The shortest people in history.

At the end of the last ice age, men hunting across what’s now Greece and Turkey stood around 5ft 9in. By 3000 BC, after their descendants had taken up farming, the average man was nearer 5ft 3in.

That’s the skeleton record Jared Diamond laid out in a 1987 essay he titled “The Worst Mistake in the History of the Human Race.” Later studies argue about how big the drop was. None of them argue about which way it went.

Farming was a bad deal for the farmer: a narrower diet, more disease, more hours bent over a field. The hunter ate better and, by Diamond’s account, had more time off.

And yet farming won, everywhere. A field feeds more mouths than a forest, and grain keeps through winter. The individual got shorter. The tribe got bigger.

I think about that every time I ask a sales leader a question I’ve used for years:

Would you rather hit your target and upset a client, or keep the client happy and miss your target?

Hunters answer straight away, and they all give the same answer. Farmers hesitate. I don’t know a quicker way to tell them apart.

One of the best hunters I ever met put it more bluntly: “Once the client signs on the dotted line, that’s it. They’re dead to me.”

On his own numbers, he’s right: the commission’s real and the quarter closes on time. If the client gets neglected, that lands on somebody else’s P&L next year.

And you won’t talk him out of it. Hunters are wired for the chase. Ask one to spend his days nursing accounts and you’ll either lose him or he’ll quietly drift back to hunting.

So the fix isn’t turning hunters into farmers: it’s deciding where the hunt ends. Microsoft worked that out.

In 2017, Microsoft’s reps were selling Azure, its cloud platform, the way they sold everything else: bundled into big multi-year agreements. Contracts got signed, commission got paid, and plenty of customers barely switched it on.

That July, Microsoft did two things. It moved the finish line, so reps were paid on how much customers actually used rather than what they signed. And it built a different kind of team. Technical people, embedded with customers to get projects running. Amy Hood, the CFO, said Microsoft had never really used that type of person in sales before.

The hunters kept hunting. Only the kill moved, from signed to switched on. And the farmers were there to keep the meter spinning.

By 2018, Judson Althoff, who runs Microsoft’s sales, said less than 20% of anyone’s plan was tied to what they sold. A year on, Microsoft’s stock was at an all-time high and the cloud business it had struggled to sell was growing fast.

You can watch a medical device company try the same move right now.

EBR Systems launched WiSE, a wireless pacing system for heart failure, in the US in 2025. Its reps did what hunters do and opened hospitals. By September it had 57 purchasing contracts. Forty hospitals had implanted at least once, and 20 had done three or more.

That’s Microsoft’s 2017 problem again. A new implant isn’t won when the hospital signs. It’s won when the surgeon books the next case without being chased.

In October the CEO, John McCutcheon, took the commercial team under his own control, pointed it at repeat procedures in hospitals already using WiSE, and took a soft quarter in public. “I take responsibility for that,” he told investors.

Microsoft’s answer took about a year to show, and EBR’s starts showing in January. Ignore the contracts. Watch the 20.

Most leaders use the hunter or farmer question to sort candidates, and it works. The more useful question is the one you ask about your own business:

Is there anyone on my team who’d ever have to choose?

If there is, you’ve either put a hunter where a farmer belongs, or told him the hunt ends somewhere it doesn’t.

The best leaders don’t just ask candidates whether they’re hunters or farmers.

They make sure nobody on the team ever has to answer.