I still remember their CEO building his original audience on TheServerSide.com, back when Enterprise Java was the thing.
They took 13 years to get to this point, a solid 8-9 years longer than most enterprise SaaS companies are expected to take (by VCs). And yet, not a single VC I've asked if they wish they were in Atlassian has said no.
You can build huge companies on your own terms and you don't have to swallow the story that everybody tries to feed you. You do need to do one thing though (and only one thing): get traction and keep it growing.
Eloqua - Founded: 1999, IPO: 2012. Years to IPO: 13
Qualys - Founded: 1999, IPO: 2012. Years to IPO: 13
Rally Software - Founded: 2001, IPO: 2013. Years to IPO: 12
E2open - Founded: 2000, IPO: 2012. Years to IPO: 12
Exact Target - Founded: 2000, IPO 2012. Years to IPO: 12
Box (Summer IPO) - Founded: 2005, IPO: 2014. Years to IPO: 9
Demandware - Founded: 2004, IPO: 2012. Years to IPO: 8
Fleetmatics - Founded: 2004, IPO: 2012. Years to IPO: 8
ServiceNow - Founded: 2004, IPO: 2012. Years to IPO: 8
Bazaarvoice - Founded: 2005, IPO: 2013. Years to IPO - 7
Workday - Founded: 2005, IPO: 2012. Years to IPO: 7
Marin Software - Founded: 2006, IPO: 2013. Years to IPO: 7
Marketo - Founded: 2006, IPO: 2013. Years to IPO: 7
I have a start up I have been working on for almost 2 years. I am bringing in money, but not enough yet to be life sustaining. So I have a full time job and working every day on this startup. It can get rather tiring...
I was questioning its worth the other day, to keep going or to wait. But I must keep going. It doesn't take a year, but rather its a long drawn out process.
The average time to an enterprise exit is 9.5 years according to http://pando.com/2013/05/21/memo-to-this-years-yc-class-its-.... So Atlassian is generally in line with the average. Talk to any person or VC that has built a large SAAS company and they will likely say the same thing.
Would every company love to grow >40% annually with spending 0% on sales and 15% on marketing...of course! The reality is that most companies can't do that. You can't look at it and say "I want to be Atlassian not Box." Box has a growth rate of over 100%, so commands a gigantic valuation. If I were starting a business I'd take either unicorn.
It is a small but important distinction, what they don't do is scale sales by throwing salaries + commissions at a highly paid sales force.
Well sure, a hit is a hit. A hit that took 3x time is still worth being a part of if your hit ratio is less than one in three. With hindsight, of course they "wish they were in Atlassian."
Both are closer to 5-6 years. Salesforce had just over a $1B valuation post-IPO, Marketo was ~$500M.
by being super specialized, the cost of sale is massively lower. no need for an army of sales people. focus is on product and delivery/implementation - once it is proven that your stuff actually works, others take notice in a tightly knit industry.
Eli Lilly, one of our customers, just got awarded for their innovation around using our products: http://www.informationweek.com/strategic-cio/executive-insig...
For context, I was comparing the origin of Atlassian and Workday. Atlassian was started by two 22 year olds through a sizeable credit card debt. Workday was started by a 65 year old ex-PeopleSoft CEO with his funds and an additional $15M in VC finance.
"Mature" would have been a better suited word. Thank you for calling me out on it.
In reality, there is a lot more needed to be successful, but Atlassian is a great example of success being 90% great product.
Upfront non-discriminatory pricing, direct purchase via credit card, immediate access to trial downloads, both installable AND cloud offerings, and source access to give customers piece of mind when it comes to solving their own problems and ensuring business continuity.
Atlassian does it right in a field where so many don't.
It feels like the universe makes sense again.
To be fair, Atlassian makes HipChat, and though it's not exactly the foundation of their valuation it is a pretty significant product.
That said, it's most of what you want from Github at much more reasonable rates. So I'm very happy with it outside of the positive network effects of forking on Github (I rarely see a lib I want to fork on BitBucket).
Saying that they're gunning for the same market seems a stretch to me. No one's going to say "I was thinking of buying Jira, but I just went with Box instead."
Leading off with drawing parallels between Atlassian and Box just strikes me as not helpful.
Atlassian is to Github as Box is to Dropbox ?
Also, Atlassian is a good deal bigger than GitHub (if measured by revenue, profit, people or valuation), which would seem to put them on the Dropbox side of the analogy
also: They really make awesome products (we use BitBucket, HipChat and Confluence)
As someone who has to use and administer JIRA I'll have to respectfully disagree. The nicest thing I can say about the product is that it was built "by engineers, for engineers". People who want empowering software that gets out of the way appear to have been an afterthought.
The backend application may very well be engineer-focused, but that speaks more to its flexibility than anything, IMO - as far as user experience, it's as empowering and minimalistic as you make it.
One of the few pieces of software I truly love.
In the real world, simplistic ticketing systems fail to meet enterprise needs, which are complex and peculiar, arising from weird contract clauses, irrational political outcomes, legacy processes etc etc. JIRA's multi-layered customisability is a huge boon in these environments.
Also, people aren't actually stupid. Yes, even non-engineers can use a complex piece of computer software! A little training goes a long way. Who knew?!
But as for the question, again not challenging, just curious:
What do you suggest as empowering, out-of-the-way, issue tracking software?
My only (but major) gripe I have with FogBugz is that it is stupidly hard to insert inline images in cases (compared to how easy it is in a github issue)..and while at it, the entire editor could use a makeover for ease of use/markdown capability
The price is steep for the most price sensitive users who probably already have experience in another tracking system. It's hard to get that user to try it the first time. Even the free version is very limited compared to other trials, allowing a total of 2 users, which probably isn't representative of the number of people involved when issue tracking is brought up.
Oh dear god.
As an employee at a medium sized company that uses Jira and Confluence, I have to disagree.
I'm sure people out there have managed to wrangle and configure Jira into something usable, but it's implementation where I work has been just horrible.
And confluence... Apart from being just slow and unreliable, it's text editor gives me nightmares. I mean, all due respect for building that (it does some pretty powerful and complicated things), but 95% of the time I just want to write an article in Markdown and post it.
Unfortunately, that's enterprise for you. Basic human logic is often insufficient to handle the demands...
Those applications are absolutely only as good as the time taken to adapt them to the needs at hand and to the team using it.
And +1 for markdown support. There are plugins out there that add that, but... don't get me started on their plugin ecosystem.
I do wish Jira had much more powerful query, and support for deeper dashboard customization. Compared to ClearQuest, Jira is very user friendly, but CQ felt much more powerful.
In the end we just scrapped it all and went with Trello. Yes, it's 'less powerful', but we've found that you don't really need much more. For all the features that Trello 'lacks', it makes up for in speed, simplicity and ease of use.
Thanks for the suggestion though. I'll give Tower a whirl.
Yes, and they all suck. They go from a pain to use (Bugzilla) to a huge mountain of useless and worthless crap with a obscene price tag (Rational tools)
It seems to me they are in the right path
Redmine ain't bad.
For feature tracking of timeline type of views, Bugzilla is fairly useless, but for just doing bugs, it does a good job.
Eg. I put a JIRA ticket number in a commit message to a BitBucket repo. Bamboo is watching that repo and runs a build with the ticket # in message. The ticket number has also already been interpreted and links straight back to JIRA where you can mark the issue as deployed.
Honestly, if they're worth $3.3 billion then the software industry is doomed. Maybe that's a bit over the top but the quality of the products based on what I read doesn't add up. How can you do so much R&D and still have a product that's so unstable and can't handle simple network hiccups?
Documentation (at least of API) is not that great. If it is available since atlassian's Confluence seems to be down most of the time. The API itself is pretty weak as well. And weird too.. getting an HTML error response on json request? Yay!
All in all, it's one of the worst tools I had to use in the recent years. It looks like a dummy product, just for the sake of it. But at least it's fast I guess. Or I don't get its purpose.
Promoting their Enterprise Advocate to Enterprise Czar (or BOFH) would be good, so that person could finally force the dev teams to think of scalability first.
They have a solid target market and a product that pretty much does what it needs to, albeit with a tonne of bloat and questionable UX principles.
you don't get to 3 billion dollars without giving away a lot of stuff for free to infiltrate the enterprise
It's hard to beat free private repos but note that they are limited to 5 users. If you have a larger team, you've got to pay.
Also don't forget, Atlassian runs BitBucket! And look at GitHub's valuation.
On the other side, my team has been bit by a bug from 2008 which prevents us from changing any ticket's state to Resolved. We can Resolve tickets, but their Resolution remains Unresolved. Since the Resolved state is central to most of Jira's built-in features, this means Jira has been broken for me since we installed it 6 months ago. The Your Issues default display shows all my tickets, including my closed tickets. The interface is not customizable enough to build workarounds for this problem.
Also, the name (bastardizing the Japanese pronunciation of Godzilla) always feels somewhat racist to me.
You can also bulk modify issues that are unresolved by transitioning them to "In Review" and bulk updating them again back to "Done" which then correctly set the resolution field. Unfortunately this technique means that all your issues are now resolved on the same day so your graphs etc show a big spike on one day.
For some reason this bug occurs after a few days with any of the default workflows in Jira, but it's not too big a deal if you spot it early and fix the workflow to resolve issues correctly.
As for the Jira/Godzilla thing, I always thought it was a reference to Mozillas Bugzilla, which I thought was kinda cool :)
Hope that helps!
Bastardized? It's the ACTUAL pronunciation, they just removed the "go" part.
It's nothing like making fun of the Japanese. I think trying to find racism where it doesn't exist has gone too far.
* Create a copy of your workflow
* Add the outgoing transitions that you want to add
* Change the workflow scheme to point at the new workflow
It's not super elegant but if you need to work around it you can.
Feel free to connect with me on twitter (@jsinkers).
That is extremely reaching, like on a hole new level. Do you get this offended in real life too if a coworker makes a joke? Genuinely wondering.
You can then bulk transition all your existing issues and the resolution will get populated correctly.
This is a company that is learning too. Since the acquisition of HipChat they have made the product dramatically better. If they can get a good model in place for further acquisitions, and they can continue to successfully launch their own products as well as they have in the last few years, then this company is on track to being worth a hell of a lot more than $3.3bn.
JIRA - For issues
Bitbucket - For code repository
Confluence - For documentation
Bamboo - For a build server and CI
I'd love to settle on HipChat as well but it's stupidly difficult to get it to pull git commits and other things directly into chat.
Flowdock is much much simpler in usage, UI and integration - so Atlassian team, if you're reading this: Let me 1-click integrate git commit notification into a HipChat room of my choosing, you own both products - make them connect in the easiest most straightforward way!
> 1/3 of the top 100 are SaaS, but not all are purely enterprise.
Well, I say Microsoft makes more profit out of enterprise then Facebook makes out of consumer. Does this prove anything? Not at all.
i.e. if Instagram had a valuation of $800 million the $1 billion purchase price could make sense if Facebook believes their combination could produce the difference in price.