Tech startups: cross-functional people or cross-functional teams?

Today was the culmination of the business plan class (ALS 458) here at KGI, with the final presentations by 6 teams — some of whom will be going on to formal business plan competitions elsewhere. So it was the day of the year that the students, and invited guests most celebrate (and ponder) the nature of tech startups.

Our students are unusual in having both science and business training: they come with a science (or engineering) degree, they take science classes, and they take business classes. So in effect, they are cross-functional individuals, with a little bit of knowledge about a lot of things in their heads. Similarly, a company like HP used to pick their marketing staff from among engineers who later got an MBA.

This is also how schools like Stanford and Berkeley set up a mini-business school (or “engineering management” program) within their engineering school. And it’s also why MIT recently set up its Engineering Leadership Program for undergraduates.

On the other hand, a number of schools run business plan classes and programs by assuming individual specialization and deliberately mixing the various specialists. The NSF-funded Georgia Tech Tiger program is perhaps the best known such program among those of us who teach tech entrepreneurship. To some degree, this reflects the supply limitations — you can’t get enough cross-functional people so you merge silo’d programs (with silo’d students) onto a temporary cross-functional team.

Obviously any good tech idea needs to be brought to market through a combination of technology, marketing, finance and manufacturing (or other operations) skills. How do you build such a team in a real startup? And who should be in charge?

I’m an engineer who went into B2B sales and marketing, so it’s easy to guess where my sympathies lie. And at a recent MIT club event on the “Gordon-MIT Engineering Leadership Program,” I heard veteran tech CEOs talk about how it takes a technical person to lead a tech startup.

Still, there are many counter-examples. For every Larry Page, there’s at least one (maybe more) Jerry Yangs.

Steve Jobs offers another model. Sure he was a great marketer — one of the best of the 20th century — and a great CEO. However, if you look at the recent Jobs biography, it was clear that his mechanic father and his childhood electronics experiments gave him an intuition about engineering design that many practicing engineers lack. (Alas, as the original Mac 128 death march demonstrated, he also had completely unrealistic expectations about how long things should take.)

So how do you form a cross-functional team to make the next great tech startup? And how do you allocate decision rights and authority among them? How does this change over the life of the firm, the industry and the technology? And what do you do with your hybrid business-engineers (or business-scientists)? They’re not going to be CTOs or CSOs (are they?), but do you put them as VP of R&D, or division managers, or CEO?

These are all interesting questions. Perhaps someone will research these answers.