Why Elon Musk’s “People as Vectors” Analogy Resonates

In his keynote at INBOUND last year, Dharmesh Shah (Hubspot’s cofounder) shared a personal story of meeting Elon Musk and their short but powerful conversation about aligning people on a team as you would vectors in an equation. Unlike many folks in the startup world, I’m not a die-hard Musk fan. I appreciate his creativity and accomplishments, but also find him to be frustratingly quiet regarding political, social, and moral issues on which I wish he’d speak up. That aside, I find the analogy from his chat with Dharmesh something I return to over and over.

The basic concept is simple:

  • We can think of the people on any given team as vectors
  • People vectors have both direction and magnitude
  • When people all work in precisely the same direction, their magnitudes are added to each other
  • When people have any degree of deviation, the varying directions subtract (at least somewhat) from the maximum amount of productivity that could be achieved

Yes. It’s overly reductive. People are not merely “vectors” (thank goodness). But, in so many ways, the mental model delivers a beautiful degree of clarity.

Let’s say I joined together a dream team of startup founders — for this fantasy and the illustration below, we’ll make it Courtland Allen (of IndieHackers), Tara Reed (of Apps Without Code), and Arlan Hamilton (of Backstage Capital) — to build a new product.

large version of the graphic available here

Arlan is our standout contributor. She does twice the work that I can (if you follow her on Instagram, you know this isn’t a stretch). Courtland and Tara are both at 1.5X my productivity. And, if we’re all working together toward exactly the same goal, with 100% alignment, we can total 6 units of productivity (assuming that I’m 1 “unit”) in some arbitrary time period.

But, say we’re in a sub-optimal (and far more common) alignment scenario. I believe we should be doing things differently and am working to see my vision through to fruition and trying to nudge the team to my perspective. Arlan is ignoring me and continuing on the original path. Courtland is slightly swayed by my input, and has adjusted his trajectory somewhat toward mine. Tara heard my ideas and found them to be so poorly-thought-through and illogical that she’s actually moved slightly in another direction such that the two of us are near-diametrically opposed. Despite everyone contributing their maximum capacity, our total production toward the goal is now only 3.5 units.

In the latter scenario, I’m literally worse than useless, dragging team productivity off course to a greater degree than if I weren’t there at all. No wonder so much business advice in the last 20 years has focused on (to use Jim Collins’ words) “get the right people on the bus and the wrong people off the bus.”

Those of you who read Dharmesh’s post might be wondering why I’m re-hashing his points… It’s because I wanted to share three additional thoughts on the analogy. They’ve been circling in my head since I heard it at INBOUND, and for me, simultaneously temper and add valuable applications to the concept:

#1: What Does 100% Alignment Mean?

Let’s get something out of the way — 100% alignment doesn’t exist. Not even for one person. Not even for Musk himself (though I’ll admit his single-minded dedication probably comes close). We are people, not automatons, and our wonderfully human imperfections, emotions, and biases drag us off course. The most single-minded, dogged pursuit in the history of humanity is still not a straight line. Thus, it’s impossible that any group of people could keep their own efforts flawlessly parallel on a project with any modicum of complexity.

Thankfully, this reveals a beautiful truth: we can *all* benefit from greater alignment.

But to do that, we need to know what “alignment” means. Dharmesh included three vectors of alignment in his post:

  1. Align people with the organization’s goals.
  2. Aligning individual teams (product, marketing, sales, service, etc.) with the organization’s goals.
  3. Aligning the organization’s goals with the needs of the customer.

I agree with those in the abstract, but found them missing some of the elements I saw in my own experiences. Those included consensus (or as much agreement as possible) on the answers to a few fundamental questions:

  • Why are we on this team together?
    We could work at many different places, on many different projects. Why this one? Why now? Am I here for the same reasons you are?
  • For whom are we building this?
    There are a plethora of potential customers/consumers. Do we agree on who will (or should) benefit from this? Are we in agreement on who’s most important to support and whose needs are less crucial?
  • What are we creating?
    Not just in the abstract, but in the highly specific details. Do we share a belief in what the best, specific thing that could come out of our efforts is? Would we put the same importance numbers next to the same priorities? Do we agree that we could ship without “A” but absolutely not without “B?”
  • Do we have shared respect, trust, and empathy for one another?
    When we feel like someone else isn’t holding up their end of the work, resentment festers. Is Arlan OK with Rand contributing only half her capacity in our project example above? Does Rand admire and cheer for Arlan’s progress, or does he envy her abilities and (knowingly or not), feel anger and fear as a result?

If you’re reading these questions and thinking “no one could ever agree on of these 100%!” you’re right. And that’s OK. We strive toward alignment, knowing that it won’t be perfectly achieved. The striving is what matters. The perfection doesn’t.

These questions also reveal a few powerful truths:

  • In smaller groups, greater alignment is easier (two pizza teams?)
  • Accepting that imperfect alignment is inevitable can make us more prepared for and more comfortable with disagreement
  • Many times, we don’t ask these questions out loud… which can lead to us silently wondering, or worse, making assumptions

When I reflect on some of the best, most successful projects I worked on at Moz, I’m sad to say that we never openly asked each other these questions (at least, not formally). But I do recall the immense feeling (right or wrong) that we shared most of the answers. Perhaps that was just an assumption on my part, but, to some degree, it worked.

#2: What Brings People Into Alignment, and What Pushes Them Apart?

I don’t know that I’ve been exposed to the full range of alignment levels in my career. But I do know that I’ve seen a lot of variance. Common threads bind together those experiences, and when I’ve tried to optimize this (since hearing Dharmesh’s talk), here’s what I’ve noticed:

  • First trust, then alignment – in most groups (though not all), there’s a high degree of initially assuming questionable intent that fades after a (variable, depending on the person) time. Stereotyping happens here, too. For a white dude, founder in his late 30s with a degree of (historic) prominence and power in an organization, I found that some teams and people assumed I’d be a power-hoarder, that I’d distrust ideas that weren’t my own, and that I’d have sexist, racist, and/or ageist tendencies. I had to consciously fight against those, sometimes for long periods, to win that trust. And I know that with some people, I never succeeded.
  • Values > Goals for some people, Goals > Values for others – I’ve worked with two kinds of people; those for whom a commitment to shared values (agreement on these things) could trump disagreement on many of the questions above (i.e. we may disagree on the what, but we share the why, and that’s good enough for me). And those for whom goals alignment outweighed values (i.e. we may be at opposite ends of the ethical/social good/political spectrum, but we share a product plan). Since I’m in the former group, I struggled to make things work with those in the latter group personally, but know folks who did reasonably well with it.
  • Psychological Safety – the science of dozens of studies all point to the same conclusion, and my experiences matches. To quote Amy Edmondson “(successful teams have) shared belief held by members that the team is safe for interpersonal risk-taking. (There is) ‘a sense of confidence that the team will not embarrass, reject or punish someone for speaking up.”

These are investments teams big and small can make, and if it can multiply the output of a group of people, without costing money or adding friction, we’d be somewhat insane to ignore it.

#3: Why Emotional Buy-In Matters, and Why “Disagree and Commit” Fails

The last element I’ve been reflecting on is the peculiar alchemy that happens when a group shares more than just the answers to the questions above, but a tangible, vivid enthusiasm for the what and the why. My final project at Moz (one to which I’m still contributing after my departure and at least until public launch) has a pervasive, unquenchable element of this binding the team together. The whole feels like more than the sum of its parts. Difficult problems feel surprisingly surmountable. People volunteer to take on more than they can chew, at a speed that’s faster than other projects, and somehow they pull it off.

I think harnessing that emotional buy-in is something every team should strive for, and early-stage companies doubly so. If you’re not truly excited about the work, in the way you might get excited about other big, potentially wonderful parts of your life, it’s hard to have the same impact. Perhaps, in the mathematics of the vector model, emotional excitement serves as a multiplier.

Conversely, I heard the phrase “we can disagree in this room, but once we leave, we need to all commit 100% to the group’s (or the CEO’s) decision.” When I was CEO, it frustrated me deeply that this was never truly possible. When people “disagreed and committed” it was never with the enthusiasm and focused force that “agreed wholeheartedly and committed ecstatically” carried. When I became an individual contributor, and disagreed myself, I discovered exactly what it felt like to be in their shoes, and I, too, couldn’t (even when intellectually and for the sake of my stock’s performance I wanted to) unleash the full force of my efforts.

Disagree-and-commit sounds rational. But it never, in my experience, achieves the same magnitude in the vectors of people who leave the room after “disagreeing and committing.”

My takeaway is to work hard in the future to: A) find people who agree B) find missions/visions/projects on which we agree C) highlight the agreements and D) consider whether those whose disagreement is at a fundamental level (i.e. I don’t think we should create this) vs. surface level (i.e. I don’t think we should use this provider, but we can always swap to their competitor if it doesn’t work) are the right ones for the bus.

And yeah, I am still trying to reconcile this with a contradicting concept in which I also deeply believe: that a group without disagreement (or in which people are afraid or unwilling to disagree) is a dangerous thing.


Thanks to Dharmesh for sharing this and to Elon for the original idea. Whether in a team of two people or two hundred, it’s served me wonderfully as a way to evaluate team cohesion and potential.