How Crowded Has Mount Everest Become Lately

I still remember the first time a summit photo stopped looking like a victory and started looking like a queue. A friend forwarded it without comment: bright suits, fixed ropes, people stacked shoulder to shoulder on a ridge that, in every old book I had read, was supposed to be empty. That picture was not a fluke. Long-running expedition logs now show more people on Mount Everest than at any point in the last three decades of consistent record-keeping. In 2025 the mountain recorded 585 visitors and 934 hired workers, 1,519 people in total. In 1995 the same kind of count sat at 395. Nearly four times the human presence, on the same narrow piece of ice and rock.

If you only follow the highlight reels, crowding sounds like a slogan. The logs make it concrete. Successful climbs have piled into spring. The people who carry loads, fix rope, and keep clients alive have grown faster than the client list. And on a handful of calm May days, hundreds can stand on the same summit within hours of each other. That is the story underneath the postcard.

The Scale Of The Everest Climbing Boom

Everest did not suddenly become popular last season. The curve has been bending upward for years, with a few violent interruptions and then a return to the same direction. Visitors in these records include paying clients and independent climbers. Workers are hired expedition staff, most of them Nepali Sherpas and other high-altitude specialists. Both groups have expanded. They have not expanded at the same pace, and that split matters more than the headline total.

Look at the early stretch. 1995: 247 visitors, 148 workers. 1998: 271 and 160. Even the year 2000, already feeling busy to people who had climbed in the 1980s, logged 406 visitors and 247 workers, 653 people altogether. By the late 2000s the combined figure was routinely clearing 800, and in 2007 it crossed 1,100. The 2010s pushed the total over 1,000 in several seasons. Then came the pandemic pause, a near-empty 2020 with 32 visitors and 28 workers, and a rebound that did not creep back. It snapped back.

A Thirty-Year Headcount, Not A Vibe

Here is the shape of the boom in plain figures. I am pulling the pattern, not every footnote, because the pattern is what changes how you read a summit photo.

Two things jump out if you read the columns instead of the total. Visitor numbers wobble. They sit in a band, roughly 450 to 600 in busy modern years, with the obvious hole in 2020. Worker numbers do not wobble the same way. They climb. In 1995 there were about three hired workers for every five visitors. By 2025 the ratio had flipped toward roughly eight hired workers for every five visitors. Hired staff made up more than 60 percent of everyone recorded on the mountain that year.

That is not a small bookkeeping change. It is the commercial model showing up in a spreadsheet. A client does not appear at Camp Four having carried every bottle, every stove, and every coil of rope. Somebody else did most of that work, often more than once, while the client was still rotating down to base camp to sleep and eat. I’ve found that people argue about “crowds” as if the only crowd that counts is the one in the summit photo. The larger crowd is the one that built the photo.

Summits Multiplied Faster Than The Old Baseline

Presence and success are related, but they are not the same number. Of the people on the mountain in 2025, 878 reached the summit. Across 2021 to 2025, the mountain averaged about 720 summits a year. In the late 1990s the annual average sat near 100. More than seven times as many successful climbs, in a window of years that still includes the odd interrupted season.

A short run of annual summit totals makes the acceleration hard to romanticize away.

• 1995: 82 summits. 1999: 116. The decade still felt selective.
• 2004: 335. 2007: 611. Commercial seasons had clearly arrived.
• 2013: 680. 2016: 678. 2018: 850. The “busy year” became the normal year.
• 2019: 899, the pre-pandemic peak in this set.
• 2021: 462, a partial rebound. 2022: 704. 2023: 685. 2024: 870. 2025: 878.

Notice the years that are missing from that highlight list on purpose. 2014 logged 133 summits and then stopped, after a disaster that killed workers on the route. 2015 logged zero summits on registered expeditions in this accounting, after the earthquake and the avalanche at base camp. 2020 logged 49. Crowding is not a straight line. It is a line that drops when the mountain or the world intervenes, then resumes as if the pause was a scheduling problem.

The mountain did not get wider. The calendar of good weather did not get longer. The number of people treating those few days as the only days that count did.

Why The Old Autumn Mountain Disappeared

Decades ago, a successful Everest climb was not a spring exclusive. In the 1980s, expedition records counted 87 spring summits and 77 autumn summits, plus a thin scatter of winter and summer successes. Autumn was a real season. Winter was rare but not mythical. Summer existed at the edge of the data.

That spread collapsed. From 1990 to 1999 the logs show 739 spring summits against 135 in autumn, five in winter, and none in summer. From 2000 through 2009, more than 99 percent of recorded summits happened in spring: 3,346 against three in autumn. The 2010s went further. 5,561 spring summits. Six in autumn. None in winter or summer. Since 2020 the pattern has barely twitched: 3,634 spring summits, 14 in autumn, and nothing logged in the other two seasons.

So when someone says Everest has become a spring mountain, they are not being poetic. They are describing a near-total seasonal monopoly. Commercial teams want the most reliable window, the one clients can book a year out, the one operators can staff, the one insurers and permit offices already understand. Autumn still has weather. It no longer has the industry. Perhaps the most interesting aspect is how fast a mountain with four theoretical seasons became a mountain with one practical season. Once the clients, the flights, the rope-fixing teams, and the lodge bookings all lock onto May, autumn stops being a alternative and starts being a curiosity.

Hundreds On The Same Ridge, Sometimes The Same Day

Even inside spring, the traffic is not spread across April, May, and early June like butter on toast. It clumps. Summit attempts wait on the jet stream. When that high river of wind shifts off the summit, a few days of lower wind open up, usually before the summer monsoon shuts the door. Teams that have spent weeks acclimatizing, burning money and patience at base camp, all see the same forecast. They move.

The clearest single spike in the dataset fell on 23 May 2019, when 354 people reached the summit from Nepal and Tibet combined. That is the highest one-day total in these records. Picture a sidewalk at rush hour, then remove the sidewalk. Replace it with a corniced ridge, a fixed line, an oxygen mask, and a drop that does not forgive a stumble. Three hundred and fifty-four summits in a day is not “busy.” It is a compression event.

I keep coming back to a simple mechanic. Weeks of acclimatization are individual. The weather window is shared. You cannot bank a calm day the way you bank a rest day. If the window is 48 hours long, every team that is ready will try to use those 48 hours. The ones who wait for a quieter tomorrow may get no tomorrow at all. That incentive, more than any poster or permit office, builds the queue.

Does every crowded day look like 2019? No. Some seasons break into two or three smaller pulses. Some years the wind never really leaves, and the summit total sags even if base camp is full. But the structure is stable enough to plan around. Operators know it. Clients are told to expect it, sometimes in the brochure language of “teamwork on the fixed lines,” sometimes more honestly as waiting. The waiting is not abstract. It is cold, it burns oxygen, and it happens at an altitude where standing still is its own risk.

What The Queue Does To A Climb

A queue on flat ground is an annoyance. A queue above 8,000 meters is a physiological problem. People radiate heat poorly in down suits, then cool too fast when they stop. Regulators freeze. Bottles run down while nobody is gaining height. A slower climber ahead does not just delay your photo. They delay your descent, which is when a lot of accidents actually happen, because fatigue and darkness arrive together.

Guides talk about turnaround times for a reason. If you have not reached a certain point by a certain hour, you go down, summit or not. Crowding makes that rule harder to keep and more important to keep. The client who paid for the top does not experience a delayed rope team as logistics. They experience it as the day slipping away. Pressure travels downhill through the team: from the client to the guide, from the guide to the rope fixers who were on the ridge hours earlier. Nobody in that chain is irrational. The system just has more bodies than the ridge has width.

There is also a quieter effect that rarely makes the video clips. Decision quality drops when you cannot move. In my experience of watching how people describe these days afterward, the stories split. Some climbers remember a strange camaraderie, strangers clipping past each other with small courtesies. Others remember fear that felt administrative, the fear of being stuck behind a problem you did not cause and cannot fix. Both can be true on the same morning.

The Workforce That Arrives Before The Clients

Long before most clients clip into the summit ridge, high-altitude workers have already carried supplies, stocked camps, and helped prepare the route. Commercial expeditions lean on guides and other staff through weeks of acclimatization and the final push. That labor is the reason a visitor count of 585 can coexist with a worker count of 934. The mountain is staffed like a short, brutal season of hospitality and construction, conducted where helicopters are a last resort and a dropped glove is not a joke.

The occupational split shows up most clearly when things break. In 2014, all 17 deaths recorded in the database were hired workers. In 2015, when the Nepal earthquake sent an avalanche into Everest Base Camp, workers accounted for 11 of the 14 deaths recorded on registered expeditions. Those were exceptional disaster years, not a typical May. They still underline a blunt fact. The people who make the commercial season possible absorb a disproportionate share of the catastrophic risk.

Other seasons tell a more mixed story, which is worth sitting with instead of flattening into a single moral. Some years visitor deaths outnumber worker deaths. Some years the totals are small. A few recent snapshots from the same logs:

• 2019: 11 total deaths, 10 of them visitors, 1 worker.
• 2023: 18 total deaths, 12 visitors, 6 workers.
• 2024: 8 total deaths, 6 visitors, 2 workers.
• 2025: 5 total deaths, 4 visitors, 2 workers, in the figures as compiled.
• 2014 and 2015 remain the grim outliers, weighted almost entirely toward hired staff.

I am not going to pretend a table of deaths is a complete safety audit. Recording practices, cause of death, and whether someone was above base camp all complicate the picture. What the split does show is that “Everest risk” is not one risk. Clients face the summit push, the altitude, their own fitness, and the queue. Workers face those things across repeated trips, plus the earlier work of route preparation, when the mountain is less photographed and the seracs have not been negotiated by a hundred prior teams that week. Calling both groups “climbers” is accurate and, sometimes, a way of hiding the job.

A season can look successful in the summit column and still be a bad year for the people who opened the route.

A pattern visible in expedition fatality records, especially 2014 and 2015

Waste Is The Crowd You Can Weigh

More people means more waste, and waste at this altitude does not walk itself down. Nepal requires expeditions to remove rubbish. Doing it is difficult and expensive, which is a polite way of saying gravity and cold are not on the cleanup crew’s side. In spring 2024, the Sagarmatha Pollution Control Committee collected 85 tonnes of waste from Everest Base Camp and the higher camps. Human waste made up 27.53 tonnes of that haul.

Eighty-five tonnes is an abstract number until you imagine bags, barrels, and frozen loads moving down a glacier by hand and by yak, then by road. Human waste is the awkward subset. It cannot be tossed in a crevasse and forgotten, not if the valley below is also a water supply and a home. Committees, deposit schemes, and carry-out rules have all been tried. Some work better than others. None of them cancel the arithmetic. A mountain that hosts well over a thousand people in a season will produce a waste stream that a mountain hosting a few hundred did not.

There is a temptation to treat trash as the whole environmental story. It is the visible part. The less visible part is foot traffic on fragile moraine, helicopter rotations when weather allows, fuel for stoves and generators, and the slow enlargement of base camp from a camp into a seasonal town. I do not think every tonne collected is a failure. Collection is the opposite of neglect. I do think the tonne count is a receipt. It tells you the scale of the visit in a unit that does not care about sponsorship logos.

Permits, Fees, And The Business Under The Ice

Everest is also an income stream, and pretending otherwise makes the crowding debate fuzzy. In 2025 Nepal raised the spring climbing permit fee for foreign climbers from $11,000 to $15,000, with the new rate taking effect in September. That fee is only one bill. Climbers pay operators, who pay guides, cooks, porters, and lodges. The Khumbu economy is built, in no small part, on people who want the highest summit on Earth and are willing to fund the apparatus around that want.

A higher permit does two things at once. It pulls more money into public coffers for each foreign climber who still comes. It also tests whether demand is as inelastic as the summit photos suggest. Some climbers will shrug at an extra $4,000 on a trip that already costs far more than that in logistics, oxygen, and time off work. Others will hesitate, or shift seasons, or look at other 8,000-meter peaks where the permit and the queue are both smaller. We will not know the full effect from a single rule change. We will know it from the next few rows of the same table: visitors, workers, summits.

From a market point of view, the mountain is a scarce date with a strong brand. The scarcity is the weather window, not the rock. The brand is the name. Operators sell access to both. When too many operators sell the same week, the product degrades in a very physical way, even if the summit is still the summit. That is classic congestion. The fix, if anyone genuinely wants a fix, is not a slogan about respect. It is a limit on how many people can be funneled into the same days, paired with wages and insurance that reflect what the work actually is. Fee hikes without headcount management raise revenue. They do not, by themselves, widen the ridge.

A rough way to hold the tradeoff:
  More permits sold in the same May window
  = more income downstream
  = more bodies on the same fixed ropes
  = more waste, more waiting, more pressure on staff

Who Actually Stands In That Line

The visitor column hides a mix. There are experienced alpinists who could function on a less managed mountain. There are strong amateurs on their first 8,000-meter peak, supported heavily, which is a legitimate choice and also a different sport from the one the old autumn parties were playing. There are a few people who should not be there at all, and every crowded season seems to produce a story about one of them. Using those stories as the whole explanation is lazy. The queue is mostly made of prepared clients plus the staff attached to them, arriving on the same forecast.

Independent climbers still show up in the records. They are no longer the center of gravity. The center of gravity is the commercial team: a ratio of staff to clients that has tilted toward staff, a spring calendar, a summit push measured in hours of good wind. If you want the mountain to feel like 1995, you are asking for a different industry, not a different attitude on the same industry. Attitudes matter on the rope. They do not set the permit count.

Comparisons People Reach For, And Where They Fail

Crowded Everest gets compared to crowded trails, crowded museums, crowded airports. The comparisons are useful for about thirty seconds. A trail can be widened, or a permit lottery can spread hikers across dawn and dusk. A museum can add hours. An airport can add a gate. The summit ridge cannot add a lane. The jet stream does not accept reservations. Acclimatization takes weeks, so you cannot tell a team to “come on Tuesday instead” once they are already in the valley, not without asking them to gamble the season.

Other high peaks have crowds too, and some have learned ugly lessons about fixed ropes and bottlenecks. Everest is simply the case where the brand, the fee, the media, and the geography all amplify the same week. Tibet’s northern side and Nepal’s southern side both feed the top, which is why a one-day total like the 2019 figure has to be read as a combined number, not a single-base-camp story. Two approaches, one small summit. That geography is not going to renegotiate.

What might renegotiate is the social expectation that a summit photo is the proof of the trip. As long as the photo is the product, teams will chase the day that makes the photo possible. Shift the product, even slightly, toward a climb that can be called complete without the top, and some of the compression eases. I am not optimistic that the market will do this on its own. The top is the sentence everyone understands. “We turned around at the South Summit in good style” is a better mountaineering sentence and a worse advertisement.

A Practical Reading For Anyone Watching The Next Season

If you follow this mountain from a distance, a few markers are more honest than the viral clip of the line.

1. Watch the worker count, not only the visitor count. A season can look stable on clients and still be loading more labor onto the route.
2. Watch how many summits fall into how many days. A high annual total spread across a dozen days is a different mountain from the same total crushed into three.
3. Watch autumn. If it stays near zero, the spring monopoly is intact, whatever anyone says about spreading the load.
4. Watch waste hauled out, especially human waste. It is the crowd made measurable after the flags come down.
5. Watch whether a higher permit fee changes the mix of who comes, or only the invoice.

None of those markers require you to scorn the people who want to stand on top. Wanting that is old. Paying a large team to make it more likely is newer, and it is the part that scaled. The records from the 1980s are a reminder that the mountain once rewarded patience across seasons. The records since 2000 are a reminder that patience lost to the calendar.

The Human Footprint, Without The Lecture

I do not find the moralized version of this story very useful. “Everest is ruined” ignores the families in the valley for whom the season is school fees and roof repairs. “Everest is fine, it has always been hard” ignores a one-day summit count that would have been a full year in the mid-1990s. Both sentences are doing publicity. The more accurate sentence is plainer. The highest mountain on Earth is still uniquely appealing, and three decades of logs show how large the human apparatus around that appeal has become.

Apparatus is the right word. Ropes, ladders, camps, oxygen systems, weather services, permit offices, rescue insurance, waste depots, teahouses, flights into a short mountain airstrip. Remove any major piece and the current summit numbers fall. Keep every piece and the numbers have room to rise until something external, a storm year, a political closure, a fee that finally bites, pushes them down. The 2020 dip is the cleanest demonstration. Take away international travel and the mountain empties. Restore travel and, within a couple of seasons, the totals are back above a thousand people and the summits are back near 700 to 900.

That rebound is what I would want a reader to remember, more than any single shocking photo. Demand was not a bubble that popped. It was a pipeline that paused. Pipelines can be managed. They do not manage themselves, and a ridge at 8,800 meters is a bad place to discover that you assumed they would.

What Climbers Say They Feel, And What The Data Adds

Talk to people who have been up in different eras and you get texture the table cannot give. Older climbers describe longer silences and more route-finding. Recent climbers describe briefings, radio checks, and the odd comfort of knowing a fixed line is there, alongside the odd dread of knowing everyone else knows it too. Sherpa climbers describe repetition: not one summit push but several rotations, not one season but a career of seasons, with the body keeping the books even when the press does not.

The data adds a check on nostalgia. The 1980s were not a museum of perfect style. People died then too, often with less backup. Autumn success was real, and so was the thinness of winter. What changed is concentration. Skill did not vanish. It got surrounded. A competent climber in a crowd still has to climb. They also have to wait, communicate, and sometimes abandon a plan because the plan of fifty other people has occupied the only safe line.

There is a version of this essay that ends with a rule. Cap the permits. Spread the season. Pay the workers more. Require human-waste removal with audits, not pledges. I think those are serious levers, and I also think they collide with incentives that are working as designed. A cap reduces the fee base unless the fee rises again. A spread into autumn requires clients and staff to trust a season the industry has spent twenty years abandoning. Higher pay without fewer rotations still leaves people in the Khumbu Icefall more often than any client. Policy here is a bundle, not a slogan, and bundles are slow.

Holding The Numbers Together

Put the pieces in one place, because they belong in one place. About 1,519 people recorded on the mountain in 2025, against 395 in 1995. About 878 summits in that recent year, and an average near 720 a year from 2021 through 2025, against about 100 a year in the late 1990s. Spring accounting for nearly every success since 2000. A single day in May 2019 with 354 summits from both sides of the mountain. A workforce that now outnumbers visitors, and that absorbed every recorded death in 2014 and most of them in 2015. Eighty-five tonnes of waste hauled from base camp and above in spring 2024, more than a quarter of it human waste. A spring permit moving from $11,000 to $15,000.

None of those figures needs embellishment. Together they describe a place that is still extreme and no longer rare in the way the old books promised. Rare was the selling point. Extreme is what remains when rare is gone. You can love the idea of the climb and still admit the idea now includes a line, a large staff, a waste problem, and a fee that just went up. Admitting it is not cynicism. It is how you read a mountain that has been counted, carefully, for long enough that the change cannot hide inside anecdotes.

The next calm week in May will produce new photos. Some will be beautiful. A few will be used as proof that everything is out of control, and a few others as proof that teamwork still works. Both uses will be thinner than the logs. If the worker column rises again, if the summits cluster into two or three days, if autumn stays empty, then the structure described here is still the structure. The crowd is not a mood. It is a schedule, a payroll, and a weather window, stacked on a ridge that has not grown.

I keep the forwarded photo in mind less as outrage than as a caption. The caption is the thirty-year table. Once you have seen the table, the line of suits stops being a surprise and starts being the predictable output of a system that funnels ambition, money, and labor into the same few days on the highest point anyone can stand. Whether that system should be smaller is a political and local question, not a caption. The size, at least, is no longer in dispute.