What I learnt at Atlassian
You don't have to be in Silicon Valley to build a great tech company.
I hear this from lot of people. There are so many tweets, blog posts, podcasts and even books dedicated to this very belief.
It's a convincing idea - you build a good product, use internet to sell it and you will have the reach to customers all over the globe. Your location doesn't matter.
Yet, you find only a very few companies successful following this path.
Personally, I had experience with two early-stage startups based in Sri Lanka and Singapore. It was not easy. Even simple logistical requirements like accepting payments online take effort (at least now there's Stripe). But there are bigger issues. How do you hire the right people? How do you spread the word? How do you get into enterprise partnerships? How do you access funding if you want to accelerate the growth? If you are in Silicon Valley there's a strong eco-system designed to answer these questions.
In 2014, one of my friends living in Saigon texted me "Have you heard of Atlassian?". "I know they build Bitbucket and Hipchat. Why you ask?". A little surprise. She was not in tech. She was a consultant at a professional training firm. "Did you know they are a 10-year old company, based in Sydney and profitable? Also, they are the best place to work in Australia". "How you know all that? "They came to us to setup a training program for some of their developers in Saigon."
That conversation made me curious to look more into Atlassian's story. It was a tech unicorn built outside of Silicon Valley. They bootstrapped. They claim they sell enterprise software without a sales-team. They have offices in different cities around the world. It seemed they have achieved the impossible.
Few months later, something interesting happened.
Back in 2011, I went to Vienna to work with Haymo Meran and his team on a prototype for a real-time collaborative editor. They evolved into a startup called Wikidocs. I saw a FB post from Haymo, that Atlassian acquired Wikidocs.
I congratulated Haymo on Messenger. He asked if I'd be interested in working with them again, but this time at Atlassian. I happily agreed.
Haymo made some intros. I talked with few different people at Atlassian. I said I want to learn how this was built. They offered me to join their product growth team, which was the perfect place to understand Atlassian's secret to success.
In early 2015, I moved to Sydney to join Atlassian.
A University for SaaS business
Life at Atlassian felt like attending a university. Free lunch. Free ice creams. Team sports. Board games. Shower rooms with towel services. Social activities. And the most important thing - heaps of opportunities to grow yourself.
The past 18 months were intense, but exciting. Every single day I felt like I'm learning something new.
I started in a SWAT team assembled for HipChat in height of chat wars. Our goal was to have some short term wins while product engineers can focus on long-term strategy. Then I moved to focus on tooling to encourage company-wide experimentation. I was in Saigon, when the company decided to move entire teams back to Sydney. I experienced company's hustle to IPO and the aftermath. And six months ago, I left the growth team to join a new team that was formed to tackle an ambitious project, where I got a first-hand expereince of founders' thinking.
The pace of change was surreal.
Great culture trumps sophisticated processes
Like in a startup, most challenges faced at Atlassian are new challenges. This is also due to the rate the industry change. What worked 2-3 years ago won't apply today.
Atlassian has built a culture that makes people resilient to change. It starts with company values. Those don't just live in posters or wall paintings. People actively practice them. It's hard to pass a day in office without hearing someone mention about values.
The value "open company, no bullshit", defines how people communicate within the company. Decisions are not made in emails that get stuck in few people's inboxes. They happen in JIRA tickets, Confluence pages or HipChat rooms. Anyone in the company can search, subscribe and read them to understand what is happening. This builds trust.
Kudos system is also a beautiful part of Atlassian's culture. Any employee can give a kudos to another as a token of appreciation. You can choose from a range of options like a bottle of wine, a gift card from Amazon or a t-shirt from Threadless. There's no restrictions on how many you can give or receive. People stack their kudos cards on their desks as a symbol of pride. It felt special when I first received a kudos from an employee from another office, whom I have talked only on Hipchat.
I was also impressed by the candor shown by the founders at all-staff meetings. People can anonymously post questions to founders and upvote the ones they feel important. Scott or Mike will then answer the top-voted ones. Usually, there are lot of curveball questions and there's no beating around the bush.
Seek first to understand
I loved the weekly digest of customer feedback that lands in my inbox at the start of the week. It was a revelation how people either absolutely loved or hated our products.
I never quite understood the appeal of JIRA to be honest. It felt bloated and unintuitive on surface. Later, I found you need to put on a manager's hat to understand its power. JIRA for managers is what Vim is for developers. There's a learning curve, but once you become proficient it makes you more productive.
If you look at the history of JIRA, it was born as an alternate for Bugzilla for open source projects. Open source developers loved it and took it to their work places. Some of these developers were then promoted to manager roles. Atlassian was smart to understand its primary market - developers turned managers.
Hierarchies and bureaucracy are still rampant in the large enterprises. Middle managers spend majority of their time in either in meetings or compiling some bullshit report on progess of a project. With JIRA, these could be distilled into few JQL queries that runs periodically and then embedded as macros. No wonder these people love JIRA.
Note - JQL stands for JIRA Query Language, which is a subset of SQL specific to JIRA domain. It's such a brilliant idea.
Not made in Silicon Valley
I often wondered if JIRA would have been JIRA, if Atlassian was a Silicon Valley startup. Such unsexy product would not be tolerated by the Silicon Valley hive mind. It is not mobile-first. It doesn't use Ruby on Rails. It doesn't even follow a multi-tenant cloud architecture (lolwat).
Most early employees who joined Atlassian came from banks and other old school enterprises in Australia and Europe. They were tired of legacy systems and wanted to one up. But they weren't drinking the Silicon Valley cool-aid.
Instead they brought in a diverse set of talent and experience. They listened to their customers. Created their own unique culture. As a nice side-effect it turned out to be a business that worth six billion dollars.
Today, I walk out of Atlassian with this new perspective. Thanks for teaching me to fish. ♥️