[Guides](/guides/) › The Alps are not full, they are concentrated # The Alps are not full, they are concentrated Every summer the same story runs: the Alps are overrun, the huts are full, book in January or don't bother. Every summer it is half true. The huts people have heard of are full. The network they sit in is not. That is not our opinion. It is the stated position of the federation that owns and runs the French refuges, and it is echoed by the Austrian Alpine Club, by a departmental tourist board, by peer-reviewed research, and by the Alpine Convention, the treaty governing the Alps, whose tourism protocol has committed its signatory states to spreading visitor demand across both time and place since 1998. Every claim of that kind below is attributed and linked. What we add is the measurement. > The problem isn't that there are too many people in the mountains. It's that this traffic is massively concentrated in the same places and at the same times. That is Charles Van der Elst, President of the FFCAM, writing in [Alpine Mag](https://alpinemag.fr/tribune-charles-van-der-elst-president-ffcam-montagne-epreuve-nos-choix/) on 20 August 2026. The Federation Francaise des Clubs Alpins et de Montagne operates 120 refuges, including much of the French side of the Tour du Mont Blanc. His own federation's numbers show the shape of it. Across 120 FFCAM refuges and 315,000 overnight stays in 2024, the ten busiest huts took about a quarter of every night slept in the network. The busiest, Col de la Vanoise, ran nearly six times the network average. At the Gouter and Tete Rousse, he notes, an entire season's places go in a few hours when bookings open. 8.3%of the FFCAM's huts... 25.4%...took this share of its overnight stays 5.7xbusiest hut vs network average **The FFCAM published the counts; the percentages are ours.** They reported 315,000 nuitees across 120 refuges in 2024, and a top ten running from Col de la Vanoise at 15,000 down to Drayeres at 6,100. Ten huts out of 120 is 8.3% of the network; their 79,000 nights are 25.4% of the 315,000; and Col de la Vanoise at 15,000 is 5.7 times the network average of about 2,600. Nothing there is a HutScanner measurement. It is division applied to their numbers, and anyone can repeat it. The same shape appears in Switzerland, in a different federation's books. Of the 409,000 nights slept in Swiss Alpine Club huts in 2025, the ten busiest huts took 79,078 of them, about one night in five, across a network of 152 huts and bivouacs. Three regions, the Valais, Grisons and Bernese Alps, took 68% of the total between them. Two federations, two countries, two independent sets of books, one pattern. Nor is it only a hut phenomenon. The Alpine Convention's own report on Alpine tourism found that 46% of tourist beds sit in 5% of Alpine municipalities, while 37% of municipalities have no tourist beds at all. Concentration is the structure of tourism in the Alps, not a quirk of one route in one August. ## What a settled month looks like We have recorded every hut's availability every day through this summer, so August 2026 is now a closed book. Across every hut that published a live bed count, 787 of them, August finished **32.6% sold out**. Two-thirds of hut-nights in the busiest month of the year still had a bed in them on the night, and over 318,000 beds were never slept in. 32.6%of hut-nights ended up sold out 318,660bed-nights that stayed empty 787huts measured, every night Each August night is counted as it stood on our last reading before that night was slept, so this is the final state of the books rather than a forecast. We use August deliberately: a September figure would still be moving, while August cannot change now. ## The evidence is inside a single route The cleanest way to see the problem is not to compare famous places with obscure ones. It is to look inside one famous route, where every hut is on the same trail, in the same guidebooks, on the same nights. Through the whole of August 2026, four huts in our Tour du Mont Blanc pack were sold out on every night we recorded. Three never sold out once. The route as a whole ran 52.7% sold out and still ended the month with around 16,800 empty bed-nights. Some of that spread is honest geography. The pack covers the massif rather than only the waymarked loop, so a few of the emptiest are Swiss or high-alpine huts a little off the walkers' line: Saleinaz and Orny are climbing bases, not walkers' stages. But geography does not rescue the picture, because the pattern holds among huts nobody would call remote. The Gite du col de la Forclaz sits on the standard circuit at a road col, Charamillon is a lift ride above Le Tour, and the Relais d'Arpette is on the Fenetre d'Arpette variant. None of them sold out in the busiest month of the year. The clearest case is a pair where none of the usual excuses apply. The Refuge de la Balme and the Auberge La Roselette are both gites d'etape, 167 m apart in altitude, 5.1 km and under two hours' walk from each other, and the leg between them grades as ordinary mountain walking, T2 on the SAC scale, the same as the rest of that section of the tour. In August, La Balme sold out on **21 of 29 nights** and finished with 17 spare bed-nights. La Roselette sold out on **four of 31** and finished with **272**. Two gites d'etape, 5.1 km apart on T2 ground - % of August nights sold out - Refuge de la Balme72% - Auberge La Roselette13% Same kind of place, same ground, an hour and fifty minutes apart. Nothing about difficulty, altitude or remoteness explains that. The difference is which hut the stage list names. ## Massifs compared, August 2026 settled Every region here is one we have curated hut by hut and mark as reviewed in the app. Reviewed regions - % of August hut-nights sold out - Ecrins92.7% - Vanoise63.1% - Tour du Mont Blanc52.7% - Karwendel35.3% - Walker's Haute Route28.0% - Otztal16.7% - Zillertal15.5% - Bernese Oberland14.3% - Valais and Zermatt8.9% Region pages: [Ecrins](/regions/ecrins/), [Vanoise](/regions/vanoise/), [Tour du Mont Blanc](/regions/tmb/), [Karwendel](/regions/karwendel/), [Walker's Haute Route](/regions/haute-route/), [Otztal](/regions/otztal/), [Zillertal](/regions/zillertal/), [Bernese Oberland](/regions/bernese-oberland/), [Valais and Zermatt](/regions/valais-zermatt/). A tenfold spread in the same month. Each row counts that region's own huts, and some regions overlap, so the rows are comparable to each other but do not sum to the Alps-wide figure above, which counts every hut once. ## Demand is rising and capacity is not It would be comfortable to conclude that the crowding is imaginary. It is not. The Swiss Alpine Club recorded 409,000 overnight stays in its huts in 2025, up 12.7% on the year, the most it has ever recorded. Winter alone rose 25.7%. The demand is real and it is climbing. What is not climbing is capacity, and here the pressure is structural rather than commercial. The SAC's own *Hutten 2050* work found that more than a third of its 152 huts and bivouacs could be destabilised by thawing permafrost, with 42 potentially exposed to rockfall from permafrost terrain. The club expects to spend 20 to 25 million francs a year on hut construction to 2040, about a third of it climate adaptation. Huts in that programme close for renovation, get relocated, and in some cases are taken down. And the concentration is measurable over decades, not just over a season. Researchers at Lausanne and Grenoble took attendance records from eight huts in the Val d'Herens and Val d'Anniviers and compared the late 1990s with the late 2010s. Summer overnight stays across the group fell 11%, while rising sharply at the huts offering quick access to emblematic 4,000 m peaks. The Moiry hut grew 35%. Demand went down; concentration went up. That is the whole problem in one dataset, and it is not a problem more beds would solve. Nor is this a comfortable business to be in anywhere on the network. Swiss reporting in 2023 put the number of SAC huts that are actually profitable at around a dozen out of 153. Unprofitability is the norm, not a quiet-hut affliction. And it is worth being careful about what that does and does not mean, because there is a tempting story here that turns out not to be true: huts do close, but the record says they close for renovation backlogs, failed water sources and wardens who cannot be replaced, not for want of guests. So the arithmetic is fixed: more walkers overall, a capped and slowly shrinking bed stock, and no realistic prospect of building the difference. Redistribution across the network is not one option among several. It is the only lever left. ## What concentration actually costs The bill does not arrive as crowded huts. It arrives in three less obvious places. **It wastes the wardens' capacity.** On Austrian Alpine Club huts, 30 to 40% of all reservations are now cancelled, against 15% in 2022. As the OeAV puts it, cancellations often come so late that the freed places can no longer be refilled. A hut turns away a walker in March and sleeps half empty in August. And when guests simply fail to arrive without cancelling, it is not always clear whether they are in trouble on the mountain, which can trigger a search. **It pushes people into carrying a tent who never wanted one.** This is the one we care about most, and it is the reason this project exists at all. In the summer of 2025 I wanted to walk hut to hut. I could not work out a chain that had beds on consecutive nights, so I ended up with the worst of both worlds: some nights in huts, and a full set of camping gear carried the whole way for the nights I could not fill, pitched at valley campsites because bivouac was not permitted where I was. I carried more weight than I needed, I could not stay high, and I missed most of the hut ambience I had gone for. The route existed. I just had no way to find it. Multiply that by everyone who reaches the same dead end and some of them do not drop to a campsite, they pitch where they should not. The consequences are already visible in French law: Haute-Savoie now runs prefectoral orders restricting bivouac in the Contamines-Montjoie and Aiguilles Rouges reserves, with capacities, designated zones, overnight-only hours and compulsory reservation in places. Courmayeur has banned it on its section outright. Van der Elst puts the scale problem precisely: "Five or ten tents, that's bivouacking. Sixty, a hundred, sometimes two hundred in the same place, that becomes something else." The activity is not the problem. The concentration is, arriving one layer down. We are not against bivouacking. Above the treeline, done properly, by people equipped for it, it is a legitimate and beautiful way to travel. What is not good for anyone is a hiker with no bivouac experience being pushed above the treeline by a booking system, onto ground that cannot absorb them, because the search tool they had could only see eleven huts. **It makes the network look smaller than it is.** A route is a graph, not a list. The moment you can see every hut within a day's walk rather than the ones named in a stage table, a "full" itinerary usually turns out to be a solved one: shift a night, split a stage, use the hut two hours further on. ## Guidebooks are not the problem, they are half the answer It is tempting to blame the stage list, and it would be wrong. A numbered itinerary is a genuinely good product: it turns an intimidating massif into something a first-timer can commit to. Guidebooks, trip reports and the big trekking communities are the reason most people get to the Alps at all, and they carry the knowledge no dataset holds: what the ground is actually like, where the weather turns, which descent is miserable in rain, what to do when your plans fall apart at Les Chapieux. The gap is narrower and more specific than "guidebooks cause crowding". **A stage list is fixed in print and availability changes every day.** A book set in winter cannot know which huts will have beds next September, and no author can fix that, because the fact moves daily. That is not an editorial failure. It is a live-data problem sitting underneath a perfectly good editorial product. Which is why we think these fit together rather than compete. A guide tells you what the route is and how to be ready for it. A community tells you what it was actually like last week. Neither can tell you which huts still have a bed on your dates, and we cannot tell you any of the first two. We are a planner, and a planner is not much use without the route knowledge that comes from people who have walked it. So if you are planning a Tour du Mont Blanc, get the context from the people who have it. Cicerone's *Trekking the Tour of Mont Blanc*, long associated with Kev Reynolds, is the standard stage-by-stage reference and worth owning. *Get TMB Ready* by Paul Homersmith collects what the big TMB community has worked out between them, which is exactly the kind of practical knowledge no dataset holds. The large TMB groups will answer a question about one specific crossing faster than anything else on the internet. Read those first. Then come and find out which huts actually have beds on your dates. ## What we think should happen - **Fill the huts first.** Where a hut exists and has a bed, that is the lowest-impact place for a walker to sleep: concentrated waste and water handling, no new ground taken. - **Vary the route before abandoning it.** Most "the TMB is full" problems are stage-boundary problems, not route problems. - **Then look further out.** Valais and Zermatt, the Bernese Oberland, the Zillertal: reviewed ranges that ran under 15% sold out through the whole of August. The walking is not a consolation prize. - **Keep bivouac for people equipped to do it**, above the treeline, where it is permitted, not as the fallback for a failed search. - **Cancel early if plans change.** A cancellation in March is a bed for somebody. A cancellation on the night is a loss for the warden and, occasionally, a search party. That is the whole argument for what we built. HutScanner treats the Alps as what they are, a connected network of several thousand huts with live bed counts, and finds chains of huts that actually have space on consecutive nights, at a walking pace you set. Not a better stage list. A way to see the two-thirds of the network that the stage list was never able to show you. ## Sources and method Availability figures are HutScanner's own, from the huts' own booking systems, read daily and committed to version control every day, so the series is per hut and per night. Each August night is counted as it stood on the last reading taken before that night was slept. Huts are counted once each: our region packs overlap, so a hut on both the Tour du Mont Blanc and the Mont Blanc massif is de-duplicated rather than counted twice. Only huts whose system publishes a bed count are included, so spare-bed figures are a floor. "Sold out" means the booking system offered no bed of any kind that night. Distances, ascent and walking times are from our own routing over the walking network, not straight lines, and each leg carries an SAC grade from that routing. Where this piece names a region or compares particular huts, it uses the twelve regions we have curated hut by hut and mark as reviewed in the app. Within those twelve, the same settled August reads 34.9% sold out across 306 huts. - Charles Van der Elst, President of the FFCAM, tribune in [Alpine Mag](https://alpinemag.fr/tribune-charles-van-der-elst-president-ffcam-montagne-epreuve-nos-choix/), 20 August 2026. - FFCAM 2024 refuge attendance, reported by [Mon Sejour en Montagne](https://www.mon-sejour-en-montagne.com/actualites/alpes-le-top-10-des-refuges-les-plus-frequentes-en-2024/) from Dauphine Libere, 7 December 2024. - Schweizer Alpen-Club SAC, ["SAC-Hutten verzeichnen so viele Ubernachtungen wie noch nie"](https://www.sac-cas.ch/de/huetten-und-touren/sac-huetten-verzeichnen-so-viele-uebernachtungen-wie-noch-nie-51553/), 21 April 2026. - Schweizer Alpen-Club SAC, ["Klimawandel stellt SAC-Hutten vor grosse Herausforderungen"](https://www.sac-cas.ch/de/huetten-und-touren/klimawandel-stellt-sac-huetten-vor-grosse-herausforderungen-46388/), 26 November 2024. - Tourisme Haute-Savoie, ["Hiking in Haute-Savoie: can you still find a refuge in August 2026 without booking three months in advance?"](https://www.tourisme-haute-savoie.com/en/news-from-haute-savoie/hiking-in-haute-savoie-can-you-still-find-a-refuge-in-august-2026-without-booking-three-months-in-advance/), 24 June 2026. - Mourey, J., Clivaz, C. and Bourdeau, P. (2023), ["Analysing the Transformation of Sports Practices in Less-Developed Mountains Based on Huts Attendance Data"](https://journals.openedition.org/rga/11141?lang=en), Journal of Alpine Research 111-1. - SRF, ["Betrieb von SAC-Hutten: eine finanzielle Gratwanderung"](https://www.srf.ch/news/schweiz/unterkuenfte-in-den-alpen-betrieb-von-sac-huetten-eine-finanzielle-gratwanderung), 21 October 2023. - Osterreichischer Alpenverein, [press release on hut cancellations](https://www.alpenverein.at/portal/service/presse/2026/2026_06_23_storno-auf-huetten.php), 23 June 2026. - Alpine Convention, [Protocol on the Implementation of the Alpine Convention in the Field of Tourism](https://www.alpconv.org/fileadmin/user_upload/Convention/EN/Protocol_Tourism_EN.pdf), signed at Bled, 16 October 1998, Article 18. - Alpine Convention, [Sustainable Tourism in the Alps, Report on the State of the Alps 4](https://www.alpconv.org/fileadmin/user_upload/Publications/RSA/RSA4_EN.pdf), 2013, p. 28. - Bivouac rules: Haute-Savoie prefectoral orders DDT-2026-0474 and DDT-2026-0472, and Courmayeur's published notice. Covered in full in our [Tour du Mont Blanc bivouac guide](/guides/wild-camping-bivouac-tour-du-mont-blanc/).