This spring, Mount Everest had one of its most crowded seasons on record—a record 494 permits, and 274 climbers on the summit in a single day, more than ever before. Most came home. The ones who didn't share a detail almost everyone skips: the client climbers who died didn't die reaching the top. They died coming down, after they'd already made it. Climbers have a name for the thing that does this: summit fever—the tunnel-visioned pull toward a goal that's right there, even when every signal says turn around.

This piece is about summit fever, and why the version that runs your team is more dangerous than the one on Everest. We'll look at why the descent—not the summit—is where people die, the one rule elite climbers use to beat their own judgment, and how to install that rule on a team before the next launch turns into a death-zone scramble. The short version: the top of the mountain is not the finish line. It's the halfway point you're most likely to mistake for the end.

What climbers mean by summit fever

Above 8,000 meters—about 26,000 feet—climbers enter what mountaineers call the death zone, where there's roughly a third of the oxygen available at sea level and the human body is, slowly, dying. You can't think clearly. You can't recover. Every hour up there spends a budget you can't refill. Summit fever is what happens when a brain starved of oxygen, hours into the hardest day of its life, locks onto the one thing it came for and stops weighing anything else.

Now the counter-intuitive part. A 2026 study in the Journal of Physiology, Updates to Mortality on Mount Everest: 1921–2024, found that "more than three-quarters of climber deaths from 1921 onward have occurred during descent from extreme altitude." Not on the way up, chasing the summit. On the way down, after getting it.

The same study found that climbers were "eight times more likely than sherpas to die during summit descent (0.8% versus 0.1%)"—a gap the authors tie to experience and acclimatization, not raw fitness. And while overall mortality has fallen sharply, from 1.4% in 1921–2006 to 0.7% in 2007–2024, the shape of the danger hasn't moved. "Most deaths still occur when climbers are severely impaired and rescue is difficult or impossible," the researchers note. The summit feels like the achievement. Statistically, it's the moment you've spent the resources you need to survive the part that actually kills you.

The turnaround time: a rule built to overrule you

Elite mountaineering's answer to the fever is brutally simple: decide when to quit before you start, and hand the decision to a clock instead of to your own judgment.

It's called a turnaround time. On the 1996 Everest expedition made famous by Into Thin Air, guide Rob Hall had one. Per the account on Wikipedia, "Hall established a turn-around time of 2:00 p.m. ... at which time all climbers, wherever they were on the mountain, were required to begin their return to Camp IV." The logic was sound: anyone still climbing after 2:00 wouldn't make it back down before dark and exhaustion turned lethal.

Then the rule met the fever. "Many climbers summitted after 2:00," Wikipedia notes, "which ... did not give them sufficient time to descend to the relative safety of the South Col and Camp IV before darkness which occurred about 6:45 p.m." A storm rolled in. Eight people "were caught in a blizzard and died on Mount Everest while attempting to descend from the summit." Hall was one of them.

The lesson isn't "have a rule." Hall had the rule. The lesson is why the rule existed in the first place: the person standing 300 feet below the summit at 2:30 p.m., oxygen-deprived and agonizingly close, is the worst possible person to decide whether to keep going. A turnaround time is a decision made by your clear-headed self, at base camp, built to overrule your compromised self on the mountain. It only works as a hard line—not a number you "take into account" when the moment comes.

Summit fever at work—and why the descent is invisible

Your team has summits too. The launch. The deadline. The big demo. The quarter-end number. And it runs the same fever: the late push where "we're so close" drowns out every signal that the smart move is to cut scope, slip the date, or kill the thing outright.

Watch what the fever does. The closer the goal, the more the resources already spent start steering the decision. This is the sunk cost fallacy at work—what Asana defines as "a cognitive bias that leads us to continue investing in something because of what we've already invested, even when the current costs outweigh the benefits." Its uglier cousin, escalation of commitment, kicks in when admitting the project should stop feels like admitting you failed—so you pour in more. The Concorde burned through $2.8 billion over 27 years past the point anyone believed it would pay off, because stopping had become unthinkable. Teams run the small version of that every quarter.

And the bill, exactly as on Everest, comes due on the descent. PMI's Pulse of the Profession has found that 9.9% of every dollar is wasted due to poor project performance—$99 million for every $1 billion invested, and that 52% of projects suffer scope creep. Those aren't summit numbers; they're descent numbers. The post-launch firefighting. The bugs shipped to hit a date. The support load nobody staffed. The burned-out owner who carried the final push and quietly starts taking recruiter calls. Post-launch burnout is the team's version of dying on the way down: you got the win, and it cost you what you needed to survive the next month.

The case for pushing through (and the exact condition where it breaks)

Now the honest objection, because it's a strong one. Isn't "set a turnaround time" just a license to quit? Persistence is real. Grit is real. Most teams that ship anything hard do it by pushing through the moment it stopped being fun, and the team that bails at the first obstacle ships nothing. "We're almost there" is often true, and turning around 300 feet from the summit can be the cowardly call, not the wise one. Steelman it fully: a rule that makes you stop early, every single time, is its own failure mode.

Here's the hidden condition that decides which situation you're actually in. Pushing through is the right call when two things hold: the cost to finish is bounded, and the finish is genuinely the end. Summit fever is what happens when neither is true. On Everest, the cost to finish isn't bounded—every minute past your turnaround time compounds—and the summit isn't the end; the descent is. At work, "we're almost there" stops being information the moment it has been true for three weeks straight. That's the tell. Bounded cost, real finish: persevere, and good for you. Unbounded cost, false finish: you don't have grit, you have summit fever—and grit is precisely the story it tells you so you'll keep climbing.

The pros don't resolve this in the moment, because the moment is when judgment is worst. They resolve it in advance.

How to set a turnaround time and beat summit fever

You install the rule the same way climbers do—before the climb, in writing, while everyone is still oxygenated and rational.

None of this is about quitting easily. It's the opposite. The turnaround-time rule is how you earn the right to push hard—because you've pre-committed to the one signal that tells you when hard has become fatal. Knowing when to quit a project isn't weakness; it's the skill that lets the survivors keep climbing.

Why distributed teams miss the turn

Here's where it gets worse for remote and async teams: across distance, summit fever is invisible.

On a co-located team, someone sees the fever. They notice the engineer who hasn't slept, the room that has lost perspective, the energy that has curdled from focused to frantic—and someone with standing can say "stop." Spread that team across time zones and the tell disappears. The 2 a.m. push to make the date happens silently. The decision to blow past the stopping point lands in a Slack thread no one with authority reads in time. What travels across the distance is the shipped artifact—the summit photo—not the cost of getting there or the state of the person who paid it. The team celebrates the summit and never sees the descent.

This is the same blind spot from a different angle. The crux of a project is the single hardest move; summit fever is the judgment failure near the top—a different danger on the same climb. It's why busy teams mistake motion for progress, and why the recovery you skip after a crunch is the decompression stop you can't afford to miss. Distributed teams also systematically underprice the descent—the post-launch risk that's easy to wave off until it floods you.

The fix is to make the turnaround time visible and shared. The stopping rule, the honest state of the climb, and the call to turn around have to live somewhere the whole team can see at once—not scattered across DMs where the fever wins by default. That's the gap a tool like Coommit is built to close: by keeping the conversation, a live shared canvas, and contextual AI in one place, a distributed team can set its turnaround time together, see in real time when it has crossed the line, and plan the descent instead of discovering it the hard way.

The bottom line

Summit fever isn't a character flaw. It's what a goal-fixated brain does under pressure, and the more invested and exhausted your team is, the stronger it gets—right when its judgment is weakest. The summit is seductive precisely because it looks like the end. It isn't. The descent is.

Remember one number: more than three-quarters of Everest deaths happen on the way down, after the win. Your team's launches follow the same curve. So set your turnaround time before you start climbing, hand the decision to the rule instead of to the moment, and plan for the descent like it's the dangerous half of the trip—because it is. The teams that come back aren't the ones that never feel summit fever. They're the ones that decided, in advance, not to listen to it.