In Ishikawa, city hotels and business hotels moved in opposite directions on estimated settled ADR during the first half of 2026. Comparing final months against each other (January–July 2026 vs the same months of 2025), city hotels came in at a day-weighted average of roughly ¥12,000 (¥11,600 a year earlier, +3.6%), above the prior year, while business hotels landed at roughly ¥7,500 (¥7,600 a year earlier, −1.3%), marginally below. The gap between the two has widened broadly since March, and by July the difference in year-on-year change reached 14.2 points (city +7.4% vs business −6.8%). Why did the same prefecture, over the same seven months, split this far apart? Using final figures and the most recent booking curves, we build three “yardsticks” that can be applied when resetting rate ranges for the second half.

Coverage: Ishikawa city hotels N=10 properties / business hotels N=109–110 properties (estimated settled ADR, final months); occupancy covers city N=10 properties and business N=98–103 properties. The price metric in this article is estimated settled ADR (a settled price level estimated from OTA and similar sales data, tax-exclusive equivalent); occupancy is an estimate based on OTA-listed inventory. Definitions for both appear at the end of the article. Data as of 4 August 2026.

H1 2026 in final months — city turned in March, business sank again in June

Start with only the final months for which a year-on-year comparison is available (January–July 2026). Both city and business hotels share one feature: they opened the year below the prior year. City hotels were at about ¥9,200 in January 2026 (about ¥9,900 a year earlier, −7.3%) and −7.6% in February. Business hotels were also soft, at −0.9% in January and −3.8% in February.

The turning point was March. City hotels flipped above the prior year at +4.3% in March and have not returned to negative territory through July. May recorded the largest gain of the first half at +11.6%, roughly ¥15,700 against about ¥14,000 a year earlier, and the category held positive year-on-year even in demand-trough months, at +7.8% in June and +7.4% in July. Business hotels, by contrast, did move above the prior year in April (+3.9%) and May (+2.8%), but slipped back into negative territory at −4.7% in June and −6.8% in July.

Ishikawa, estimated settled ADR (tax-exclusive equivalent). All comparisons are final figures against final figures (based on past actuals). City N=10 properties (10 properties in both 2025 and 2026); business N=109–110 properties (107–111 properties in 2025). “Gap” is the city YoY minus the business YoY, in points. The weighted average is weighted by the number of days in each month. City figures are a median-based aggregate of 10 properties, so the level should be read as a range rather than a point.

Source: Compiled by the HotelBank Editorial Team from MetroEngines Research.

The column worth watching is “Gap” on the far right. From a floor of −6.4 points in January, it ran −3.8 in February → +4.9 in March → +8.8 in May → +12.5 in June → +14.2 in July, widening throughout, with one narrowing in April (+0.7 points). Averaged across the first half it is +4.9 points, but that average simply smooths together “balance in the early months” and “divergence in the later ones”. The current reality is far more split than the average suggests.

Laying the years on top of one another shows that the seasonal shape of city hotels has itself changed. In 2024, October and November were the annual peaks at ¥17,112 and ¥17,900; in 2025 those levels fell to ¥14,126 and ¥15,586. In 2026 the year opened below the prior year, but from March onwards the line runs above the 2025 curve, and by May it had recovered to ¥15,680 — above even May 2024 (¥14,984).

Estimated settled ADR for Ishikawa city hotels (N=10–11 properties). August–December 2026 (dotted) are not final figures but current estimates based on the present sales position.

Source: Compiled by the HotelBank Editorial Team from MetroEngines Research.

Business hotels have a “flat shape” — the constraint of a 1.28x seasonal amplitude

Overlaying the years for business hotels, the lines lie flat compared with city hotels. Taking the range across final months for January–July 2026, the low is ¥6,837 in January and the high is ¥8,750 in May, so highest ÷ lowest is just 1.28x. Over the same period, city hotels ran from a low of ¥9,176 in January to a high of ¥15,680 in May — 1.71x. Same prefecture, same seven months, and the swing in price differs by more than 1.3x.

This difference in shape feeds directly into the year-on-year gap. Because city hotels have room to lift substantially in peak months, they can overtake the prior year decisively in any month when demand returns. Business hotels face a ceiling even in peak months, so there is little material to push year-on-year higher. In the first half, business hotels recovered ground in only two months, April and May, and by narrow margins of +3.9% and +2.8%.

Estimated settled ADR for Ishikawa business hotels (N=92–118 properties). August–December 2026 (dotted) are not final figures but current estimates based on the present sales position.

Source: Compiled by the HotelBank Editorial Team from MetroEngines Research.

Taking the level ratio between the two (city ÷ business) across final months makes the widening tier gap even more direct. June 2025 was 1.27x and June 2026 was 1.44x; July 2025 was 1.43x and July 2026 was 1.64x. In yen, the June gap widened from about ¥2,000 to about ¥3,100, and the July gap from about ¥3,300 to about ¥4,700. Accommodation pricing in the prefecture is shifting from a “two-storey” structure to a “two-storey structure with a higher upper floor” — that is the reality of the first half. This divergence between categories is not unique to Ishikawa; it also appears as a nationwide pattern in our breakdown of year-on-year change by prefecture and category, Japan Hotel ADR Polarization 2026.

Demand arrives differently — booking curves at 45 days out, 30 days out and the latest observation

To see where the price gap comes from, look at recent booking progress. Plotting booking curves from 45 days before the stay date through to the latest observation, city and business hotels clearly differ in “when inventory fills”.

For arrivals on Wednesday 5 August 2026, for example, estimated OCC (based on OTA-listed inventory) for city hotels ran 69.6% at 45 days out → 70.6% at 30 days out → 79.9% at the latest observation (2 days out; as of 4 August 2026, N=10 properties), a build over the period of +10.3 points. Business hotels on the same date ran 60.8% at 45 days out → 64.2% at 30 days out → 86.0% at the latest observation (2 days out; as of 4 August 2026, 83–101 properties observed), a build of +25.2 points. City hotels have already secured close to 70% at 45 days out, and the late-stage gain is small. Business hotels sit at 60% at 45 days out and then stack up sharply in the final two weeks. This pattern — inventory filling differently by category — is not specific to Ishikawa: the same directional difference has been observed in the Tokyo market, where business hotels gained +25.7 points against +14.5 points for city hotels over the same window.

Estimated OCC (based on OTA-listed inventory) for arrivals on Wednesday 5 August 2026. The x-axis is days remaining until the stay date. City N=10 properties; business 83–101 properties observed.

Source: Compiled by the HotelBank Editorial Team from MetroEngines Research.

This asymmetry is not confined to a single date. Taking fixed points (45 days out, 30 days out, latest observation) across several August and September arrival dates, the business build exceeded the city build on every date.

Estimated OCC (based on OTA-listed inventory), from observations running from 45 days before the stay date through to the most recent reading. “Latest obs.” is the most recent observation point for each stay date (Aug 5: 2 days out; Aug 8: 5–6 days out; Aug 14: 11 days out; Aug 17: 14 days out; Aug 29: 26 days out; Sep 1: 29 days out). “Build” is the point difference from 45 days out to the latest observation. Properties observed: city 9–10, business 82–101. Data as of 4 August 2026.

Source: Compiled by the HotelBank Editorial Team from MetroEngines Research.

The day-of-week shape also differs between the two. Averaging the daily actuals for July 2026 — a final month — by day of week, the monthly averages are close: 82.8% for city (July 2026, final month / N=10 properties) and 82.0% for business (July 2026, final month / N=98–103 properties). But the peaks and troughs within that are larger for business hotels. Business runs 90.3% on Saturday against 75.1% on Monday and 78.1% on Sunday, a maximum spread of 15.2 points. City runs 87.8% on Saturday and 79.3% on Monday, a maximum spread of just 8.5 points. Business hotels lean more heavily on weekends and fall further at the start of the week. Which day the trough lands on varies by region — in some markets, such as Aichi and Hiroshima, the trough falls on Sunday rather than Monday.

Estimated OCC (based on OTA-listed inventory) for Ishikawa, calculated as a simple average by day of week of the daily actuals for July 2026 (a final month). Includes the three-day weekend of 18–20 July (Sat–Mon) containing the Monday 20 July public holiday.

Source: Compiled by the HotelBank Editorial Team from MetroEngines Research.

The composition of that long weekend is telling in itself. Saturday 18 July 2026 was almost full at 99.4% for city and 99.4% for business, and Sunday the 19th held high at 90.2% for city and 94.7% for business. Yet the final night of the break, Monday the 20th (a public holiday), fell to 75.3% for city and 69.4% for business. It was a month in which Ishikawa’s pattern showed up clearly: inventory absorption on “the nights up to the middle of a long weekend” and on “the final night” are entirely different things.

Current estimates for the second half (August–December 2026)

For future months, what is available is not final figures but current estimates based on the present sales position. City hotels stand at ¥12,887 for August and ¥15,288 for October; business hotels at ¥9,689 for August, ¥12,146 for October and ¥12,212 for November. All of these can move with future sales activity, and because the basis of calculation differs from that of final months, any straight comparison against prior-year final figures should wait until month-end finalisation. Here we limit ourselves to confirming the shape — which months of the second half are expected to peak.

Current estimates of estimated settled ADR (tax-exclusive equivalent). These are not final figures and may move with future sales activity. A straight comparison against final figures requires waiting for month-end finalisation. Data as of 4 August 2026.

Source: Compiled by the HotelBank Editorial Team from MetroEngines Research.

These current estimates cannot be read straight across as year-on-year change, because the basis differs (final figures are built on past actuals; current estimates on inventory still on sale). Even so, setting out “what level the final figures reached over the past two years” gives a useful bearing when redrawing second-half ranges.

Ishikawa, estimated settled ADR (tax-exclusive equivalent). The 2024 and 2025 columns are final figures (based on past actuals); the 2026 column is a current estimate (based on inventory still on sale), so the basis of calculation differs. The difference between the right-hand column and the two left-hand columns therefore cannot be interpreted as a year-on-year change. City N=10–11 properties; business N=92–118 properties. Data as of 4 August 2026.

Source: Compiled by the HotelBank Editorial Team from MetroEngines Research.

What the comparison shows is that the size of the basis difference varies by category. The city current estimates (¥12,887 for August, ¥15,288 for October) sit in the same band as the 2025 final figures (¥13,519 and ¥14,126). The business current estimates (¥12,146 for October, ¥12,212 for November), by contrast, run roughly 30% above the 2025 final figures (¥8,891 and ¥9,465). The safer reading is that business hotels carry considerably more room for the level to come down as the month finalises. Setting second-half range ceilings directly to the current estimates risks aiming too high, particularly in the business category.

For revenue managers running city and business hotels in Ishikawa — implications and an action plan

(b) Action plan

All figures in this table are the Ishikawa aggregates presented in this article. How well they apply will vary with each property’s circumstances.

Summary — three yardsticks to take away from Ishikawa’s first half

Yardstick 1: read the tier gap as a ratio. Ishikawa’s city ÷ business ratio went from 1.43x in July 2025 to 1.64x in July 2026. In yen, the gap widened from about ¥3,300 to about ¥4,700. Check where your property sits within its band using the ratio, not the prefecture average.

Yardstick 2: read seasonal amplitude as highest ÷ lowest. On final figures for January–July 2026, city is 1.71x and business 1.28x. If that spread is wider than your own annual range, there may still be headroom in your peak-month ceiling.

Yardstick 3: the level at 45 days out means different things by category. For 5 August (Wed) arrivals, city gained +10.3 points from 45 days out to the latest observation, business +25.2 points. The same “70% at 45 days out” is close to the final shape for a city hotel but still has room to move for a business hotel. Comparing your progress with another property’s without knowing your own category’s pattern gives you nothing to decide on.

Ishikawa’s first half was seven months in which city hotels grew and business hotels could not quite follow. But that difference is not a matter of one category being superior to the other; it arises from differences in how bookings arrive and in the shape of the price range. If you are redrawing your range for the second half, start by confirming which pattern your own property is running on, at the three points of 45 days out, 30 days out and the latest observation.

About the data

Source: Compiled by the HotelBank Editorial Team from MetroEngines Research.

■ Data source

MetroEngines Research’s proprietary database, which collects and aggregates publicly listed inventory and listed prices on OTAs on a daily basis. For the two Ishikawa categories of city hotels and business hotels, monthly series by prefecture and category (estimated settled ADR, tax-exclusive equivalent) were retrieved for the range from January 2024 to December 2026, and rooms remaining on listings and total rooms for each stay date were observed daily from 45 days before the stay date through to the most recent reading. The 62 monthly values for 2024–2026 presented in this article were confirmed to match the aggregates at the time of publication in every case.

■ Calculation assumptions

Year-on-year change is calculated only between final figures (on a past-actuals basis); no straight comparison is made between current estimates and prior-year final figures. The January–July weighted average is weighted by the number of days in each month. Seasonal amplitude is the highest value divided by the lowest among final months. “Gap” is the city year-on-year change minus the business year-on-year change, in points. Estimated OCC is calculated as “100 − 100 × rooms remaining on listings ÷ total rooms”, and the day-of-week aggregation is a simple average of each day falling on that weekday. The 2026 column of Table 5 is a current estimate (based on inventory still on sale) and uses a different basis of calculation from the final figures in the 2024 and 2025 columns.

■ Limitations and caveats

City hotel coverage is thin at N=10 properties, so a rate change at a single property can move the prefecture average. The level needs to be read as a range rather than a point. Estimated OCC is an estimate based on how inventory sold on OTAs is absorbed, and is defined differently from actual room occupancy, which includes direct and corporate-contract bookings (it reads higher). Year-on-year change in occupancy is not covered in this article, as the accumulated history is under two years. Because the days remaining at “latest observation” differ by stay date, the build figures are not strictly equivalent across stay dates. No reconciliation has been made against external official statistics such as the Overnight Travel Statistics Survey.