For the past 5 years I’ve been a startup CTO (Chief Technology Officer). Before that I worked in R&D within large companies directly under the CTO. Even with a decade of technology experience, I still find exactly what the responsibilities of a CTO to be a bit fuzzy. In this post I will describe what I think a CTO should and shouldn’t do. More importantly, I will try to answer why a startup should (or shouldn’t) have a CTO.
Let’s start with what the primary responsibility is. The CTO owns the company’s technology vision. Just like the CEO is responsible for formulating a big, hairy, audacious goal into a compelling business vision, the CTO must create a similar vision but around technological goals. This vision then forms a roadmap for all of the engineering and scientific staff — a goal that guides their day to day decisions.
A technology vision isn’t an architecture, design document, or platform. Those are concrete manifestations of the vision. The vision is at a higher level. Perhaps the most famous technology vision is Moore’s law — that the number of transistors on a chip will double roughly every 18 months. That is a compelling vision and as an engineer helps you decide exactly what chip design and process improvements your should focus on.
At Distil, our business vision was that games should be as easy to make as a powerpoint document. This led to a technology vision of WYSIWY game creation. For example, when playing a game you notice a spelling mistake. This can be corrected right in the game without launching a separate tool. This led to us investigating how tools like powerpoint and word handle document rendering and editing simultaneously, and then the creation of an XML based game “Document” that can be read and edited by a game “viewer”. And all other technical decisions flowed from that vision.
Next, the CTO is an outward facing role. The CTO is NOT the top engineering manager — that is VP of Engineering. The CTO is also NOT the company’s best engineer. The CTO is the best person at communicating complicated technology to people who are not technical, and getting them excited about the possibilities.
As a CTO I spend most of my time talking with customers, understanding their problems from a technical perspective (my co-founder does a similar task from the business perspective), and then help them understand a bit about our technology.
I also spend a good amount of time dealing with IP issues. IP issues come in two forms: protecting our existing IP and finding IP that can solve our problems. As CTO I need to determine what technology we want to be the world’s best at. I work on our IP strategy. I decide which technical problems we should invest in solving (since they will give us a competitive and IP advantage) and which we should look to license or solve using open source software.
Being a startup much of the IP I develop is done in partnership with University groups. This is a very cost effective way to do applied research that results in IP we own. So I spend time getting to know University professors and the technology transfer departments and the Universities. Just as a CEO is expected to have a deep network of commercial and investment partners, a CTO will have a similarly deep network of researchers and IP partners.
So does your start-up need a CTO? As the technical founder should you call yourself CTO or VP Engineering or Lead Developer? Of course it depends. In the early days at a startup I’m doing all 3 jobs. The answer depends on the future vision for the company.
If my company is building a web or mobile app, there is probably no need for a CTO. Almost all customer interaction is occurring through a website or customer support. There is little need to work with customers to understand their technological problems and limitations. Chances are my product is competing based on a feature set or design optimization. There is probably little need for proprietary algorithms running under the hood.
But if I’m building a company that will disrupt an industry — change how Hollywood movies are made and distributed or reinvent email — then a CTO is a necessity. Obviously there is some technical miracle that needs to occur to make the business vision happen, otherwise this disruption would have happened earlier. It is the CTO’s job to find this technical miracle, articulate it as a compelling technology vision, and then develop & protect the essential technology that fulfills it.
It may seem pretentious calling yourself CTO when your are just 4 people hacking together in a room. But doing so helps clarify the overall vision for the business. It also helps remind you that at some point you will stop being involved in the day to day engineering and become almost exclusively outward focused.