Bootstrapping Atlassian: $10k In Credit Card Debt To $20M In Revenue In 5 years
thedisruptors.beehiiv.comYikes, I have to create an account and subscribe to read the second half of the article? Think the complaints about that are obvious. Leaving a comment in case anyone else wants to avoid that particular experience.
Shame, it was an interesting and well written article.
And neither 12ft.io nor google cache are able to bypass it.
Guess I’m not reading it then.
Even though I have emails created specifically for stuff like this, I refuse to uphold the practice.
I used the author's email and that was accepted. No confirmation necessary.
I assume the article is a transcript of an episode of "How I Built This", if podcast is more your thing
https://podcasts.apple.com/ca/podcast/atlassian-mike-cannon-...
Great! Lucky I haven't clicked the link yet so not giving traffic to them.
Atlassian, before Atlassian, asked me to open source my wiki engine, because they wanted to create a wiki. Did that, LGPLed it, and Confluence was born ;-)
[Edit] "Engine" is ambigous, wiki render engine would be more precise [0]. To be transparent and because it is not and was not my intention to mislead people: the code was a library that rendered Wiki markup into HTML and had a plugin/extension system to e.G. render syntax highlighted code, diagrams and integrate with third party systems. It implemented this with a macro system using {...} syntax. You would need to add at least load/safe e.G. with Hibernate to have a (minimal) Wiki application. But with Radeox you could create your Wiki app in perhaps 100 LOC.
[0] https://www.researchgate.net/publication/200773455_The_Radeo...
[edit] This is a response to the original unedited comment that claimed he wrote the first version of Confluence
I'm a former Confluence developer (2008-2013). I don't know anything about you or the detailed history of Confluence.
- When I started there were 12 people on the team including me. I wasn't there at the very beginning but some people on that team were and they would discuss design decisions from the start of the project.
- It was a spring-hibernate-webwork Java web application. Very complex and structured. Not your typical weekend project. Mike Cannon-Brookes (co-founder) of Atlassian wrote a book on this type of architecture.[0]
- They were always very open and forthcoming about open source they were leveraging. I don't remember this ever being discussed or mentioned.
[0] https://www.wiley.com/en-ie/Java+Open+Source+Programming:+wi...
[Edit] To make it clearer I annotated my initial comment because I don't want to mislead people.
[Deleted b/c it was inappropriate]
The wiki engine was called Radeox and rendered wiki markup into HTML. It was written in Java and open sourced from an existing Wiki application called SnipSnap [0]. If you have worked at Atlassian as a developer, you should have seen that name. It was also used as an example by Grails developers, e.G. in "The definitive guide to Grails"
References to it can be e.G. found here:
https://docs.atlassian.com/ConfluenceServer/javadoc/7.12.2/c...
The way macros are done and I have invented can e.G. be found here:
https://support.atlassian.com/confluence-cloud/docs/insert-t...
I have discussed things with Mike Cannon-Brookes long before you've joined the company. Sadly the oldest mail I can still find from Mike to me is from 2006 and was not about this topic.
In fairness, "Wiki engine" to me made me think Confluence was based off of XWiki or something similar and that you contributed that, rather than a Wiki macro renderer.
Additionally, Confluence being born from it seems like something someone should have the receipts to as well.
Maybe someone can reach out to Mike about it.
Yes you're right, apologies for creating a misunderstanding (see my edit above).
Not sure, it was 20+ years ago, but I might remember XWiki also using my wiki render engine back then, but I could be wrong ;-)
(it could be a reimplementation of the interface)
"The XWiki Syntax 1.0 was supporting Radeox macros."
https://extensions.xwiki.org/xwiki/bin/view/Extension/Old%20...
The group of people back then was small.
I've started work on this even before Wikipedia was started, just as a point of reference.
Asking this in a friendly manner, but don't you feel bad that something you created was taken essentially for free by a corporation and monetized? Money isn't everything, but I would think it fair for the creator to receive their share
Mostly not.
But sometimes yes, company created billions [0] and I haven't seen a cent :-)
[0] Yes, also JIRA
It's quite the resume bullet point I assume!
Worked as a manager/CTO for decades, it's not that important.
Now I work as a CTO coach for some years now - which I enjoy a lot - and it has even less impact there.
But perhaps I'll add "From the guy who put {macro} into Atlassian® Confluence®" :-) Thanks!
it is an impressive feat, sorry if it came across as negative! hopefully they treat you to dinners at least
Every open source developer has been asking this question for the better part of three decades. Possibly four.
I think there's still a difference of using small libraries which contribute to a single or couple functions here and there, versus taking an entire open source product and wrapping a commercial structure on top of it. Could be hypocrisy on my behalf though
If you don't want that to happen, then don't release your software as free software. Use a CC-NC license or something similar.
I'm not nearly talented enough to build something open source someone else would want to commercialize
Those small libraries of “Hey, I figured this out so you don’t have to” aren’t asking for $10,000/mo. in support. The issue is open source software projects that become monetized by another company as a product for profit without following the Red Hat model of supporting the project.
So you're the one ultimately responsible for all the suffering it's wrought? It looks nice, but not having access to the plain text source for pages is infuriating.
No they wanted to have a wiki, it was fashion back then. I brought a lot of suffering with the {...} notation of macros though (also not using {/...} but {...}, but I was young)
I'm responsible for everthing like e.G.
https://support.atlassian.com/confluence-cloud/docs/insert-t...
It's always telling when the top comment is about who gets to take the credit for whatever. I never understood that - you either sell your stuff or be fine with it being used by others for free
Yes, as I've said, I'm mostly fine and today would choose MIT over LGPL.
Why?
> my wiki engine, because they wanted to create a wiki. Did that, LGPLed it, and Confluence was born ;-)
What is the name / URL to your wiki engine?
This is 20 years ago, I don't think it exists (not on Github).
A reference can be found on the (back then) famous C2
It was cut out from a Wiki application called SnipSnap
Did you do that before or after GPLv3 was released?
Before, but today I would choose Apache or MIT.
Why those in particular?
As free as possible. I want open source to be used without any restrictions (like the GPL) but without liability. I write closed source software to earn my living (just my view).
The thing that blew my mind at the time was that the initial licence was like $9, something you could throw onto your credit card as a dev and just have a heaps better way of tracking bugs.
Most of their growth playbook from back then is obsolete or basic knowledge these days.
Yep, exactly. The more I read and it seems that these huge incumbents were all building in a blue ocean market.
Have you ever found any useful guides for bootstrapping in competitive markets?
It seems based on indie hacker building a personal founder brand is the only viable method for low customer acquisition.
I have very little personal brand (I went from writing about React to around 900 subscribers to running an uptime monitoring SaaS - https://onlineornot.com), but am still managing to grow in a red-ocean market.
The trick is:
- being default-alive (the business will never go under, as I have a full time job)
- continuous progress (2 hours a day, every workday)
- learn everything you can about sales and marketing, the engineering is the easy part :)
> learn everything you can about sales and marketing
Are there resources you'd recommend for this?
I've mainly been learning by doing (and failing), though I've read these books which helped solve problems I had at the time:
- Badass: Making Users Awesome by Kathy Sierra
- Obviously Awesome by April Dunford
- The Mom Test by Rob Fitzpatrick
- Every book Seth Godin has written, particularly This is Marketing
- Deploy Empathy by Michele Hansen
- Forget the Funnel by Claire Suellentrop and Georgiana Laudi
I guess the meta lesson is find today's blue ocean markets.
The "obvious" things the Atlassian founders did back then were not obvious at all at the time.
> ... but that product had terrible support and documentation. So they decided to provide documentation and support services for it, charging customers $360 to solve their issues. However, they faced two problems: First, their documentation was so good that customers rarely needed to call them...
Interesting to see how documentation & support are such pain points to sell enterprise software, and yet how few companies manage to differentiate themselves by making truly great documentation & support experiences. Stripe is another good example
Classic survivorship bias, 100s of other startups went bankrupt and left their founders with massive high interest debt.
How do people pay the bills when they start a company? I'd go start a company right now if I could figure out how to pay my car payment while doing it.
- Have a family that can back / help you out financially.
- Have a working spouse that can float you.
- Have funding secured BEFORE launching your startup.
- Have a solid network of friends, acquaintances, co-workers, business partners, etc. that have enough money to support you.
And so on.
Not really the best kept secret, but a LOT of startup founders come from social classes where money isn't that difficult to get. And others work in lucrative industries before launching their entrepreneurial careers.
For ages there's been this debate on being able to afford (taking) risks, which tends to be strongly correlated with your socioeconomic background.
Taking out credit-card loans, mortgaging your house, etc. def isn't the norm.
> Not really the best kept secret, but a LOT of startup founders come from social classes where money isn't that difficult to get.
While maybe not a secret it is striking how many startups had initial "friends and family" pre-seed/seed.
I think it's very important to call-out for those like OP where having ready access to five-six figures is one of the main reasons why a lot of founders can "take the leap" while it seems nearly impossible/terrifying for those who don't.
Needless to say if things don't work out those same friends and family can help the failed founder "get back on their feet", repeat for the next one, or keep dumping in cash for runway.
Having ready access to such funds and the safety net these same networks often provide makes a world of difference.
Maintain low expenses, be OK with going down in life quality, be willing to burn through your life savings and be OK if you lose it all.
I did it once, now doing it for the second time. I think most people will not bear it, but for me it feels like the only natural thing to do.
I can never imagine enjoying either running a hyper-growth VC funded company or being an employee, and I realize most people are not like that.
So on that end, maybe ask yourself if you were OK with cutting costs like losing the car, moving to a smaller apartment (even back with your parents) etc. and be happy about it even if your company flops.
Don't have car payments? Just buy a cheap or used car for cash instead of a fancy car that you can't afford.
Live below your means, so you're not stuck in working longer and longer hours just to pay for your lifestyle.
Accept help from family. If your parents have a big house, move back in with them. If your wife has a good job, ask her if she's okay with being the primary breadwinner for some time.
Build a small fraction of your company on the side, while still working as an employee.
Maybe even find some contract work to do part time.
Or find an investor who pays you a salary to work on your startup :)
Start with sales.
It may take a few of months ( <6 ) to cover your cost of living, but if you can sell, it’s readily achievable. The trick is pricing high enough to actually achieve this on 2-4 customers.
In my experience, almost all new businesses price far too low in the early days.
I started my business with a product that cost 5€. There's no single path to success, there are a million ways to start a profitable company.
Charging anything is more than many new businesses charge.
1) Have jobs that pay well, and do a pretty good job of saving for years and years. There are several reasons that, despite the prominence of young founders in pop culture, successful founders in fact tend to be quite a bit older—that's one of them.
2) Family money. Either have access to it directly, or be personally "poor" but with an excellent safety net in the form of rich (-enough) relatives who won't let you really suffer if things go poorly (and may help fund your business, even). Doesn't need to be, like, Rockefeller money, either—parents who can guarantee you a place to live, comfortably cover your living expenses for a while, give you the occasional four or five figure interest-free loan without putting their financial security at risk, that kind of thing, can be enough to eliminate most of the personal risk of trying to start a business.
I spent the first 15 years of my career grinding and saving cash + equity as a SWE. It turns out that some of that "worthless" startup equity eventually CAN turn in to Real Money and now I'm using it to pay my bills while I build my own thing.
Work nights and weekends until you start getting paying customers and then hopefully you can reach a point of walking away from your existing job.
I got pretty lucky with this; found a cofounder who came from money and was happy to bankroll us for 6 months until we closed our first round.
Still not profitable after 20 years, so I wouldn't call this success.
I can make millions in revenue by selling dollar for 50c as well.
It’s certainly been profitable for the founders who’ve been selling millions of dollars in stock every few weeks for the past year.
It's interesting they had "don't fuck the customer" on value list...
It's a great story, but at least in the software world, this sort of thing is exponentially more difficult to pull off today than it was 10 or 15 years ago. The internet was the most economically revolutionary invention of all time. Sell to anyone! Connect anyone in the world instantly! It created a lot of low hanging fruit. So much that it took 20 years or so to grab it all. However, in 2023, all that low hanging fruit is gone, and the market is largely saturated. A quick Google search of nearly any idea will pop up at least a handful (sometimes 30+) of established competitors fighting over scraps, none of which ever have a chance of achieving Atlassian-level success. Even B2B SAAS is reaching that saturation point. The golden age of the garage tech startup becoming a billion dollar business in a few years are probably behind us.
the headline raises important questions that we aren’t guaranteed an answer for in the article; what did revenue start at? and how much additional debt (of any kind) was incurred on the way to $20M revenue?
Strange to think of, that at some point Jira was a small dream made by a handfull of real people, labeling Jira beer at the office to give devs.
It's hated software by developers.