A different look at building the right corporate culture

Bridgewater rates employees in real time; Toyota lets any worker stop the assembly line. Both win wildly. Why copying famous corporate cultures is a trap — and the three questions leaders should ask instead.

company culture
Photo: stock.adobe.com
Share this article

Consider two corporate cultures that could not be less alike, both of which win. At Bridgewater, the world’s largest hedge fund, colleagues rate one another to their faces during meetings, in real time, and the meetings are recorded. On a Toyota line, any worker who spots a defect can pull a cord and halt the entire factory.

One sounds merciless; the other almost tender. Both are deliberate, both have been studied for decades, and both have produced extraordinary success. So, if one were to try to create an ideal corporate culture, which model would be best to follow? 

This spotlights the real question leaders need to ask. The question is not whether a culture is humane or harsh, flat or hierarchical, candid or polite. It is whether the culture fits what the company is actually trying to do.

Culture is less a set of values you choose than a set of behaviours you need. Get the need right, and even an extreme culture can be correct. Get it wrong, and the finest words on the wall will change nothing.

What a company needs comes down to three things: its mission, or what it is optimising for; its tasks, or what the daily work really demands; and its people, meaning those it has, those it can reach, and those it hopes to attract.

When you read the famous and oft-studied corporate cultures against those three questions, they stop looking like simply philosophies to admire, and start looking like sensible answers to particular problems.

Take mission first. Netflix optimises for rare, enormous creative hits, where one exceptional person is worth more than ten adequate ones, so it guards talent density with its keeper test and parts generously with everyone else. Amazon optimises for scale and the customer, so it sets a high bar, replaces slide decks with six-page written arguments, and tells its people to disagree and commit: argue hard, then row together once the call is made.

Closer to home, DBS decided its mission was to survive the digital age rather than defend the branch, and rebuilt itself around experimentation and hackathons, behaving, as its own leaders liked to say, like a startup of 20,000+ people. In each case, the appropriate behaviour follows and supports the mission.

Then the tasks. Where the work is high-stakes, and the cost of being wrong is enormous, candour has to be brutal. That is why Bridgewater rates people in the room, and why McKinsey writes an obligation to publicly dissent into the job description. Where the work is safety-critical and repetitive, you push authority downward instead, which is the whole point of Toyota’s cord: The person nearest the problem must be able to act on it.

Different tasks demand different, and sometimes opposite, instincts, but each is right for the job.

And the people. A firm drawing on a deep, replaceable pool of ambition can afford to be ruthless. Jack Welch’s General Electric ranked its staff every year and cut the bottom tenth, and could do so because there was always another eager hire at the door. A firm whose edge is a small team that is hard to replace cannot.

For it, loyalty and safety are not sentiment; they are strategy. Who you have and who you are trying to attract sets how hard you can push.

Which brings me to our own firm. The culture that currently exists at Quantedge is not one that we picked out of studying the cultures of other firms. It can be traced back to those same three questions.

Our mission is to compound capital patiently over decades, and a culture in a hurry cannot hold a position through the time it takes to be proven right, so we are, by necessity, patient with our people and our goals. Our task is to let a systematic process do its work, which means evidence must be able to beat status, and the model must be allowed to override human instinct. That requires ideas to surface without fear, and stoic self-discipline to hold the course through brutal drawdowns. And our people are not replaceable on a whim.

In any case, radical transparency of the Bridgewater sort is not quite the Asian way. So we chose acknowledgement of experience over brutal candour, trust, and psychological safety over surveillance, and we bind it all together with ownership.

The employees of Quantedge are invested in the same single fund that we’ve built together, which means disagreements focus on the fund we collectively manage rather than personal trifles. We row together, regardless of rank, desk or function.

None of this makes our corporate culture superior to any others. It makes it ours. A firm with a different mission, different work and a different talent pool would be right to build something far harder than we have, and we would be foolish to copy ours onto it, however much we admire it. There is no best culture. There is only the one that fits the job.

And every fit has a price. Patience can lose an organisation to people who are inherently in a hurry and seek faster gratification. Trusting freely forces us to be careful about whom we let in. A culture built on rowing together keeps no seat for the brilliant soloist.

We pay these willingly because they are the cost of the culture our work requires. That is the real test of a culture — not whether it is admired, but whether you would pay its price again for what you are trying to build.

So, is culture chosen or grown? I like to believe that the leaders of every organisation are responsible for designing it through value statements, policies, and, most importantly, how we behave. In truth, it grows from what we need to do, who we need to do it with, and what we are prepared to give up.

The celebrated cultures are neither templates to copy nor warnings to avoid. They are, instead, proof that the question isn’t which culture is best, but which one is right for the work in front of you. Build that one.

Share this article