A booking curve is not something to stare at every day. It is something you cut at a fixed number of days remaining and compare. Measured across Okayama business hotels (N=88 properties, 10,457 rooms), estimated OCC for Saturday 8 August 2026 moved from 76.2% at 45 days out to 77.8% at 30 days out — a gain of only +1.6pt — and then stacked up +11.8pt in one push to reach 89.6% at the latest observation (8 days remaining). In the same segment, Saturday 22 August 2026 was already at 85.3% at 45 days out and 91.3% at 30 days out, running high from early on, with only +2.1pt added by the latest observation (22 days remaining). Same hotel type, same day of the week — and completely different curve shapes. This article uses Okayama’s measured figures as a teaching set to lay out how to run a booking curve on just three fixed points: the level at 45 days out, the slope from 45 to 30 days out, and the room left to gain in the final stretch.

Scope: Okayama business hotels N=88 properties (10,457 rooms), city hotels N=16 properties (1,924 rooms), ryokan N=45-46 properties (882-903 rooms). The price metric in this article is estimated settled ADR (a settled price level estimated from OTA and other sales data, tax-exclusive equivalent); occupancy is an estimate based on OTA-listed inventory. Full definitions appear at the end of the article. Data retrieved 2 August 2026 (the latest observation cross-section for booking curves is 31 July 2026).

Same Saturday, different shape — three fixed points for Okayama business hotels

First, let us confirm with measured data what it means to compress a booking curve into three points. The chart below overlays estimated OCC (based on OTA-listed inventory) for Okayama business hotels across three Saturdays — 8, 22 and 29 August 2026 — with days remaining until the stay date on the horizontal axis. Observations are limited to the window from 45 days out to the most recent reading.

Source: Compiled by the HotelBank Editorial Team from MetroEngines Research

All three lines rise to the right; on that count they are the same. What matters operationally is where the slope concentrates. For 8 August, the 15 days from 45 days out to 30 days out produced only +1.6pt, and everything after that added +11.8pt. Almost the entire gain is packed into the final two weeks — a “late-surge” pattern. By contrast, 22 August was already high at 85.3% at 45 days out, added +6.0pt from 45 to 30 days out, and then only +2.1pt: it finished stacking early — a “front-loaded” pattern. 29 August ran from 77.7% at 45 days out to 82.0% at 30 days out (+4.3pt), with the latest observation (29 days remaining) at 82.2%, leaving it mid-course with the evidence not yet complete.

Okayama, business hotels, estimated OCC (based on OTA-listed inventory). The base is 87-88 properties in total. Observed properties number 81-85 at the 45-days-out cross-section, 82-87 at the 30-days-out cross-section and 52-83 at the latest observation. For 5 and 12 September the stay date is more than 30 days away, so the 30-days-out cross-section has not yet been reached.

Source: Compiled by the HotelBank Editorial Team from MetroEngines Research

The operational implication of this table is clear. Looking at the level at 45 days out alone does not determine where a date will finally land. 8 August came from an unremarkable 76.2% at 45 days out all the way to 89.6%, while Wednesday 26 August went from 65.1% at 45 days out to 70.5% at the latest observation — only +1.9pt of gain with 26 days still remaining. Two dates that both sat “in the 70s or below at 45 days out” went on to diverge. That is exactly why, on top of the level at 45 days out, you need to watch the slope from 45 to 30 days out. Against +6.0pt for 22 August and +4.3pt for 29 August, 8 August managed +1.6pt and 19 August +2.5pt. That 15-day slope is the first signal separating a date that “settles early” from one that “keeps moving until the last minute.” The shape in which 15 August shows the lowest 45-days-out level among August Saturdays is not unique to Okayama, either: the same three-point cut confirms it in Hokkaido city hotels, where only 15 August trails among August Saturdays.

One caveat on 22 August: at the latest observation (22 days remaining), the share of properties for which no listed inventory can be confirmed is estimated at 50.6% (52 properties observed), and 49.4% at the 30-days-out cross-section (82 properties observed). Because the number of observed properties swings between 52 and 87 depending on the cross-section, this share has to be read together with the thickness of that cross-section. Never resting a conclusion on a figure from a thin cross-section alone is a precondition of running the three-point frame.

The 45-days-out level spans 30pt by hotel type — the same date reads differently

When you share the three-point frame with other properties or other areas, the question that causes the most misunderstanding is “what percentage at 45 days out counts as healthy?” Okayama’s measured figures show that for the same stay date, the level at 45 days out differs completely by hotel type. At 45 days out for Saturday 12 September 2026, estimated OCC was 83.6% for city hotels (16 properties observed / base of 16) against 53.1% for ryokan (45 observed / base of 45) — a gap of 30.5pt. Business hotels sat between them at 73.4% (84 observed / base of 88).

Source: Compiled by the HotelBank Editorial Team from MetroEngines Research

The chart above overlays the curves of the three hotel types using Saturday 15 August 2026 as the example. City hotels went from 76.2% at 45 days out to 78.8% at the latest observation (15 days remaining), a gain of +2.6pt — their shape was essentially set by 45 days out. Business hotels ran from 68.1% to 78.0% (+9.9pt) and ryokan from 55.0% to 66.6% (+11.6pt), stacking up toward the back half. Here sits an obvious fact that is nonetheless easy to miss in day-to-day operations: the meaning of “the level at 45 days out” differs by hotel type. For city hotels, 76.2% at 45 days out means “close to the expected landing,” whereas for ryokan, 55.0% at 45 days out is nothing more than a waypoint on the assumption that more will be stacked from here. The underlying question — whether hotel types that run high at 45 days out have less room left in the final stretch — is tested type by type in Tochigi booking curves by hotel type for August-September 2026.

Okayama, estimated OCC at the 45-days-out cross-section (based on OTA-listed inventory). Observed properties: business hotels 81-85 (base of 87-88), city hotels 14-16 (base of 16), ryokan 38-45 (base of 45-46).

Source: Compiled by the HotelBank Editorial Team from MetroEngines Research

Build thresholds from your own track record — landing level, day-of-week shape and monthly ADR

So how do you set the threshold for “at what percentage 45 days out do I act?” Applying the market average directly to your own property does not work. The procedure has three stages: (1) establish the landing level for your property and hotel type from completed-month results; (2) break out the shape by day of the week; and (3) cross-check that level against the monthly estimated settled ADR. Okayama’s completed-month results let us walk through it.

Okayama: within-month daily averages and day-of-week averages of estimated OCC (based on OTA-listed inventory), derived from daily results for completed months. Each month is a separate measured figure.

Source: Compiled by the HotelBank Editorial Team from MetroEngines Research

Use this table as the “yardstick” for the three-point frame. The July 2026 daily average for business hotels was 88.5%, with Thursday highest at 92.3% and Sunday lowest at 81.2%. Once you know the landing level and the day-of-week shape, you can translate the level at 45 days out into “how many more points are needed to land.” Saturday 8 August, cited at the top, was at 76.2% at 45 days out but reached 89.6% at the latest observation (8 days remaining), putting it in a position consistent with the landing-level benchmark. Wednesday 26 August, by contrast, stood at 70.5% at the latest observation (26 days remaining) — still some distance from the Wednesday landing benchmark. Two dates that are both “low at 45 days out” call for different moves once their distance to the landing level differs. As for reading day-of-week shape itself, the Sunday trough and the Thursday peak are the two anchors to fix first.

Here is the practical way to build the threshold. From your property’s past 12 months of results, derive landing occupancy by hotel type, day of the week and month. Next, record where your property stood at 45 days out and 30 days out relative to that landing, and build the distribution of the gap (the points remaining). The median of that gap is “the normal room left to gain for that day of the week in that season,” and the moment your measured 45-days-out figure departs from it is the moment to intervene. The market’s 45-days-out level is best used not as an absolute target for your own property, but as a reference line for checking whether external conditions are running as usual.

Let us run this procedure once against Okayama’s measured business-hotel data. Taking the same day-of-week average from the most recent completed month (July 2026) as a proxy for the landing level, the table below lines up how far each stay date’s 45-days-out figure sat from it, and how far that distance had closed by the latest observation. Translated from an absolute 45-days-out value into “distance to landing,” two dates in the same 70s range call for different moves — as the table makes visible.

Okayama, business hotels, estimated OCC (based on OTA-listed inventory). The landing benchmark is a proxy built from July 2026 day-of-week averages (Sat 92.2%, Fri 86.3%, Wed 91.0%, N=83-88 properties) and does not incorporate seasonality specific to August. Distance = landing benchmark minus estimated OCC at the given cross-section. A negative value indicates the benchmark has already been exceeded. When running this at your own property, replace this column with your own landing results for the same month and the same day of the week over the past 12 months.

Source: Compiled by the HotelBank Editorial Team from MetroEngines Research

The third stage is price. An occupancy yardstick alone cannot tell you whether to fill the remaining room or defend rate. For Okayama business hotels, taking the most recent month where confirmed figures can be compared like for like, estimated settled ADR was ¥6,700 in June 2026 (N=94 properties) against ¥6,900 in June 2025 (N=97 properties), or -3.9% year on year. May 2026 was ¥7,600 (N=97 properties), +2.3% year on year. By hotel type, confirmed June 2026 figures were ¥9,700 for city hotels (N=17 properties, -16.6% YoY) and ¥10,700 for ryokan (N=58 properties, +7.1% YoY).

Source: Compiled by the HotelBank Editorial Team from MetroEngines Research

Solid lines in the chart are confirmed figures; dotted lines are estimates based on current sales conditions. For business hotels, August 2026 is currently estimated at ¥10,400 (N=93 properties) and September at ¥10,100 (N=90 properties). Because these rest on a different basis from confirmed figures, a straight comparison with confirmed values has to wait for month-end finalisation. Even so, the outline — August sitting high within the year as a seasonal shape, easing somewhat into September — is a useful reference when drawing up a rate-revision calendar. For city hotels the current estimates are ¥16,100 for August 2026 (N=17 properties) and ¥16,200 for September (N=17 properties); for ryokan, ¥14,200 for August (N=54 properties) and ¥14,500 for September (N=55 properties). All of these can move, and should be assessed once the month closes.

For revenue managers running hotels and ryokan in Okayama — implications and an action plan

Source: Compiled by the HotelBank Editorial Team from MetroEngines Research

Summary — three yardsticks to take away

Yardstick 1: read the 45-days-out level against a type-specific reference line. For Saturday 12 September 2026 in Okayama, city hotels stood at 83.6%, business hotels at 73.4% and ryokan at 53.1% — a 30.5pt spread. A comparison that does not use your own type’s reference line does not hold up in the first place.

Yardstick 2: the 15-day slope from 45 to 30 days out determines the pattern. The +6.0pt on Saturday 22 August is front-loaded; the +1.6pt on Saturday 8 August is late-surge. The slope is a proxy for “how much time is left to move price,” and it connects to action more directly than the level at 45 days out.

Yardstick 3: measure the room left in the final stretch as distance from the landing level. 8 August went from 77.8% at 30 days out to 89.6% at the latest observation (+11.8pt); 22 August went from 91.3% to 93.4% (+2.1pt). Even within business hotels, the size of the remaining room differs by an order of magnitude from date to date. Hold the distance to your own landing level in advance (on the market side, Okayama business hotels averaged 88.5% daily in July 2026) and you will not be whipsawed by late movements.

The value of a booking curve lies not in the elegance of the line but in being able to cut it at a fixed number of days remaining and compare it with last time. Compress it into three points — 45 days out, 30 days out and the final stretch — and you can hold the same discussion every week, in the same format and the same vocabulary. Start by re-laying your own past 12 months along those three points.

About the data

Source: Compiled by the HotelBank Editorial Team from MetroEngines Research