’s the split between you and Mike?

On a good day, he does 80 percent of the work and I get 80 percent of the credit.

That’s the way we’d like it to go. But practically, both of us have done every job in the business, from cleaning the bins out when we first started Atlassian, to running every department. We sort of mix it up every other year or so. Today, I run the go-to-market functions and the sort of G&A [general and administrative] functions, so legal, HR, and finance. And Mike runs all the product functions. So all the product management, the design, the engineering, and so forth. And so that’s how we split it up, but we’ve done everything at various stages.

Well, the reason I ask, that seems like the natural split, right? There’s the sort of policy side of the house, and the product side of the house, that goes to the other question I asked you. But is there a time that you made a trade-off in the product, because of a policy concern?

GDPR is probably a great example of where we have done that. As sort of the broadest-reaching legislation. We are now looking at data residency, is an interesting one, and companies and laws are such that data needs to physically reside in a data center on that country’s land, in order for it to be applicable to certain rules. And it’s a little bit backward-looking, because you know, kind of where the bits physically live on disc is a weird way to think about the cloud, which most people don’t think about when they open their browser. They don’t really think about the physical location of the bits.

But from a policy perspective, that’s the way the world has ended up. However, things like privacy shields, in Australia there’s the Cloud Act between the US and Australia that allows certain things to happen under those data residency regimes. And again, the ability and what we need to put in data centers, what data needs to reside in different locations. There’s things like identity, which kind of need to reside everywhere, because you know, you don’t sign in to log into Australia, you say, “I want to log in,” and then we redirect you to Australia. So all those sorts of laws about how data is protected in each country, and how it can be shared between countries, really affect concrete product decisions and the efforts that we put in.

Jira is used everywhere, it’s used by the biggest companies in the world. My understanding of big companies as customers and clients, is that they are themselves very demanding, especially when they are themselves global. How do you balance, “Okay, the Fortune 50 all uses Jira, those are big contracts, here’s what they want from us, we might have to tell them no, because of the GDPR.”

As a company, we serve from 10-person teams to 100,000-person teams, and so you’re right. Like, we run the full gamut of that. At the top end, it’s not a matter of saying, “No, I can’t serve you, because of these laws.”

It’s a matter of how do we actually comply with these laws. There’s a company in Australia that are looking to use our cloud product, and because of the regulation, they’re saying, “Well, we ourselves operate in 13 different countries, and before we can sign off on your cloud product, we need to get sign-off from our regulatory authorities in 13 different countries to do that.” And again, I imagine the laws governing this are pretty similar across the board, but the fact that they’re different in 13, even if they’re 1 percent different, means they need to get sign-off in all those different locations.

Again, I think as a technology industry, we could do a better job of helping these governments harmonize their laws, or interpret them and say, “Well, all cloud companies interpret the Australian law this way, so how can we standardize on that,” and that will make it easier for people to consume that. It would make it easier for governments to sort of harmonize their laws.

So, there’s definitely things we can do on our side. But I think there’s also things governments can do, to take the same law someone else does, or say, “I’ll do it your way.” In fact, it’s interesting, we have a neighbor called New Zealand, and you know, I don’t know, it’s like in the US, in New Zealand, anything that’s great, that’s [from] New Zealand, we claim it as Australian. Maybe everything good that comes out of Canada gets claimed as American.

Yeah.

But we do that for all of New Zealand. One of the things they’ve done is, a lot of their laws — because they’re a much smaller country, a couple million people — many of their laws just say, “We do what Australia does.” So, for example, drug testing, kind of the equivalent of the FDA in the US, New Zealand doesn’t have one, they really just say, “We’ll do what Australia does.” I think the world would be a lot simpler if we could do that more often and harmonize laws across different jurisdictions.

So I obviously have a very US-centric view. I grew up here, I passed the bar exam here. So my entire world view is from the United States. You’re describing something that the United States would never do, right? Like, the United States would just never be like, “We’re just gonna do what Australia does.” Like, it is impossible. Beyond New Zealand, do you see that happening in other places?

You know, there’s a lot of things we do harmonize on. We’ve got the Paris Accord for climate change, and you could argue whether it’s perfect or not. But we got 200 countries around the world to come together all effectively to have a standardized measurement system and standardized targets. And we can have a whole different discussion about how Australia is doing that, and I think we’ve bent the rules slightly in terms of how we’ve interpreted them.

But I think if we can do that on climate change, you know, really complicated things, then there are other areas where we can put together international bodies to work out what is the union of all the different laws and things people are considering. Data privacy is a good example. I would settle for one data privacy act across all of the United States, let alone internationally.

Because you’ve got one in Europe, you don’t have one US-wide, and California is looking to do one. And then, of course, there’ll be other states that’ll follow in California’s footsteps. And so if we can harmonize those things, the Cloud Act between Australia and the US is a good one, Privacy Shield between the US and EU where we try and harmonize laws. I’m optimistic that there are big things we can do there.

As an Australian CEO, you’re obviously raising your children in Australia; it’s remarkable that you know about the law change in California at a state by state level. How distracting is the United States to you as an Australian CEO running a global company?

There’s a couple of aspects on that. One is employees. And I will say that, not just legislation, but it’s been a very tumultuous time in the US over the last 12 months and we have about half of our staff globally there. And a lot of them have had to go through a pandemic, very difficult racial tensions that have happened. A very tumultuous election has happened. And so I’m just really feeling for everyone in the US at the moment, in terms of like, what they’ve had to go through over the last 12 months.

And as CEO, I’m responsible and try to do our best to help that. And then, you know, the United States, beyond that, has a big influence on the world, and a very large customer base of ours. Just under half our customers are in the US. And so I’m very acutely aware of what happens at a politics level, and at a regulatory and a competition level over there.

When you say “acutely aware,” do you get a briefing book in Trello every morning that says, “Here’s what the United States government did overnight?”

It is funny. When I got my first investment from Accel, so this was sort of 2009, 2010, I would fly over there and I’d go into the offices. And I remember one day, I think this is very early on, there was a revolving door of CEOs at Twitter. I can’t remember who was in or who was out right then.

And I was in a meeting and I just brought that up. And they were sort of amazed, like, “How did you find that out in Australia?” And I was like, “What do you mean?” He’s like, “Well, how do you... How do you know?” I was like, “Well, I read the same news articles you read.” [Laughs]

It’s not like you’re in the Twitter board room and understood it firsthand. Like, you found out about it from the same communication channels. And so I don’t think there is a difficulty these days, being the expert on any local area, because all that information is available online. And so we can be just as informed about what’s happening in the US from Australia. And the same thing, I think I’d advise people in the US to be up to date with what’s happening in Australia, given how it’s sort of a test bed for legislation that may come down the line. And I think we’ve already seen this. Just today we saw, I think it was Congress, basically talk about a bill where they would allow media companies to collectively bargain against big technology.

And in some ways that would be a better solution than what we have in Australia, because then if all the media companies collectively bargain and say, “Well, our news is worth X.” And Google says, “No, it’s not,” and they shut them off. Well, then that sounds like it’s actually pretty fair because they’ve equally bargained on both sides. But I think that would not have come to the floor if it wasn’t for Australia’s legislation paving the way. So I think people, globally, should be kind of looking to Australia and other countries that are enacting legislation as really a stepping stone to what might happen locally.

Yeah, it’s interesting, all of the antitrust action in the United States began with that media collective bargaining activity, which has grown from that to something much bigger. Also, I have to compliment you, you said today but it is tomorrow for you. So well done. I would’ve not gotten that time zone right.

I want to end by asking about something that is much harder to see, right? I can go read Australian news coverage, you can come read American news coverage. It is very hard for us to see what is happening in China. You said China is Australia’s biggest trading partner, I’m assuming Atlassian does business in China?

We do a small amount of business in China. It is, yeah, a small amount of business. It’s not zero, but it’s not a huge amount compared to, I guess, the population that is there.

As I think about the splinternet and the fracturing in the global internet. Obviously, the difference between the Chinese internet and the rest of the global internet is vast. Again, Australia is in a unique position, you have a different perspective than I do.

How should we think about that market, which a lot of companies want to go into and grow, versus how that government treats some of its citizens, how that government treats the flow of information on the internet there, how the government exports technology. I’m just imagining you have a much different perspective on it than I do.

I don’t know your perspective, but mine is that, well, first of all, we don’t have engineering in China. We have, I think some partners in China, but not employees there.

By the way, by perspective, I just meant I live in the United States. That’s all I meant.

Yeah. From a China perspective, it’s interesting to see. Like they’re a rising superpower, right? The US was a rising superpower after the second World War. You know, prior to that, they didn’t really have global dominance, and China is trying to do that themselves, as well. However, you know, their values are so different from what we’ve come to see as liberal democracy, Western values where rule of law from a democratically-elected perspective is really respected, competition, other things.

Just a very different set of cultural values. I’m sure they would look at ours and say we’re weird, but there’s a lot of things, I think, that mean it’s going to be very difficult to harmonize these two worldviews over time. And so, from a commerce perspective, almost every CEO says, “I’ll make China work.” You know? “I’ll be the first CEO to ever make China work.”

And I think there’s enough data points now to show that it is near impossible for someone outside of China to go in and make it work. Whether it’s intellectual property theft or it’s just the rule of law over there, or just other things that you don’t understand culturally about it. And so, I feel optimistic that those two worldviews can coexist. But I don’t think they’re going to merge anytime soon. And so when I look at global growth markets, I don’t put China as something that’s gonna be, you know, 20 percent or 30 percent of Atlassian’s revenue, even though there’s a lot of stuff that we do that would hugely benefit from the way we work.

I think even some of the cultural aspects around collaboration and being open and transparent would make business work better there. It’s not a big area of investment compared to, say, India, which has a largely English-speaking population, and also, a much more solid rule of law. I think those types of economies have bigger growth prospects for technology companies like ours.

Let me wrap up by asking just a big “step back” question. You and Mike, you’re two kids, you went through a workshop program, you start shipping a product, you realize your support product is more interesting than the thing that you were trying to do. This is like a classic founder story, right?

You’re two people, you start a business, you pivot into the tool you made to solve your own problem. That grows into Atlassian. Atlassian is a gigantic global company. Now you are more aware of California privacy law than even I am. Like, that’s a great journey. But it all depended on you being able to address a global market from the beginning without — in 2001, 2002 — the pressure of multiple regulatory regimes and privacy regimes, and all that happening. Is that still possible for two founders in Australia now, or two founders somewhere else?

I think it’s still possible to do that. I think it’s still possible to build a global company from anywhere in the world, and that’s great for everyone. It’s why you see huge amounts of venture capital plowed into really early companies because basically, the size of the market now is way, way, way bigger than we ever perceived it to be. There’s billions of people on the internet, and all of them could potentially use your product. So I think that that is still very possible.

I wanna just keep the balloon up, as a big warning that we shouldn’t take that for granted. It didn’t exist 30 or 40 years ago. Like, there were multinationals, but it took them 50 years to become a multinational. Whereas these days, you can be multinational from day one. I don’t take that for granted because it could go away really, really quickly, and I think we would all lose out in that scenario as consumers.

All right. Well, Scott, you’ve given me far more time of tomorrow than I anticipated. Thank you so much for being on Decoder.

Nilay, thank you, it’s been a pleasure.

Policy

Elizabeth Holmes says Theranos’ former president abused her

Policy

Amazon must redo Bessemer union election, orders labor board official

Policy

Frances Haugen will testify before Congress again this week

View all stories in Policy