Meet the CTO is a new series about CTOs, their daily lives, roles, and responsibilities. In our first edition, we talk to Morten Primdahl from Zendesk.
What does a typical day look like for you?
My day is pretty diverse. We’re a rather large company now, which means I get to see a lot of different things. For example, yesterday I was at an event and talked to 70 CEOs from the Point Nine Capital portfolio and some other friends, talking about my journey over the last 10 years with Zendesk. Later in the day I was talking technology with a customer of ours, and ended the day with a CEO dinner for the Heureka conference. So, yesterday was very atypical day.
Next week, I’ll be back at the office in San Francisco. Being back with the teams and working with the infrastructure teams, working with the security teams, back in the grind. Such days are filled with email, correspondence, communication, trying to help people stay online. That’s pretty much my role today.
What motivated you to start working at Zendesk?
This was 2004/05, my early thirties, and I worked on some customer service solutions for the enterprise, specifically on a product called Remedy. I worked for a German company called Materna, which had an office in Copenhagen. We were selling these big systems and they cost a million dollars. A million dollars to host on premise. A million dollars to operate, and it was still just a really expensive sales model where the salesperson goes in, sells to some executive, and he’s out of there. This was bad for many reasons. It was complicated to use, lots of input fields, lots of buttons, and had a horrible UI, bad experience from start to end.
That was basically the motivation. How do we build something simple, elegant, beautiful and cheap that worked for small companies.
What interested you about technology? Was there any moment in your childhood where it clicked?
There’s been a couple of moments. Early on I was playing with Commodore 64. Like every other kid, I would record on the tape player signals over the radio, and that would turn into a game. That was fascinating. The local radio stations actually broadcast a game that I played. That’s how technology first got into my life.
Later when I was starting at university, I studied biotechnology because I felt that computers are more of a hobby. That was a mistake obviously.
I was very passionate about computers. I just didn’t realize how to transfer that into a university career.
But I did switch from biotechnology to computer science later on, and did my masters in that eventually. That’s how I ended up. Technology’s really for me, and it’s something I’m passionate about, and just embracing that.
Do you have some tips for people getting started in web development?
I think there are many different kinds of people. I don’t think everyone needs or should do a university degree necessarily. Things get very academic and very technical if you start computer science; whereas, doing web development today it’s more of a practice.
You’re more like a carpenter than a neurosurgeon.
You’re more of a practitioner and you follow established practices, and you build some products. You don’t need to know everything about compilers and kernels to do that. It takes a long time before a company meets that level of competence. If that’s what you’re passionate about building, then you don’t have to start through all the hardest algorithm courses of the craziest university.
How important are college degrees when you hire new developers?
Two of my very best engineers have no university degree. Even at our scale today, with more than 800 people, I don’t judge a book by its cover.
If you come in and you blow minds on the interviews, and the good guys always do, you just know right away when you’re hiring someone excellent. I don’t care about your degree at all. If you’re passionate about building software, and you can add to the great team we have at Zendesk, absolutely we’re a match.
If you had to decide between “pursuing your startup idea” and “going to college” – which one would you choose?
The reason that our generation and my generation in particular, all got degrees, is because our parents come from a generation where they could not all get degrees. If you get a degree, you were pretty much guaranteed to have a pretty stable life.
My parents knew about the oil price, and their parents knew about starvation and the failing harvest. That’s why most people of my generation got a degree. Today, I think the world’s slightly different. I wouldn’t recommend one over the other. It depends on what you feel comfortable with in your gut. If you want to build companies and you can feel that, I don’t think you need to get degrees to do that.
Can you share some challenges you’ve faced during the journey of Zendesk – from startup to public company?
Today I’m talking at the Heureka conference about the challenges from an engineering perspective. I get asked quite a lot about, “So, you went public, then what happened?” The truth is, not much happened because the preparation in engineering was actually all in the years before that.
We went from a scrappy start-up of 3 guys with everything in their heads to a big organization that can take on scalability, operations, databases and product development. If you really get passionate about business in your life, that’s where you need to go. When you grow from a couple of dozen guys in 2009 to the size we have today, the company changes a lot all the time, and you need to adapt to that.
The challenges are both technical and organizational. Scaling your business means growing fast and hiring engineers. And that’s a big global challenge. To put it that way: It’s just hard.
So it’s more about organizational and cultural challenges than technical ones?
As a founding CTO, you see everything. We’ve definitely had lots of severe scalability challenges we had to solve as well as architecture problems. But dealing with people and culture is very hard. You go from a place where everyone is involved in everything, you communicate broadly, and you move to a more need-to-know basis because all of a sudden your financial data becomes sensitive as a public company.
When you got your series A funding you moved to Boston. What was the reason for Boston?
2008/09 is a very long time ago, obviously. We had spent some time looking for funding for a while, and we couldn’t find any initially, so we had friends and family in this.
Christoph Janz from Point Nine Capital was an initial investor, and that was 2008. We were talking to various investors on the west coast as well, and then the credit crunch happened in 2008. That means that everyone who dealt with money got very conservative all of a sudden.
Then, out of the blue, 2009 or late 2008, Devdutt Yellurkar from Charles River Ventures, who are based in Boston, flew to Copenhagen to meet us. He spent the day with us in Alexander’s loft. Sitting there having coffee with us around the work table turned out to be a series “A” offer. They’re based on the east coast, in Boston, and of course it made sense for us to move to Boston. We used their network, and they could see “These guys are coming into the country and they’re going to do well here.” That’s why we started there. We didn’t stay there for long.
I think it was just a couple of months, right?
Yeah, it was. The moment we moved our intellectual property legally into the US, and we committed to move to the US ourselves, we got a few series “B” offers. We decided to take one of those. That was a west coast company, so we decided to move the company to San Francisco.
If you could travel through time, just from the beginning of Zendesk to nowadays, what time would you choose?
I think the sweet spot for me was just after series A. Getting Alex here was really hard. Funding was very hard to get back then. You didn’t have the luxury to build things right from the get-go because you’re starting so small. At that point you have a commitment and some money to start building a proper organization.
With the knowledge I have today, I would have tackled some things differently. I would like to be in that position again where you still have the open opportunity ahead of you, and you have a decent foundation financially and the backing of the VCs.
What advice would you give yourself as a 14 year old?
It’s going to be all right. Do what you like. Do what you’re passionate about. Don’t do whatever anyone else thinks. Figure out how to be yourself, and beat your own path. If you want to be an artist, be an artist. If you want to be a painter, be a painter. If you enjoy computer programming as a 14 year old, you’re going to be an amazing engineer later in life. I’ve never met a 14 year old who started programming that early and who didn’t turn out to be a completely amazing engineer.
So you think it’s ok if you learn how to code later on in your life?
Your door’s wide open at 14 and 24, 34 and 44. It’s never too late, I promise you. We started our company, we were all in our 30s. I was 32. The other guys were a little older than me. That’s pretty late for a startup, but you’re mature at that age, and you can accomplish something obviously.
Looking into the future, what challenges do you think we will be facing in software engineering?
I think the demand for talent is going to keep rising. I think more and more engineers will realize that there is an opportunity to be an entrepreneur. And it’s going to be even harder for well-established companies to hire good people. I think that’s a very real challenge.
What is your personal motivation to build software?
When I was in high school, I really liked math, because you sat down and you had a complicated equation or diverse question and all of a sudden they were solved and you’d be like “Pi squared, and whoa it’s amazing, a beautiful result.”
It just felt great. Later you get that success feeling when you do Sudoku, “Oh, it adds up.” It feels fantastic. That’s pretty much what software engineering is. You have a problem statement. You come up with a solution. You know the solution’s efficient or you know it’s inefficient. You know something about the solution, and you have the result, which is like instant feedback.
I think that gratification, getting that fix 10, 20, 30 times a day, feels good.
And how do you handle testing at Zendesk?
We have so much testing going on. Early on testing was not a priority. We were 3 guys trying to survive and Alex and me were doing the coding. We have this terrific graph that shows the evolution of the number of lines of our unit tests and how we used up less for unit tests than application code. Now we’re up to twice as much. We do unit tests, virtual tests, situation tests, functional tests. We deploy many times a day.
We have more than 94,000 paid customers that we need to cater to their needs, we can’t just interfere with their business. We have a great QA team that tries to get a human angle on it. Do some lightbox testing. Do some automated testing there. It’s a big challenge. And we invest heavily in it.
Can you share some tools you’re using in your tech department?
We use a wide array of Saas products, like Travis CI. GitHub, of course, since the early days. We’ve used New Relic for many, many years. Early on New Relic was completely a mind blowing product, now they’re more competitors in that space. We use sublime text for editing. And we’re all on OS X.
Any communication tool for staying in touch?
We use Flowdock. Most people tell us: “Use Slack”, but we’ve just used Flowdock for longer. We use a lot of travelling to bring people together because face-to-face time is really valuable. Even though we have Skype and Google Hangouts and Flowdock, we try to bring people together a few time a year. When you join the company, you spend two weeks in San Francisco and meet the team, and get to understand the experience.
One last question: Can you share some influential resources you have read?
Obviously, I recommend my co-partner Mikkel’s book “Startupland” where he tells his story as the founding CEO, and some of the challenges he went through. It’s very inspirational. There’s some good, fun stories in there. It talks about the dynamics in an early stage company.
On the technical side, I’ve come to reading “Big Data” by Nathan Marz, where he elaborates the architecture and how you provide stream processing and batch processing for real time analytics. It’s super fascinating. I also read the confluent.io blog by Martin Kleppmann.
Is there one thing you really love about your job as a CTO at Zendesk?
Talking to people like you.
Nice try 😉 And besides that?
I get to work with the smartest people I’ve ever met. I really have, and I know everyone says this, but I really have the best group of engineers I’ve ever met in my 15 years of career in software development. I work with the smartest people. It so inspiring and so humbling at the same time to see the problems we get to solve. I love to see the diversity of problems. I get to touch on product issues, on database crashes, communication issues. You just get to learn so much, and drink from the fountain of knowledge day in day out.