When MetroEngines Research’s estimated settled ADR (tax-excluded equivalent) for Hokkaido is set side by side — finalized September 2026 values against finalized values for the same month a year earlier — the picture splits by hotel segment. Business hotels rose +8.9%, to ¥10,800 (N=437) from ¥10,000 (N=419) a year earlier, while city hotels were essentially flat at +0.1%, at ¥16,500 (N=75) versus ¥16,500 (N=74). In August, city hotels had come in at ¥17,800 (N=75), −11.3% below the prior year’s ¥20,100 (N=74), so September erased that drop and brought them back to last year’s level. Ryokan (−0.2%, N=294) and resort hotels (+0.8%, N=147) also landed in line with the prior year. This article breaks down the year-on-year differences between finalized months by segment, checks whether the same conclusions hold when the set of properties is held fixed, and then lays out the “baselines” worth referring to for pricing decisions from autumn into winter.

Coverage: Hokkaido city hotels N=75, business hotels N=437, ryokan N=294, resort hotels N=147 (number of properties in the September 2026 sample). The price metric in this article is estimated settled ADR (the level of achieved prices estimated from OTA and other sales data, tax-excluded equivalent). Definitions are at the end of the article. Data as of October 28, 2026.

August and September YoY — The Four Segments Did Not Move Together

First, we line up the year-on-year changes for the finalized months of August and September across the four segments. In August, only business hotels were above the prior year at +2.5% (N=438, prior year N=417); the other three segments all fell short — city −11.3%, resort −6.9% (N=147, prior year N=141), and ryokan −3.7% (N=293, prior year N=280). City hotels in particular had risen a steep +16.8% in August of the previous year versus the year before, and that high comparison base showed up in the size of this August’s decline.

In September, however, the picture changes. Business hotels widened their gain to +8.9%, while city hotels, ryokan, and resorts all landed within −0.2% to +0.8%, nearly erasing their August declines. In other words, Hokkaido in September is best read as a two-tier structure: “only business hotels were well above last year; everything else was at last year’s level.” Segment moves as of the August close and summer RM tactics are covered in Hokkaido August Settled ADR: City −11.3%, Business +2.5%.

Source: MetroEngines Research; compiled by HotelBank Editorial Team (Hokkaido, estimated settled ADR, tax-excluded equivalent, area median of the properties sampled each month)

One caveat: the number of properties in the sample is not the same this year and last. For business hotels, the September sample grew from 419 to 437 properties, and for ryokan from 275 to 294; since June 2026, the sample has been larger than the prior year’s in month after month. A year-on-year difference in medians contains both “price moves at the same properties” and “turnover in the property set,” so the next section checks the results with the properties held fixed.

Fixed-Sample Check — The Conclusions Hold, but Ryokan Show “Flat Median, Rising Individual Prices”

We fixed the sample to properties with finalized values in all four months — August and September 2025 and August and September 2026 — and recalculated YoY on the medians of that same set. The fixed sample comprises 73 city hotels, 396 business hotels, 257 ryokan, and 139 resort hotels.

Source: MetroEngines Research; compiled by HotelBank Editorial Team (fixed sample: city N=73, business N=396, ryokan N=257, resort N=139)

The September conclusions barely move. Comparing fixed-sample medians, city hotels went from ¥16,500 to ¥16,500 (+0.0%), business hotels from ¥10,300 to ¥10,900 (+6.5%), and resorts from ¥16,100 to ¥16,200 (+0.8%) — the same direction as the full sample. The business-hotel gain narrowing from +8.9% to +6.5% suggests that newly added properties were nudging the median up somewhat. Even so, 74.5% of the fixed sample was above the prior September, confirming that the increase was not concentrated in a handful of properties.

For city hotels in August, on the other hand, the fixed sample shows a deeper fall than the full sample’s −11.3%: from ¥20,600 to ¥17,800, or −13.6%. Only 28.8% of fixed-sample city hotels beat the prior August, meaning the August decline was spread across the whole segment. Combined with the share of properties above the prior year recovering to 49.3% in September, city hotels’ August can be summarized as a “temporary dip” and September as a “return to last year’s level.”

Ryokan require careful reading. The fixed-sample median was flat in September, from ¥10,100 to ¥10,100 (−0.3%), but the median of property-level YoY changes was +4.5%, and 64.6% of properties were above the prior September. Even though the area median did not move, most individual ryokan were selling at higher levels than a year earlier. For ryokan operators, judging “flat versus last year” from the area-median YoY alone risks missing the price increases happening around them.

Source: MetroEngines Research; compiled by HotelBank Editorial Team (properties with finalized estimated settled ADR in all four months: Aug and Sep 2025 and Aug and Sep 2026)

City Hotels — Last Year’s Strong Autumn and Winter Become a “Comparison Wall”

To see why city-hotel YoY sank into the summer, we look at three years of monthly trends. Through almost all of 2025, city hotels ran well above 2024: October was ¥14,700 (+25.4% YoY), December ¥13,600 (+24.6%), and January 2026 ¥14,700 (+29.4%) — increases of more than 20% persisted through autumn and winter. After that, YoY slipped into small negatives — −0.9% in April 2026, −1.4% in May, −3.6% in June, −5.7% in July — before bottoming out at −11.3% in August.

Source: MetroEngines Research; compiled by HotelBank Editorial Team (Hokkaido city hotels, estimated settled ADR, N=73–76 per month. October–December 2026 are estimates based on current sales and may change)

As the chart shows, the 2026 line pulled well away from the 2025 line in August and then rejoined it in September. What to keep in mind is that the coming October–December period corresponds to last year’s upswing. October–December 2025 were all 15–25% above 2024, so city hotels in autumn-winter 2026 face a high bar on a YoY basis. A negative YoY does not necessarily mean weak demand; reading it in two steps, together with 2024 levels, gives a clearer picture of what is actually happening.

Business Hotels — Resilient, Above the Prior Year for 12 Straight Months

Business hotels stand in contrast to city hotels. Over the 12 months from October 2025 through September 2026, YoY never turned negative, ranging from +0.3% at its lowest (April 2026) to +14.4% at its highest (January 2026). September’s +8.9% matches February and March 2026 (+8.1% to +8.2%) and marks a re-acceleration from the summer’s +2.5% to +4.1%.

Source: MetroEngines Research; compiled by HotelBank Editorial Team (Hokkaido business hotels, estimated settled ADR, N=410–441 in finalized months. October–December 2026 are estimates based on current sales and may change)

The three lines for 2024, 2025, and 2026 keep nearly the same seasonal shape while shifting upward each year. The fact that the summer peak and winter trough keep their shape while only the level rises indicates that the price increase is not concentrated in particular peak periods but is progressing as a year-round lift. Whereas city hotels are absorbing the rebound from last year’s sharp rise, business hotels’ gains were relatively moderate, which arguably makes it easier for YoY gains to accumulate. For which days of the week carry demand for Hokkaido business hotels, see Hokkaido business hotels by day-of-week estimated OCC.

We also check the 12-month trends for ryokan and resorts. Ryokan rose sharply in winter, including +31.6% in December 2025 and +28.4% in January 2026, and stayed positive through June (+1.4%) before falling below the prior year in summer at −3.3% in July and −3.7% in August, then returning to last year’s level at −0.2% in September. Resorts posted a run of declines from −4.0% to −6.9% from May through August 2026, then recovered to +0.8% in September. In terms of a large rise last winter, ryokan are in a position similar to city hotels, whereas resorts’ YoY last winter stayed within +0.1% to +6.9%.

Source: MetroEngines Research; compiled by HotelBank Editorial Team (Hokkaido, YoY change in estimated settled ADR; all comparisons are between finalized months. The number of properties sampled varies by month: city N=73–76, business N=414–441, ryokan N=241–297, resort N=123–147)

Autumn-Winter Baselines — Prior-Year Finalized Values Alongside Current Estimates

From October onward, values are not yet finalized, so YoY cannot be calculated. Instead, we set the prior year’s finalized value for each month as a “baseline” yardstick and place the estimate based on current sales next to it for reference. Estimates may change toward month-end, so a straight comparison with finalized values has to wait until the month closes.

Source: MetroEngines Research; compiled by HotelBank Editorial Team (October–December 2026 are estimates based on current sales and may change. A straight comparison with finalized values must wait for month-end close)

The key to reading these baselines is that each segment’s prior-year finalized value sits at a different point in terms of “how much it had already risen.” For city hotels and ryokan, October–December 2025 was in an upswing of +14.9% to +31.6% versus two years earlier. By contrast, business hotels rose +9.3% to +11.4% over the same period and resorts −5.4% to +10.5%, relatively smaller moves. When using the prior-year finalized value as a yardstick, keep in mind whether the yardstick itself sits high or at a more moderate level. The gap between winter listed prices and estimated settled rates is analyzed in detail in our winter hotel rate comparison across five snow-country prefectures.

For Revenue Managers Running City Hotels, Business Hotels, Ryokan, and Resorts in Hokkaido — Implications and Action Plan

For all of these actions, the figures in this article are Hokkaido-wide segment medians. The right level for your own property differs by location and room mix, so use the baselines as a starting point for decisions and make final settings in light of your own booking pace.

Summary — Three Yardsticks for Reading YoY by Segment

The first yardstick is “YoY between finalized months.” In September 2026, Hokkaido showed business +8.9%, city +0.1%, ryokan −0.2%, and resort +0.8% — a two-tier structure in which only business hotels were well above last year. This yardstick also captures city hotels falling to −11.3% in August and returning to last year’s level in September.

The second yardstick is “the fixed-sample check.” With the property set held fixed, the conclusions did not change: business +6.5%, city +0.0%. For ryokan, however, 64.6% of properties were above the prior year even though the median was flat, showing the need to look at both the median and the distribution.

The third yardstick is arguably “the height of the baseline.” Prior-year finalized values for autumn-winter sit at levels reached after large increases last year for city hotels and ryokan, and after relatively moderate increases for business hotels and resorts. Once October and later months are finalized, reading YoY alongside the change versus two years earlier will capture price changes more accurately.

Frequently Asked Questions

Q. How did Hokkaido’s estimated settled ADR in September 2026 compare with the prior year?

Business hotels were ¥10,842 (N=437), up +8.9% from ¥9,960 in the prior September; city hotels were ¥16,485 (N=75), +0.1%; ryokan ¥9,764, −0.2%; and resort hotels ¥16,198, +0.8%. It is a two-tier structure in which only business hotels were well above last year while the other three segments were at last year’s level.

Q. Doesn’t the difference in the number of properties sampled versus last year affect the results?

Even when the sample is fixed to properties with finalized values in all four months of August and September 2025 and August and September 2026, September shows business +6.5% (N=396), city +0.0% (N=73), and resort +0.8% (N=139) — the same direction as the full sample. The business-hotel gain narrows by the amount attributable to turnover in the property set.

Q. How can the ryokan median be flat while “most raised prices”?

The median of the 257 fixed-sample ryokan was flat at −0.3%, but the median of property-level YoY changes was +4.5%, and 64.6% of properties were above the prior September. Even if the price of the property at the median did not change, properties that raised prices were the majority across the distribution.

Q. Should city hotels’ −11.3% in August be seen as a problem specific to my property?

Only 28.8% of fixed-sample city hotels beat the prior August, so the decline was spread across the whole segment. The rebound from a high prior-year August (+16.8% versus two years earlier) was a major factor, and in September the share of properties above the prior year recovered to 49.3%.

Q. What should I use as a benchmark for pricing decisions in October–December?

Set the prior year’s finalized value for the same month as the baseline and place the current estimate next to it for reference. For city hotels and ryokan, October–December 2025 sits at a level reached after rises of +14.9% to +31.6% versus two years earlier, so once values are finalized, checking both YoY and the change versus two years earlier makes it easier to separate apparent declines from real changes in the market.

About the Data

Source: MetroEngines Research; compiled by HotelBank Editorial Team