I’m incredibly proud to be able to celebrate the 10th birthday of Puppet Labs.
When I started this company in 2005, infrastructure was boring. One of the main struggles was convincing people — developers, investors, sysadmins, my parents — that what I was working on even mattered. They kept asking, “Why bother?”
Well, no one’s saying that anymore. So much has changed, and so much of that change makes it clear that today, every company is a technology company. Everyone we talk to is trying to move faster, while somehow spending less money.
Growing an Automation Company, Growing a Market
People always ask me, “Has the company played out the way you planned?” or, “Have you grown as quickly as you thought you would, in the way you expected?” etc. Hah! That would imply I had a certain level of planning when I started. I did have three related goals, but really no idea how I was going to accomplish them:
- Build software that makes its users better at their jobs, and more fulfilled doing them. Specifically, build automation software that would enable sysadmins to automate the menial, repetitive parts of their work, and reduce inconsistencies and outages, so they could focus on long-term, larger and more interesting work.
- Enable companies to move faster without sacrificing stability or security, turning technology into a competitive advantage.
- Drive adoption of the best technology throughout the industry by reducing the friction of switching to newer and better technology.
Puppet Labs started as an open source software company because, really, I didn’t know anything else. In its early years (when it was mostly just me), the company was really about understanding the shape of our users’ problems, and building a coherent product to solve them. Along the way, the community of people using Puppet really became the bedrock of our success. They kept us honest, and helped guide us to build the right products in the right way.
Then our next challenge was, how do we build upon great open source software and a healthy community to create a software company that survives on the value its customers find in its software — while still maintaining the open source software, supporting its users and at the same time, advancing the commercial product? How do we scale the company, the community, and the product?
We have found that Puppet’s balance of simplicity and power has scaled really well. We continue to evolve our product suite as our customers hire it for more complex jobs, and we add more functionality to enable that — and still, the focus on being simple enough to be used by most people, while being powerful enough to do most jobs, has been excellent. As we grow in scope, we’ll always seek to maintain that balance, and we’ll always eschew creating fast but insufficiently powerful tools, or crazy-awesome but confusing tools. We’ve made some missteps, and our customers have made these clear — and that has helped us build something better.
Now the challenge in front of us is, how do we move from just 15 percent of the market using automation (yes, it’s true) to 50 percent? That’s the big hill in front of us today. It’s not about taking share in the automation market, but about changing the expectations and habits of the market so that automation itself becomes a default — similar to how VMware has taken the market from essentially 100 percent physical machines to just about entirely virtualized over the course of its growth.
What’s Changed in the IT Industry over the Past 10 Years
One of the most important trends we’ve seen is the move away from outsourcing IT, and toward companies trying to turn IT into a competitive advantage. In the early 2000s, many companies thought, “Hey, I’m not a technology company — I’m a media/retail/finance/etc. company,” and partnered with someone to make the tech problem go away.
By now they’ve all found that, actually, they are a technology company, and technology has to be a core competency. Now they’re bringing it back in house, all while trying to build up a brand new skillset. This has resulted in a lot of churn among vendors, and lot of reorgs at our customers, but also a strong commitment to doing things differently from now on.
We’ve also seen how important it is to focus on getting software into production as quickly as possible. Companies have been relentlessly finding ways to shorten the time between a developer’s keyboard and production, through smaller process, more automation, heavy use of virtualization, and much more.
Speaking of virtualization, the last 10 years has seen it move from merely interesting to the de facto substrate on which much innovation and change is done. To name just two big trends, both AWS and Docker are entirely enabled by the pervasive adoption of virtualization, and the rebuilding of everyone’s workflows around that. When I started this company in 2005, both would have seemed inconceivable (even though AWS was launched just the next year).
My Three Takeaways from Puppet’s First Decade
The first and biggest thing is, yes, you really can build a great software company by building software your users love, even when you’re selling to the enterprise. I couldn’t commit to a company that didn’t stay focused on this, so it’s kind of important to me personally.
Second, customer success is everything. It doesn’t matter what you want to sell, it matters what your customers want to use. We now have an amazing organization built around making sure our customers don’t just buy, but that they actually use our software to succeed at what they hired it to do.
Lastly, what’s more clear than anything is that this business is a marathon, not a sprint. It’s not just about getting people to adopt a new tool — it’s about retraining every sysadmin; rebuilding workflows around software deployment; rebuilding buyer patterns; dramatically changing expectations for the CIO; and getting the rest of the technology industry to realize that what they build will be managed by software, not by people. We’re proud to be the largest pure-play IT management company, to be focused on bringing automation to the masses, and we’ve got clear eyes for how big the job really is.
Puppet has grown from nine employees in 2011 to more than 360 employees around the world today, with offices in Portland, Belfast, London, Pilzen and Sydney. I want to say “thank you” to so many people who helped us get here — from the earliest employees, such as Teyo Tyree, James Turnbull, Dan Bode and others; to our first believers, like Digant at Stanford; to the people who have believed for so long, like Nigel Kersten, Jeff McCune, Scott Campbell and Eric Sorenson; and to everyone now working at Puppet Labs. We’re still here because you wake up every day, ready to attack the hill in front of us.
While it’s been a long road, the thing I still love best is when someone comes up to me (usually at one of our events) and says something like, “Puppet changed my life. I love my job so much more now, and it’s all because of Puppet.” Nothing we’ve done would be worth it if our users didn’t feel this way.
I’m excited to see what the next 10 years will bring.
Luke Kanies is the founder and CEO of Puppet Labs.