Autumn is the season for walking tourism. On both the Kumano Kodo and the Shikoku Henro the mode of travel is on foot, and a day’s progress generally falls between 15 and 25 km. For these two pilgrimage routes, lodging is therefore not an accessory to the tourism resource — it is the infrastructure that makes the itinerary possible at all. So how much capacity actually exists? Using OTA-listed properties tracked by MetroEngines Research, we counted properties and rooms within 3 km, 5 km and 10 km of 10 Kumano Kodo waypoints and 9 Shikoku Henro temples, and measured October–November room rates and the share of foreign-language guest reviews.

A walkable day is 15–25 km — first, count the inns at each waypoint

The Nakahechi, the best-known walking route on the Kumano Kodo, runs roughly 38 km from Takijiri-oji to Kumano Hongu Taisha. The model course on Wakayama Prefecture’s official tourism site puts Takijiri-oji to Chikatsuyu-oji at 13 km, and the following leg from Chikatsuyu-oji to Kumano Hongu Taisha at 25.7 km. The Shikoku Henro is much the same: the notorious “henro-korogashi” from No. 11 Fujii-dera to No. 12 Shosan-ji covers about 12 km with 1,150 m of elevation gain in a single climb. On a walking journey the itinerary is set not by “how far can I get today” but by “where can I sleep tonight.”

So for all 19 points — 10 Kumano Kodo waypoints and 9 Shikoku Henro temples — we counted OTA-listed properties within a 3 km, 5 km and 10 km radius. The results are in the table below: the Kumano side totals 89 properties and 2,958 rooms within 3 km, the Shikoku side 15 properties and 382 rooms. Of the 19 points, 5 had no lodging at all confirmed within 5 km.

Source: MetroEngines Research (OTA-listing confirmed basis); compiled by the HotelBank Editorial Team. Waypoint coordinates are representative points based on publicly available location information for each shrine, temple and hot-spring district. * Figures for Takijiri-oji and Chikatsuyu-oji are withheld pending recalculation, as an error was found in their representative-point coordinates.

Source: MetroEngines Research (OTA-listing confirmed basis); compiled by the HotelBank Editorial Team

The Nakahechi mountain section is thinly served — and the gap is also room to build

The clearest numbers come from the mountain section of the Nakahechi. The inns listed on OTAs around Takijiri-oji and Chikatsuyu-oji are mostly minshuku and guesthouses of 1 to 13 rooms, and no substantial room stock can be confirmed. That is the scale of publicly listed inventory a walker can choose from after finishing a 13 km day.

This thinness does not mean there is no demand. If anything the opposite: capacity is not concentrated at the points where an itinerary forces an overnight stay. Small inns along the Nakahechi that take bookings only through travel operators do exist, so the property counts here are limited to what is listed on OTAs. Even so, a publicly bookable room count in the low double digits marks clear room to enter for small-scale lodging businesses, kominka (old folk house) conversions and whole-house rentals. The nationwide structure of municipalities with thin lodging inventory is set out in our article aggregating municipalities with one hotel or fewer.

Hongu, Yunomine and Kawayu form one walkable cluster — 15 properties, 324 rooms, median 9 rooms

Around Kumano Hongu Taisha the picture is different. Hongu, Yunomine Onsen and Kawayu Onsen all sit within 2–3 km of each other, and merging their three 5 km radii while removing overlaps gives 15 properties and 324 rooms. By type that is 8 ryokan, 3 minshuku, 2 guesthouses, 1 farm-stay minshuku and 1 pension, with a median of 9 rooms per property and a mean of 21.6. Two ryokan of 90 and 78 rooms account for more than half the room count, and the rest are single-digit-room properties.

This “two large boxes plus many small ones” shape suits a walking route well. Even a property that can offer only 2 to 9 rooms on its own contributes to real area capacity when a dozen or more inns line up within walking distance, making it easier to serve the multi-night stays, luggage transfer and onsen-hopping that pilgrimage routes generate. In fact the 3 km radius of Yunomine Onsen and that of Kawayu Onsen both return exactly 15 properties and 324 rooms — statistically, they behave as a single cluster.

Shikoku’s mountain temples have zero lodging within 5 km — how to bridge a 100 km leg

The Shikoku side is starker still. At five temples — No. 1 Ryozen-ji, No. 12 Shosan-ji, No. 21 Tairyu-ji, No. 60 Yokomine-ji and No. 66 Unpen-ji — not a single OTA-listed inn could be confirmed within 5 km. Distances to the nearest inn are 5.2 km at Ryozen-ji, 5.7 km at Shosan-ji, 5.7 km at Unpen-ji, 6.1 km at Yokomine-ji and 9.6 km at Tairyu-ji. Many of these mountain temples sit at around 700 m elevation, and on legs where the visit alone takes half a day, the working assumption is that walkers descend to the valley before looking for a room.

Capacity is thickest, conversely, at No. 38 Kongofuku-ji on Cape Ashizuri, with 8 properties and 237 rooms within 3 km. This is the end point of the leg that Kochi Prefecture’s official tourism website Yosakoi Net describes as “said to be about 100 km from the previous temple, Iwamoto-ji — the longest stretch between any two temples.” Inns clustering at the end of a long leg while the mountains in between stay thin captures the character of the Henro route’s lodging stock well.

Widening to a 10 km radius around the 9 temples gives 53 properties and 1,461 rooms, dominated by 19 ryokan and 18 business hotels. That contrasts with the Kumano side, where 24 temple lodgings (shukubo) at Koyasan fall inside the 3 km radius: on Shikoku, the “commute to the temple from a business hotel in town” pattern shows up directly in the inventory mix.

Three cases at 15 km, 20 km and 25 km a day — how far the gaps eat into an itinerary

So far we have counted inventory by waypoint. What really matters on a walking journey is not volume itself but whether an inn is reachable within that day’s itinerary. So for the five temples with no lodging confirmed within 5 km, we estimated how much of a walking day the distance to the nearest inn consumes, setting the daily walkable distance at 15 km (conservative), 20 km (standard) and 25 km (strong walker) in three cases. On pilgrimage routes the standard pattern is to descend to an inn and return to the same point the next morning, so both the outbound and return legs are counted as added walking.

Source: Calculated from the waypoint stock table in this article (Table 1). Added round-trip walking = distance to the nearest inn × 2. Compiled by the HotelBank Editorial Team.

The result is uniform. Even in the most permissive strong-walker 25 km case, the added round trip accounts for 41% or more of the day’s itinerary at all five temples. In the conservative 15 km case it reaches 69% at No. 1 Ryozen-ji and 128% at No. 21 Tairyu-ji, exceeding the assumed daily walking distance outright. In other words, on these legs the shortage of inns is not merely a matter of limited choice — it is at a level that forces the day’s itinerary itself to be redesigned. Widen to a 10 km radius and the inventory does exist — 557 rooms at No. 1 Ryozen-ji, 243 rooms at No. 66 Unpen-ji — but that also means it sits beyond walking reach.

In practice, however, many inns along the Henro route offer pickup and drop-off, and where that is available this added walking does not occur. The estimate above is an upper bound for a walk-only itinerary; whether a property offers transfers, and its service hours, are not captured in this tabulation. Even so, the fact that five temples are connected only through a transfer operation points, from the supply side, to exactly where measures such as small inns with transfers, whole-house rentals and leg-by-leg taxi partnerships would work best.

October–November rates — flat on the Kumano side, double-digit gains on Shikoku

Next, rates. To compare only months on the same basis, we placed the estimated transacted ADR for October and November 2024 alongside October and November 2025, both now finalised. The split is clear. On the Shikoku side, Takamatsu rose 20.0% year on year in November, Matsuyama 18.4%, Saijo 16.5% and Kanonji 10.9% — all double digits. The Kumano side, by contrast, was close to flat: Tanabe +1.5% in October and +3.3% in November, Shingu -0.5% and +1.8%.

Source: MetroEngines Research; compiled by the HotelBank Editorial Team. N is the number of properties covered by the estimated transacted ADR calculation (minimum). Koya is not shown, as only one property qualified.

Source: MetroEngines Research; compiled by the HotelBank Editorial Team

Overlaying Tanabe’s monthly trend year by year shows why. Tanabe’s estimated transacted ADR spikes in August (¥14,900 in August 2025), while October and November sit at roughly the annual average (¥11,800 in 2025) — autumn produces no peak. The season that suits walking best is still flat in pricing terms. There is room here to extend the high season from August into October and November. Autumn on the Kumano Kodo brings stable weather and is the easiest time of year for planning an itinerary, and in demand-quality terms it is a segment that can carry a higher rate.

Source: MetroEngines Research; compiled by the HotelBank Editorial Team (estimated transacted ADR, finalised months only)

For reference, the levels listed on OTAs for October and November 2026 as of the survey date are shown below. These are a work in progress that will move with future sales, and they rest on different calculation assumptions from the year-on-year figures above, so please do not subtract one from the other directly.

Source: MetroEngines Research; compiled by the HotelBank Editorial Team. N is the number of properties covered.

Foreign-language review share: Koyasan 65.3%, Hongu area 44.7%, Henro area 4.5%

Where does the “walking autumn” clientele come from? We aggregated guest reviews by language and compared the non-Japanese share by area. Koyasan’s 3 km radius is highest at 65.3% (N=3,631 reviews), followed by the Kumano/Hongu area at 44.7% (N=3,507). The 3 km radii of Tanabe, Shingu and Katsuura — Kumano’s gateways — come to 18.7% (N=10,665). Against that, the 10 km radius around the 9 Shikoku Henro temples was 4.5% (N=5,347). For comparison, a random national sample of ryokan, minshuku, guesthouses and similar properties gives 10.0% (166 properties, N=21,255).

Source: HotelBank Editorial Team research (NLP analysis of guest reviews, most recent 24 months)

Source: HotelBank Editorial Team research

The language mix is equally contrasting. At Koyasan, alongside 1,238 English reviews sit 246 in Italian, 232 in French, 214 in German and 165 in Spanish — a thick layer of European languages. The Kumano/Hongu area likewise leads with 938 in English, plus 116 in French and 73 in Spanish. The 9 Shikoku Henro temple areas, by contrast, show 5,106 in Japanese against 123 in English and 35 in Korean, below the national average. Note that language reflects the language a review was posted in rather than nationality itself, so it does not break down where guests live.

For the Shikoku Henro, that gap is straightforward upside. The product characteristic — a pilgrimage you walk — is shared with the Kumano Kodo, and the fact that European-language reviews have accumulated on the Kumano side is evidence that the same experience can be valued in overseas markets. Shikoku has domestic demand as its base and is at the stage where overseas awareness is only now beginning to build. Length of stay among inbound visitors and the reception structure by property type are covered in our article comparing average inbound stay length by prefecture and our article on inbound share by property type.

Conclusion — capacity designed as an area, not a property, sets the rate

To draw the measurements together. First, pilgrimage lodging stock is not evenly spread: it gathers at the gateways and end points and thins out in the mountains. Against 89 properties and 2,958 rooms within 3 km of the 10 Kumano Kodo waypoints, the 9 Shikoku Henro temples have 15 properties and 382 rooms within 3 km, and five temples had no OTA-listed inn confirmed within 5 km. Second, at points like Hongu, Yunomine and Kawayu, where several small inns line up within walking distance, real area capacity forms even at a median of 9 rooms per property. Third, October–November rates rose by double digits on the Shikoku side while the Kumano side tracked last year’s level, leaving room to price in the walkable autumn.

A walking trip is judged by the whole itinerary, not by a single night. That is precisely why inns here are less likely to end up competing with the one next door: connecting capacity leg by leg lifts both the area’s total capacity and its rates. Seen that way, the thinly served legs are the easiest upside to act on.

Methodology and sources

Property and room counts cover OTA-listed properties tracked by MetroEngines Research, limited to those registered as operating and with an observed selling price within the past 120 days (441 of 605 properties within 20 km of any of the 19 waypoints). This is not a complete census; inns not listed on OTAs and inns that take bookings only through travel operators are excluded. Distance is judged by surface distance from each waypoint’s representative point and does not match administrative boundaries. Representative-point coordinates are based on publicly available location information for each shrine, temple and hot-spring district.

Rates use municipality-level estimated transacted ADR, and year-on-year figures are calculated only between finalised months (October and November 2024 vs 2025). Koya, where only one property qualified, is excluded from the rate tabulation. Review language tabulation covers the most recent 24 months; the national comparison sample randomly drew 200 properties with 50 or more reviews from ryokan, minshuku, guesthouses, pensions and farm-stay minshuku, of which the 166 with confirmed reviews were tabulated.