In a Niseko winter, everyone is selling the same snow on the same mountain, yet the price of one night varies widely depending on the form of the product. In Kutchan in January 2026, resort hotels listed at ¥126,000 per room (N=10 properties), vacation rentals at ¥103,000 per house (N=7) and cottages at ¥65,000 per house (N=4) — a spread of nearly 2x across the three product types. This spread is about the price ceiling, i.e. how high a product can be sold. It is not a supply analysis that counts rooms, nor a capex analysis that builds up the construction cost of a house. Starting from winter data for the Niseko area, this article examines how far the ceiling on achieved rates diverges between hotel rooms, vacation rentals and onsen villas.

The intended readers are operating companies and funds weighing resort development, owners of existing ryokan considering adding detached annexes or villas, and regional-bank staff responsible for business-viability assessments. The overall picture of new supply covered in our earlier article (Vacation Rentals Lead Japan’s 2026 Hotel Supply) and onsen running costs (Buying Onsen Water: ¥8k–190k per Room a Year) are taken as given here, and the focus is narrowed to the price ceiling by product type.

Defining the product types — what you sell as one unit sets the rate ceiling

First, let us sort out the product types compared in this article, because how a lodging product is priced changes with what is sold as one unit. A hotel room is a product that sells one room to two guests, and its rate ceiling is set by room size and view, brand, and the demand in the area. A detached annex (hanare) is a standalone building on a ryokan’s grounds, sold to one party while sharing the main building’s meals and large baths. A vacation rental is a house with a kitchen and several bedrooms, sold as a whole to one party (usually 4–8 guests); even when the per-house price looks high, divided back per room it is often cheaper than a hotel room. A branded villa is a detached product served by a hotel brand’s operating team, planned either for sale to owners (branded residences) or as rental inventory.

On top of this sits a second axis: whether onsen water is piped in. A detached product with onsen becomes a different product from a vacation rental without one, because it can justify its rate in spring and autumn stays too, not just on winter ski demand. Table 1 summarises the categories used in this article and which property-type category each corresponds to in the MetroEngines Research & Consulting data.

Table 1: Product types compared / Source: MetroEngines Research & Consulting

Niseko area in winter — hotel rooms in Kutchan at ¥126,000, vacation rentals at ¥103,000 per house

We start with January 2026, for which results are final. In Kutchan, listed prices were ¥126,000 per room for resort hotels (N=10), ¥103,000 per house for vacation rentals (N=7), ¥65,000 per house for cottages (N=4) and ¥28,000 per room for business hotels (N=5). The all-type listed price for Kutchan was ¥78,000 (N=42), and the estimated transacted ADR was ¥62,000 (N=20). Niseko Town sits one step lower than Kutchan, with resort hotels at ¥79,000 (N=7), ryokan at ¥62,000 (N=4) and pensions at ¥55,000 (N=3); its all-type estimated transacted ADR is just ¥42,000 (N=10). Rusutsu shows a high all-type listed price of ¥154,000 and an estimated transacted ADR of ¥68,000, but the estimated transacted figure covers only N=2 properties, so it must be treated as a reference value.

The point to note here is how extremely small N is for winter vacation rentals. In Kutchan, 70 vacation rentals were listed in August 2026 and 75 in October, but prices could be observed for only 7 in January and 9 in February. In a Niseko winter, vacation-rental inventory either fills early or is sold through channels other than domestic OTAs, so we are seeing only the prices of houses that remain on domestic OTAs. The winter figure of ¥103,000 per house is therefore “the listed price of the houses that were left”, not the ceiling of the villa market as a whole. If anything, it is natural to assume the ceiling lies with the houses that dropped out of these listings.

Table 2: Listed prices and estimated transacted ADR by property type in three Niseko-area municipalities (per room, 2 guests, tax included; listed price = average of all plans; estimated transacted ADR = tax-excluded equivalent). January 2027 is the listing snapshot as of early October 2026. Categories with N<3 are hidden / Source: MetroEngines Research & Consulting

Figure 1: Listed prices by property type in Kutchan and Niseko Town (January 2026 actuals, per room or per house, 2-guest basis, tax included). Numbers above the bars are N / Source: MetroEngines Research & Consulting

In the January 2027 listing snapshot, resort hotels in Kutchan show ¥148,000 per room (N=11), vacation rentals there ¥108,000 per house (N=7), vacation rentals in Niseko Town ¥321,000 per house (N=4) and in Rusutsu ¥262,000 per house (N=6). Per-house prices for vacation rentals in Niseko Town and Rusutsu jump because large houses for big groups remain on sale; given the small N, they are best read as an indication of the ceiling. Since these listings are more than three months beyond the research date, the possibility that future price adjustments bring them down should also be factored in.

Figure 2: Distribution of properties by type in the Niseko area (Kutchan, Niseko Town, Rusutsu and others within a 22km radius). Circle size = number of rooms (vacation rentals and cottages count 1 house = 1 room); dark blue = resort hotels, ryokan and deluxe hotels; light blue = vacation rentals and cottages. N=541 properties / Source: MetroEngines Research & Consulting

Two axes, two different gaps — ¥100,000 per house is ¥26,000–¥34,000 per room

A vacation rental at ¥103,000 per house and a hotel room at ¥126,000 per room are both “the price of one night”, but in different units. Vacation-rental listed prices are aggregated on a 2-guest basis, yet the product is actually used by 4–8 guests, and a house often has 3–5 bedrooms. Dividing ¥103,000 per house by the number of bedrooms gives ¥34,000 per room for 3 bedrooms, ¥26,000 for 4 and ¥21,000 for 5. On a per-room axis, then, vacation rentals are sold at one-fifth to one-quarter of the resort-hotel level.

On a per-house axis, however, the ceiling is by no means low. Lining up the monthly listed prices of individual properties, vacation rentals in Kutchan in February 2026 (N=13) have a lower quartile of ¥70,000, a median of ¥116,000, an upper quartile of ¥216,000, and a high of ¥399,000. Resort hotels (N=10) have a lower quartile of ¥122,000, a median of ¥178,000, an upper quartile of ¥237,000 and a high of ¥325,000. Hotel rooms are higher at the median, but at the top of the distribution whole-house rentals exceed hotel rooms. This is what the price ceiling of the “sell it as one house” product type really is: by taking a large group as one party, a house reaches a rate that a single room cannot.

Table 3: Distribution of monthly listed prices by property in Kutchan (February 2026, per room, 2 guests, tax included). Estimated transacted ADR can be calculated only for resort hotels / Source: MetroEngines Research & Consulting

One more thing to pin down is the gap between listed prices and estimated transacted ADR. For resort hotels in Kutchan, the estimated transacted ADR was ¥87,000 against a listed price of ¥126,000 in January 2026, and ¥71,000 against ¥140,000 in February, so the transacted level is only 50–70% of the listed level. Vacation rentals have no estimated transacted ADR, so the same discount cannot be applied mechanically, but putting a listed price of ¥100,000 per house straight into a business plan as the rate is optimistic; assuming a listed-to-transacted gap similar to hotel rooms is the conservative approach. How wide the gap between listed and estimated transacted rates opens in snow-country winters is examined across five prefectures in Snow Country Winter ADR at 40–50% of Listed; Only Hokkaido Up in Feb.

Figure 3: Monthly listed prices for vacation rentals (per house) and resort hotels (per room) in Kutchan, January–December 2026 (October–December are estimates from current listings). Vacation-rental N swings widely, from 70 properties in summer to 7 in winter / Source: MetroEngines Research & Consulting

Figure 3 overlays the monthly listed prices of vacation rentals and resort hotels in Kutchan. Resort hotels show sharp seasonality, at ¥126,000–¥140,000 in January–February versus ¥52,000–¥59,000 in summer, with winter 2.4 times summer. Vacation rentals, by contrast, look higher in summer, at ¥122,000 (August, N=70) versus ¥103,000 in winter. But this is a composition effect: in winter the houses left on domestic OTAs are skewed toward a few small and mid-sized ones, while in summer more than 70 houses including large ones are listed. It does not mean the same house raises its price in summer. When looking at the price ceiling of detached products, house size and N need to be aligned.

Across prefectures, do detached products price above hotel rooms? — yes in snow resorts; level in onsen and beach areas

Is the phenomenon of detached products pricing above hotel rooms unique to Niseko, then? Lining up the ratio of vacation rentals (per house) to resort hotels (per room) on December 2026 listed prices across major resort areas, the answer is clear. Hakuba has vacation rentals at ¥244,000 (N=35) against resort hotels at ¥81,000 (N=19), or 3.0x; Niseko Town ¥215,000 (N=7) against ¥67,000 (N=6), or 3.2x; and even Hokkaido as a whole ¥76,000 for vacation rentals (N=317) against ¥47,000 for resort hotels (N=129), or 1.6x. In snow resorts, detached products clearly sit above hotel rooms.

In contrast, in onsen towns and beach resorts, detached products and hotel rooms are roughly level. Atami has vacation rentals at ¥69,000 (N=77) against resort hotels at ¥66,000 (N=30), or 1.04x; Onna ¥45,000 (N=79) against ¥46,000 (N=39), or 0.96x; Fujikawaguchiko ¥60,000 (N=146) against ¥65,000 (N=24), or 0.92x; and in Yufu, vacation rentals at ¥55,000 (N=95) sit below resort hotels at ¥79,000 (N=11), or 0.69x. Kusatsu is in between, with vacation rentals at ¥71,000 (N=13) against resort hotels at ¥52,000 (N=8), or 1.37x.

Table 4: Listed prices for December 2026 (per room, 2 guests, tax included, average of all plans; listing snapshot as of early October 2026). Vacation rentals counted as 1 house = 1 room / Source: MetroEngines Research & Consulting

Figure 4: Ratio of listed prices, vacation rental (per house) ÷ resort hotel (per room), December 2026. 1.0 = same level / Source: MetroEngines Research & Consulting

What this distribution shows is that the factor pushing up the rate ceiling of detached products is the combination of snow and long stays by large groups. Skiers stay a week at a time with family or friends and want a whole house in a ski-in/ski-out location. Here the price of a house is justified by the number of guests and days. In onsen towns and beach resorts, the unit of lodging has long been optimised for the 2-guest, 1-room ryokan/hotel model, so detached products are still positioned as “an option to stay cheaply in more space”. In other words, in Yufuin and Atami, the price band in which onsen-equipped detached products sit above resort hotels is currently empty.

Running costs of piping in onsen water, and investment per room equivalent

So what changes when onsen water is piped into a detached product? Let us check from the cost side. Atami’s municipal onsen supply for commercial Class A use has a base charge of ¥43,378 per month (base volume 80 cubic metres, tax included), or ¥521,000 a year. Assuming a house with an indoor bath and an open-air bath totalling 2 cubic metres of tub capacity, fully refilled every day, consumption is 60 cubic metres a month and stays within the base volume, so piped water costs about ¥520,000 per house a year. If 10 houses are combined under one contract, consumption is 600 cubic metres a month, about ¥4.92M a year including excess charges, or ¥490,000 per house — not much different. At 40% occupancy, or 146 nights a year, that is ¥3,400–¥3,600 per house per night, a cost ratio of 3–4% against a rate of ¥100,000 per house. Tariff structures differ by municipality and onsen cooperative, so Atami’s level cannot be applied as-is to other areas, but in order-of-magnitude terms it is far lighter than drilling. For details, see our earlier articles “Buying Onsen Water: Atami Tariffs” and “Drill, Pipe or Buy — Onsen Cost Math”.

Drilling your own well at a depth of around 1,000 metres requires an initial investment of roughly ¥71M–¥130M (Onsen drilling costs and payback years). Shared across 10 villas, that is ¥7.1M–¥13M per house, or ¥360,000–¥650,000 a year over 20 years — the same order of magnitude as ¥520,000 a year for piped water. In other words, onsen adds a few million to just over ¥10M per house, compared with the construction cost itself; what drives the investment decision is not the cost of onsen but how far onsen can support rates and occupancy outside winter.

Table 5: Indicative investment for one onsen villa (simple estimate excluding land, exterior works, furniture, and design and supervision) / Source: Compiled by MetroEngines Research & Consulting from archi-book (2025 compilation based on MLIT’s Building Construction Starts Statistics), Atami City “Onsen Tariffs”, and the editorial estimates in our earlier article on onsen drilling costs and payback years

As Table 5 shows, the investment for an onsen villa per room equivalent is ¥23M–¥39M, in the same order of magnitude as ¥24.3M–¥30.4M for a hotel room. The difference is not in the investment but in how it is sold. Requiring a 6% NOI yield (net operating income ÷ investment) on ¥100M per house (rough estimate: timber building + onsen share + exterior works) means ¥6M of NOI a year. At a GOP margin (gross operating profit ÷ revenue) of 35–45%, that requires revenue of ¥13.3M–¥17.1M, which at 40% occupancy (146 nights a year) means an annual average of ¥91,000–¥117,000 per house per night, and at 50% (183 nights) ¥73,000–¥94,000. Recalling that vacation rentals in Kutchan list at ¥103,000 in winter and ¥122,000 in summer, this level is reached on a listed basis; but if transacted rates fall to 50–70% of listed, it is hard for the winter peak alone to sustain the annual average. This is where onsen matters: if spring and autumn rates can be lifted 20–30% on the strength of onsen, the annual average reaches the required rate. This is a simple estimate; actual investment decisions require a detailed feasibility study.

Required rate as occupancy and GOP margin move — sensitivity table and three scenarios

The estimate above assumed ranges of 40–50% occupancy and 35–45% GOP margin. Table 5-2 shows how far the annual average rate that meets a 6% NOI yield changes when each of these two assumptions moves by one step. NOI is approximated as revenue × GOP margin, without deducting property tax, insurance or repair reserves.

Table 5-2: Annual average rate per house per night that meets a 6% NOI yield (¥6M a year) on ¥100M per house (simple estimate). NOI approximated as revenue × GOP margin. Shaded cells are reachable at or below the optimistic scenario’s transacted rate in Table 5-3 (¥85,000) / Source: Prepared by MetroEngines Research & Consulting

The required rate moves across a roughly 4x range, from ¥182,000 at 30% occupancy and 30% GOP margin to ¥47,000 at 70% occupancy and 50% GOP margin. At 40% occupancy and 40% GOP margin, raising either by 10 points brings the required rate down by about 20%, from ¥103,000 to ¥82,000. Rather than trying to solve it with rate alone, a design that lowers the required rate through both operating days and operating efficiency is what works.

Next, let us set the rate side in three scenarios. Applying the listed-to-transacted gap observed at resort hotels in Kutchan (50–70%) to the town’s vacation-rental listed prices (¥103,000 in winter to ¥122,000 in summer) gives an annual average transacted rate in the range of ¥52,000–¥85,000.

Table 5-3: Three scenarios for one onsen villa (investment ¥100M; simple estimate). Transacted rates are assumptions obtained by applying the listed-to-transacted gap observed at Kutchan resort hotels (50–70%) to the town’s vacation-rental listed prices; they are not actual results / Source: Prepared by MetroEngines Research & Consulting

The base scenario’s NOI yield is 4.5%; the gap to 6%, translated into the annual average rate, is about ¥23,000 (from ¥68,000 to ¥91,000). The optimistic scenario reaches 7.0%. How much of this ¥23,000 can be filled by rates and occupancy in the months outside winter is the point of piping onsen into a detached product.

The Niseko-area pipeline — branded villas have started moving as for-sale products

How onsen-equipped detached products are actually being planned in the Niseko area is shown by ONE HANAZONO VILLAS, jointly announced on September 1, 2026 by Hyatt Hotels & Resorts and Pacific Century Premium Developments (PCPD). According to the two companies’ press release, a total of 13 standalone residences will be developed in the Niseko area of Hokkaido as the first Park Hyatt-branded villas in Japan; each house will be about 700–1,000 square metres with 4–6 bedrooms, and every house will have onsen. Two model villas are scheduled for completion in winter 2026, and operations will be handled by the team at Park Hyatt Niseko HANAZONO (100 rooms), which opened in January 2020. The project was announced as a for-sale product for owners (residences), and no prices as rental inventory have been published, so it is not included in this article’s listed-price comparisons. Note that this villa plan is not included in Hyatt Japan’s “Press Releases, April 2026”; the information is based on the September 2026 announcement.

Why for-sale models are going first can be read from the previous section’s estimates. Recovering an investment of over ¥100M per house through rental operation alone requires an annual average rate of ¥90,000–¥120,000 per house, but a for-sale model recovers the construction cost before opening, and the operating team can probe the rate ceiling by renting out the periods owners don’t use. As Table 6 shows, the planned projects we can confirm in the Niseko area are mostly hotels; detached projects are too small to show up readily in building-permit statistics (for the current state of high-end hotel supply and its gaps, see Niseko’s Top Tier: Just 2 Hotels, 75 Rooms — Where Fairmont’s 165 Fit).

Table 6: Planned projects in the Niseko area (scheduled completion October 2026 or later, or undecided; duplicate registrations on the same site merged by GFA, floors and completion). *Based on building-confirmation applications at the time of research. The count is expected to rise as applications are added; use as the lower bound of the confirmed pipeline at present / Source: Prepared by MetroEngines Research & Consulting from MLIT’s Building Dynamics Statistics Survey

What stands out in Table 6 is the project scheduled for completion in June 2029: “30 rooms, 2 floors, 10,000㎡ GFA”. A floor area of 333 square metres per room is a different order of magnitude from the 40–50 square metres of a hotel room, suggesting a detached or large-suite concept. With the project name undecided, nothing can be said for certain, but it can be read as a sign that new supply in the Niseko area is broadening from stacking hotel rooms toward product types that allocate generous floor area per unit.

Detached products now lead new supply — 828 of the 1,717 properties opening in 2026 are vacation rentals

Detached products deserve a price-ceiling discussion because they are not an exceptional product; they already lead new supply. MetroEngines Research & Consulting tracks 1,717 properties opening in 2026 on an OTA-listing-confirmed basis, with an average of 24 rooms; by property type, vacation rentals account for 828, or 48.2% of the total. They are followed by business hotels (210), hostels (125), resort hotels (100), guesthouses (91) and cottages (79); vacation rentals and cottages together make 907 detached properties, or 52.8%. By prefecture, Hokkaido has the most at 167, followed by Tokyo with 103, Okinawa 96, Shizuoka 95, Kyoto 94 and Nagano 74.

Figure 5: Properties opening in 2026 by property type (OTA-listing-confirmed basis, N=1,717 properties, as of early October 2026). OTA listings appear several months before opening, so counts for recent and later months may rise with future listings / Source: MetroEngines Research & Consulting (OTA-listing-confirmed basis, N=1,717 properties)

Note, however, that this is a count of properties, not rooms. Most of the 828 vacation rentals are one listing per house, so in room terms the 100 resort hotels are larger. Even so, the fact that more than 800 operators facing the decision of what price to sell one house at have emerged in 2026 alone carries weight. The rate ceiling of detached products is not a story about a few luxury projects; it has become a common theme that determines the earnings of small operators nationwide.

Areas where the same product could work — onsen towns with a deep high-rate tier where detached products have not yet risen above it

The conditions for the onsen-villa product type to work can be broken into two. One is that the area already has a deep high-rate tier. The other is that detached products have not yet reached that high-rate tier. Without the former there is no basis for the rate; without the latter there is no gap. Overlaying the ratios in Table 4 with the level of resort-hotel estimated transacted ADR brings candidate areas into view.

Table 7: Depth of the high-rate tier and position of detached products in candidate areas (estimated transacted ADR is an estimate based on the December 2026 listing snapshot; ratios from Table 4) / Source: MetroEngines Research & Consulting

Yufu has a deep high-rate tier, with 110 ryokan listing at ¥59,000 and 11 resort hotels at ¥79,000, while its 95 vacation rentals sit below the ryokan at ¥55,000. Its basis for rates as an onsen town is ample, making it the area with the largest gap for onsen-equipped detached products to price above. In Kusatsu, 13 vacation rentals at ¥71,000 are starting to emerge above the ¥47,000 tier of 52 ryokan, already demonstrating that onsen-equipped detached products can work. Atami has an institutional entry point in municipal piped onsen supply, and 5 deluxe hotels set a ceiling of ¥214,000, so the ceiling that a whole-house product would aim for is visible. In Hakuba, detached products already sit 3x above, and onsen plays the role of supporting rates in the months outside winter. In every area, the figures in this article indicate room for growth; they do not evaluate the pricing of individual properties.

Operational room that whole-house products have — cleaning routes, check-in, and management contracts

The price-ceiling discussion needs to be paired with the operating-cost side. A whole-house product earns more revenue per party than a hotel room, but cleaning involves travel between houses, check-ins are dispersed, and equipment inspections multiply with the number of houses. There are three areas of room here. First, cleaning routes: arranging 10 houses radially around a single clubhouse or back-of-house keeps cleaning staff travel time down. ONE HANAZONO VILLAS’ design, with an owners-only clubhouse at its core, is also rational from an operating-route perspective. Second, check-in: combining smart locks with online pre-registration lets on-site handling be consolidated even across dispersed houses. Third, management contracts: existing hotel and resort operating teams taking on nearby detached properties under contract is already common in the Niseko area, and it is a natural extension for ryokan adding annexes or villas in onsen towns too.

When an existing ryokan adds annexes or villas, if it can pipe water from the main building’s source, the additional onsen cost is limited to piping and tubs, and the drilling share in Table 5 does not apply. Sharing the main building’s large baths and meals while selling a house to one party lets the product sit in a price band above the main building, so for a ryokan it becomes a way to create a new rate ceiling higher than its main-building rooms. The depth of the ryokan tier in Yufu and Kusatsu can be read as a foundation for this add-on type of onsen villa to work.

Summary

The Niseko-area winter showed a stepped gap: ¥126,000 per hotel room, ¥103,000 per vacation-rental house and ¥65,000 per cottage. On a per-room axis detached products are one-quarter of hotels, but on a per-house axis they exceed hotel rooms at the top of the distribution. Across prefectures, detached products price above hotel rooms only in snow resorts — 3.0x in Hakuba and 3.2x in Niseko Town — while in onsen towns and beach resorts they are level at 0.7–1.4x. Piping in onsen water costs around ¥500,000 per house a year, and even shared drilling is ¥7M–¥13M per house — not large enough to sway the investment decision. Investment per room equivalent is also in the same order of magnitude as a hotel room.

The value of the onsen-villa product type, therefore, lies neither in construction cost nor in onsen cost, but in whether it has a basis for rates in the months outside winter and can lift the annual average to the required level of ¥90,000–¥120,000 per house. Onsen towns like Yufu, Kusatsu and Atami, with a deep high-rate tier that detached products have not yet risen above, have the most room. In the Niseko area, for-sale branded villas have started moving first, and prices as rental inventory are yet to be observed.

Frequently Asked Questions

Q. How many guests does a vacation rental’s “per house” price cover?
A. Listed prices in this article are aggregated on a 2-guest, 1-room basis. Vacation rentals are actually often used by 4–8 guests, and additional-guest charges are not included. Per-room figures are approximations divided back by the number of bedrooms.

Q. Why is there no estimated transacted ADR for vacation rentals?
A. Estimated transacted ADR is calculated only for property types whose correction factors have been validated against property-level results disclosed by REITs. Vacation rentals and cottages have no disclosed results to check against, so they are treated with listed prices only.

Q. Are the sale prices or room rates of ONE HANAZONO VILLAS known?
A. The September 1, 2026 announcement does not state sale prices or room rates. It was announced as a for-sale product for owners and is not included in this article’s listed-price comparisons.

Source: Hyatt Japan, “Press Releases, April 2026”