Along the bottom here is the aging of the venture. Over at zero on the left is when it is founded and then as it is playing out, as it is aging overtime. As we go higher up on this, this is the stability of the founding team. At founding all the way on the left 100 percent of the founders are still there. And then as the ventures age that’s where the founding teams start losing people.

One of these lines is teams that founded with friends. The other line is teams that didn’t found with friends. Which do you think is the higher line on here, the friends team, or the non‑friends team? The friends teams, after an initial honeymoon phase that they go through where the two teams are relatively equally stable, the friends teams being less stable significantly from a statistical perspective. And also as we get further down there of the remaining teams this is a very significant difference between them. And to the extent there, that this is an example of where the most common of decisions is one that is most fraught with peril.

This is my data on how often are people co‑founding with friends versus these other profiles. And by far the biggest bar on this chart, 43 percent, is co‑founding with friends. If we add on the family side, add those two bars together that’s more than half of the teams. And so it’s one of those things to be thinking about. Why do we look to our friends as a source of founders? Why do we go and we head down the path of punting on these issues when they are so dangerous for our most cherished of relationships? The family teams by the way are even more extreme than friends ones in terms of the stability of it.

The most stable of the teams are co‑founding with your prior coworkers. Friends teams, friends and family teams that we combine them, are even less stable than co‑founding with acquaintances, co‑founding with relative strangers. Also think about with our friends, about assumptions we’re making. We assume we know them. We assume we trust them. We assume we will be on the same page. All of these other things that go in we’re not going to test those assumptions within the team. Versus if we are co‑founding with acquaintances we know we don’t know each other. We’re going to tee those up. We’re going to make sure that we have more of that getting to know you as professional coworkers within that team. And so several things that we are handcuffed by when we are co‑founding with that very common set of possibilities looking at the friends and family as the people that we’re going to be co‑founding with.

And one other element that we haven’t talked about on the friends side is an assumption that we’re making about the skills of our friends. How well do we know our friend and how well that person works? Are they going to be able to scale? Are they going to be up for it? Telling your best friend he’s not scaling, he’s going to have be demoted, going to be have to be replaced.

Let’s get a little bit more granular about some of these challenge that these types of teams face. So the first of those is the likelihood that we’re going to discuss the elephants in the room, those conflict‑ridden issues, those tension‑filled things that we are going to have a tendency to punt on. Is that going to differ by these types of teams? For me the key piece of this is the left‑hand side of this slide. This is going to be where the elephants in the room are going to be least likely to be teed up. Second element, the damage if things blow up, if things blow up within the relationship, in the venture, repercussions on the outside. What is this going to look like across these different types of teams. To me the divergence between the two of these is something that I call the playing with fire gap. So the critical thing is when we’re dealing with teams at that left end to keep in mind both sides of the playing with fire gap and find ways to reduce that gap, find ways to push up that likelihood of discussing the elephants, push downwards the repercussions on the personal side for if things blow up within the venture.

If we can harness the potential glory of the left side here, to me the left side here is the high variance, what economists call high beta in terms of where it can be. It can either be the most glorious of teams or it can be the most disastrous of teams. And by hopefully going and creating these firewalls, forcing those conversations, reducing the playing with fire gap that’s where we can hopefully cut off the bottom end of that distribution so that more of those friends and family teams can see that glory happening.