Home > Area & Property Analysis > Solo 44.7% vs Family 44.4%: Rebuilding Rate Design by Guest Mix

Solo 44.7% vs Family 44.4%: Rebuilding Rate Design by Guest Mix

Posted: 2026.09.03

Area & Property Analysis

Aggregating the companion attributes recorded in hotel reviews nationwide, 2025 came in at solo 44.7% and family 44.4% — a gap of just 0.3 points (N=407,873 reviews). In 2021 the split was solo 46.5% and family 38.6%, a 7.9-point spread, so in four years the picture of “solo travellers first, families second” has all but disappeared. Over the same period, the share of reviews posted from outside Japan fell from 44.7% in 2024 (N=4,896,647 reviews) to 38.4% in 2025 (N=4,256,486 reviews). Meanwhile, comparing settled-ADR estimates month-on-month against the same month a year earlier, July 2026 shows a number of areas where city hotels fell further than business hotels. This article turns these three datasets into a working sequence: inventory allocation by guest segment → rate plan design → mix verification.

Scope: nationwide review-attribute aggregation (companion and purpose N=407,873–407,874 reviews / domestic-overseas composition N=4,256,486 reviews, all 2025), plus business hotels and city hotels across six major prefectures, N=30–913 properties. The price metric in this article is settled ADR (an estimated transaction price level derived from OTA and other sales data, tax-excluded equivalent). Definitions appear at the end of the article. Data as of August 31, 2026.

Key Takeaways
  • — Solo 44.7%, family 44.4% — a 0.3-point gap. Down from a 7.9-point spread in 2021, the two have drawn level in four years (review-poster attributes, N=407,873 reviews).
  • — “Solo” contains two different things. Among business-purpose stays, 89.8% travel alone (N=83,714 reviews), while leisure accounts for 74.7% of all posts. The single-room inventory you need on a weekday is not the inventory you need on a weekend.
  • — Year-on-year moves in settled months split by property type. In July 2026, settled ADR in Tokyo was −5.3% for business hotels and −17.1% for city hotels, while Fukuoka was positive for both (+3.6% / +3.1%).
  • — The overseas share of reviews is not a demand indicator. It fell from 44.7% to 38.4% in 2025 — the very year inbound arrivals hit a record high. Do not use it as a pricing trigger.

The solo-family gap narrowed from 7.9 points to 0.3 points

First, a clear caveat. What is measured here is not the composition of actual guests, but the attribute composition of the people who posted reviews. Not every guest writes a review, and those who do skew by age band and trip purpose. The figures below are therefore not proof of causation — “this type of guest increased in the market” — but a reference frame for auditing your own guest mix. Even so, lining up five years computed the same way makes the direction of travel readable.

The companion composition moved as follows. Solo fell 1.8 points from 46.5% in 2021 to 44.7% in 2025, while family rose 5.8 points from 38.6% to 44.4%. Couples and married pairs shrank from 6.8% to 3.9%, friends from 4.9% to 4.5%, and colleagues from 2.4% to 2.0%. In other words, the change over these five years looks less like “solo declined” than “family grew until it drew level with solo.” How to rebuild rate and inventory design around that shift in companion mix is set out in detail in Solo vs Family Stays Just 0.3pt Apart — Rethinking Rate and Inventory.

Source: MetroEngines Research; compiled by the HotelBank Editorial Team

Table 1: Companion composition of hotel review posters, 2021–2025 (nationwide)
YearSoloFamilyCouplesFriendsColleaguesSolo − FamilyN (reviews)
202146.5%38.6%6.8%4.9%2.4%+7.9pt488,360
202246.3%40.7%5.2%5.0%2.1%+5.6pt739,842
202343.8%43.6%4.7%5.3%2.0%+0.2pt633,015
202445.6%42.5%4.2%4.9%2.2%+3.1pt606,094
202544.7%44.4%3.9%4.5%2.0%+0.3pt407,873

Source: MetroEngines Research; compiled by the HotelBank Editorial Team

The breakdown by trip purpose is worth holding onto as well. In 2025 the split was leisure 74.7%, business 20.5%, other 4.8% (N=407,874 reviews) — 5.0 points more weighted toward leisure than 2021, when leisure was 69.7% and business 22.5%. And within business trips specifically, companions are solo 89.8% and colleagues 7.7% (N=83,714 reviews), which is very nearly pure single occupancy. The 44.7% solo figure is the sum of “business solo” and “leisure solo,” and those two differ in the days of week they want and in booking lead time. Treat the solo share as a single block and you will misread how much single-room inventory you need on weekdays versus weekends.

Age skew is also worth checking. Among 2025 leisure posters, 31.4% were in their 50s, 26.2% in their 60s and 20.7% in their 40s (N=124,011 reviews); among business posters, 42.7% were in their 50s, 25.3% in their 60s and 20.6% in their 40s (N=31,355 reviews). Both lean middle-aged and older. Because the population that writes reviews at all skews that way, this is not the age composition of the market itself. It does, however, tell you that the cohort shaping your property’s public reputation through reviews is concentrated here — useful when prioritising how you handle review responses.

The domestic-overseas split of reviews, and where it diverges from official statistics

The domestic-overseas split needs to be read even more carefully. On a review-poster nationality basis, the overseas share climbed sharply — 15.5% in 2021 (N=3,751,759 reviews), 14.9% in 2022 (N=4,802,050 reviews), 41.4% in 2023 (N=4,704,930 reviews), 44.7% in 2024 (N=4,896,647 reviews) — and then fell 6.3 points to 38.4% in 2025 (N=4,256,486 reviews).

Official statistics point the other way. The Japan National Tourism Organization (JNTO) estimates annual inbound arrivals for 2025 at 42,683,600, up 15.8% year on year — more than 5.8 million above the previous record of 36,870,148 set in 2024, and a new annual high. Actual visitor numbers rising while the overseas share of reviews falls is direct evidence that this ratio reflects the composition of review-writing behaviour, not the nationality composition of guests. Posting rates shift easily with channel mix, language and stay format.

Source: MetroEngines Research; compiled by the HotelBank Editorial Team

The practical implication is simple. Reading a falling overseas review share as “inbound demand has weakened,” and easing your pricing accordingly, is dangerous. What the metric is good for is observing which languages and which cohorts your own review base is accumulating from. A property whose overseas share sits far from the national average is worth checking for listing copy, photography and review-response language design that has drifted toward one particular cohort. Judgements about demand itself should keep official statistics, your own results, and sales-data-derived metrics separate. For how inbound guests actually distribute across property types, City Hotels: 45.9% of Guest Nights Are Foreign — Ryokan +3.7% breaks that down separately on an official-statistics basis.

Year-on-year settled ADR in closed months diverges by property type

So what about the price side? Taking the most recent settled-month pairing available — July 2026 against July 2025 — we lined up settled ADR for business hotels and city hotels across six major prefectures. The year-on-year moves are as follows.

Source: MetroEngines Research; compiled by the HotelBank Editorial Team

Table 2: Settled ADR year-on-year by property type, six major prefectures (July 2026 vs July 2025, settled figures on both sides)
Area / property typeJuly 2025
settled ADR
July 2026
settled ADR
YoYN (properties)
Tokyo — business hotels¥12,990¥12,306−5.3%913
Tokyo — city hotels¥27,055¥22,434−17.1%107
Osaka — business hotels¥12,380¥8,429−31.9%497
Osaka — city hotels¥16,798¥12,582−25.1%97
Kyoto — business hotels¥11,752¥10,230−13.0%325
Kyoto — city hotels¥20,926¥17,420−16.8%58
Aichi — business hotels¥7,584¥7,491−1.2%323
Aichi — city hotels¥11,684¥11,466−1.9%30
Fukuoka — business hotels¥10,023¥10,382+3.6%340
Fukuoka — city hotels¥16,106¥16,608+3.1%40
Hokkaido — business hotels¥10,712¥11,154+4.1%432
Hokkaido — city hotels¥18,891¥17,819−5.7%75

Source: MetroEngines Research; compiled by the HotelBank Editorial Team (all comparisons are settled figures against settled figures)

Three things stand out. First, in the three major metropolitan regions there are areas where city hotels posted a deeper year-on-year decline than business hotels: Tokyo at −5.3% for business hotels against −17.1% for city hotels, and Kyoto at −13.0% against −16.8%. Second, Osaka fell sharply regardless of type (business hotels −31.9%, city hotels −25.1%), which may include a reaction to an unusually high base in the same month a year earlier. Third, Fukuoka was positive for both types (+3.6% / +3.1%) and Hokkaido’s business hotels rose +4.1%, so conditions are not uniform nationwide. How that split within Hokkaido (city −5.7%, business +4.1%) formed across the first half of the year is broken out month by month in Hokkaido ADR H1 2026: City −5.7%, Business +4.1%, April the Turn.

A guest mix in which solo and family are near-identical is not unrelated to this divergence by property type. A property type built around single-room inventory and one built around rooms designed for two or more people will see transaction prices move differently in the same demand environment. That said, the data here cannot establish causation between the two. What we take from it is an operating rule: look at the price movements of the property type your own hotel belongs to, not an all-types average.

Note that months from August 2026 onward are not yet settled, so they should be treated as estimates based on current sales conditions. For reference, business hotels in Tokyo currently estimate at roughly ¥14,200 for September 2026 (N=902 properties) and roughly ¥18,200 for October (N=884 properties); both can move with sales conditions. Comparing these directly against settled figures requires waiting for month-end settlement.

For revenue managers running hotels in a market where solo and family are level — implications and an action plan

(1) Do not treat “solo 44.7%” as a single block. Among business-purpose stays, 89.8% travel alone (N=83,714 reviews) — effectively all single occupancy — but leisure solo guests differ in which days they turn up and when they book. Re-aggregate your own solo stays into two streams, “weekday business” and “weekend leisure,” and the days on which you should hold more single inventory separate from the days on which you would do better converting to double occupancy. Decide inventory allocation from the aggregate solo share alone and that split stays invisible.

(2) A 44.4% family share is a reason to revisit the granularity of your rate design. Once the family share is this high, a design that adds per-person supplements on top of a double-occupancy base tends to leave three- and four-person prices floating above the market. Line up your own results by party size, and if bookings for three or more are thinner than expected, the issue may be the supplement increments — or the visibility of rooms that can take three or more people at all. This is not an argument for discounting; it is a check on whether you are capturing demand you should already be able to serve.

(3) Make property-type price movements your yardstick. In the settled-month comparison here, Tokyo city hotels were −17.1% and business hotels −5.3%. Comparing your own results for that month against an all-types average tells you little. Matching them against the same property type in the same area is the first input into whether to move price at all.

(4) Do not use the overseas share of reviews as a demand indicator. In 2025, the year inbound arrivals set a record, the overseas share of reviews fell from 44.7% to 38.4%. These two things measure different phenomena. The overseas share is usable for checking which cohorts your review base is accumulating from, but draw the internal line clearly: it is not a pricing trigger.

Placed on a timeline, that comes out as follows. None of these guarantees an immediate effect; they are a way of ordering the sequence of judgement.

Table 3: Action plan by time horizon for translating guest mix into pricing
HorizonActionTrigger (tied to figures in this article)Purpose
Today to this weekTake stock of the last 12 months of results across four buckets: solo (weekday), solo (weekend), two people, three or moreWhen your own family and group share diverges sharply from the 44.4% family figure in the review attributesStart the inventory-allocation discussion on actual party-size data
Today to this weekCheck the settled-month year-on-year figure for your own property type and area, separately from the all-types averageWhen your results for that month move in the opposite direction to the same property type (e.g. Tokyo city hotels −17.1% / business hotels −5.3%)Separate market factors from property-specific factors
Within two weeksRevisit per-person supplement increments and verify that prices for three or more people are not floating above the market range for your property typeWhen bookings for three or more are clearly thin relative to the level of the family shareReduce leakage of demand you could serve
Within two weeksTidy the number of rate plans so that there is at least one clear entry point each for solo, for two people, and for three or moreWhen the plan listing still makes solo and multi-occupancy options hard to tell apartAlign how each segment sees you and secure an entry point for comparison
Toward next monthCompare current ADR estimates (e.g. Tokyo business hotels, roughly ¥14,200 for September 2026 and ¥18,200 for October) against your own settings for those monthsWhen your settings for that month remain persistently below the estimated level for the same property type and areaCreate room to sanity-check your settings without waiting for settlement
Toward next monthBuild guest-mix verification into the monthly routine, recording share and unit price side by side for each party-size bucketWhen checking, in the month after an inventory reallocation or plan change, whether bucket shares have moved in the intended directionMake the link between action and result traceable after the fact

Source: MetroEngines Research; compiled by the HotelBank Editorial Team

In summary — three yardsticks

The first yardstick: solo and family are near-identical. In the 2025 review attributes, solo was 44.7% and family 44.4%, a gap of 0.3 points (N=407,873 reviews). Inventory allocation or plan structures premised on one or the other tend to fall on the side of leaking the other’s demand. Start by laying your own party-size results alongside this reference frame.

The second yardstick: “solo” contains two things. Among business-purpose stays, companions are solo 89.8% (N=83,714 reviews), while leisure accounts for 74.7% of all posts (N=407,874 reviews). Unless you count solo guests separately as weekday business and weekend leisure, you will misread which days call for heavier single inventory.

The third yardstick: compare within property type. Settled-month year-on-year figures do not line up in a single direction by type or by area — Tokyo at business hotels −5.3% and city hotels −17.1%, Fukuoka at +3.6% and +3.1%. Using an all-types or national average as your benchmark is too coarse to act on. And it is worth agreeing internally on the line that the overseas share of reviews is not a proxy for demand.

About the data

■ Data sources

• Guest-segment data is an aggregation of publicly posted reviews from OTAs and similar channels. Purpose, companion and age band come from posts on domestic channels where attribute completion rates are high; the domestic-overseas split is based on the posting country of reviews across all channels. These are the attributes of review posters, not a traveller survey, and they differ from the composition of actual guests. Age bands skew toward the population that writes reviews. Mode of transport, spend per traveller per trip, average length of stay and similar measures are not included in this aggregation.
• Breakdown of N: companion and purpose, 2025 aggregation N=407,873–407,874 reviews; domestic-overseas composition, 2025 aggregation N=4,256,486 reviews, 2024 N=4,896,647 reviews. For settled ADR: Tokyo business hotels N=913 properties / city hotels N=107 properties; Osaka N=497 / 97; Kyoto N=325 / 58; Aichi N=323 / 30; Fukuoka N=340 / 40; Hokkaido N=432 / 75 (all as of July 2026).

■ Estimation assumptions

• Definition of settled ADR: a transaction price level (tax-excluded equivalent) estimated from OTA and other sales data (lowest-plan level × property-type coefficient, ensembled across multiple channels). Past months are settled figures; the current and future months are estimates based on current sales conditions. Median error against published operating results is 6.6%.

■ Limitations and caveats

• The 2026 domestic-overseas composition is still mid-aggregation and structurally biased, so it is not used in this article (only 2021–2025 are shown).
• This article does not use occupancy (estimated OCC) or booking-curve data.
• Data as of August 31, 2026. Sales conditions and inventory change daily, so the figures here are a snapshot at the time of extraction.

References and sources

・Japan National Tourism Organization (JNTO), “Foreign Visitors to Japan (December 2025 estimate)” https://www.jnto.go.jp/news/press/20260121_monthly.html
• Japan National Tourism Organization (JNTO), “Foreign Visitors to Japan in 2025 (total)” https://www.jnto.go.jp/statistics/data/_files/20260121_1615-4.pdf
• Japan Tourism Agency, “Overnight Travel Statistics Survey” https://www.mlit.go.jp/kankocho/tokei_hakusyo/shukuhakutokei.html

Frequently asked questions

Q. Is the “solo 44.7% / family 44.4%” in this article the composition of actual guests?

A. No. It is the attribute composition of people who posted hotel reviews — neither a traveller survey nor the composition of actual guests (2025, N=407,873 reviews). Not every guest writes a review, and those who do skew by age band and trip purpose. Use it as a reference frame when auditing your own guest mix, not as proof of the demand composition of the market as a whole.

Q. Can I take 44.7% solo straight across as a target for my single-room inventory ratio?

A. Not recommended. “Solo” is the sum of business-purpose solo and leisure-purpose solo. Within business trips, companions are solo 89.8% (N=83,714 reviews) — very nearly all single occupancy — but leisure solo guests differ in the days they want and in booking lead time. The practical route is to re-aggregate your own solo stays into weekday business and weekend leisure, then set inventory allocation by day of week.

Q. What is settled ADR, and how does it differ from the listed price?

A. Settled ADR is a transaction price level estimated from OTA and other sales data (lowest-plan level × property-type coefficient, ensembled across multiple channels), on a tax-excluded equivalent basis. It is not the listed selling price itself. Median error against published operating results is 6.6%. Treat past months as settled figures, and the current and future months as estimates based on current sales conditions.

Q. Does a falling overseas share of reviews mean inbound demand has weakened?

A. No. The overseas share of review posters fell from 44.7% in 2024 to 38.4% in 2025, but annual inbound arrivals in that same 2025 reached a record 42,683,600, up 15.8% year on year (JNTO estimate). The ratio reflects the composition of review-writing behaviour, not the nationality composition of guests, and posting rates shift easily with channel mix, language and stay format. Do not use it as a pricing trigger.

Q. Can I compare the September and October 2026 estimates in this article directly against settled figures?

A. Avoid a straight comparison. Every year-on-year figure in this article compares settled months against settled months (July 2026 vs July 2025). By contrast, Tokyo business hotels at roughly ¥14,200 for September 2026 (N=902 properties) and ¥18,200 for October (N=884 properties) are estimates based on current sales conditions and can move with those conditions. Comparison against settled figures requires waiting for month-end settlement.

Q. The family share has risen to 44.4% — should I add more rooms that sleep three or more?

A. Before deciding on a change to your room composition, check your own results by party size. The data in this article is not grounds for adding rooms. First look at whether bookings for three or more are thin relative to the level of your family share; if they are, check whether the per-person supplement increments are floating above the market range for your property type, or whether rooms that can take three or more people have enough visibility. This is not an argument for discounting; it is a check on whether you are capturing demand you should already be able to serve.

Q. Should I compare my own results against the all-types average?

A. Compare by property type and by area. The settled-month year-on-year figures for July 2026 do not line up in a single direction by type or by area: Tokyo at business hotels −5.3% and city hotels −17.1%; Fukuoka positive for both (+3.6% / +3.1%); Hokkaido at business hotels +4.1% and city hotels −5.7%. Using an all-types or national average as a benchmark is too coarse to act on.

Related Articles

  • JNTO Announces March 2026 Foreign Visitor Arrivals to Japan Reached 3,618,900, Up 3.5% Year-on-Year and a Record High for March

  • Golden Week 2026 Hokkaido Hotel Price Analysis: Niseko +29% and the Drivers Behind the Surge in Sell-Out Rates

  • Post-Golden Week Hotel Prices Drop Up to 44%: Why Mid-May Is the Best Time to Book

  • Golden Week 2026 Hotel Price YoY Analysis Across Six Major Cities: Unpacking the Drivers Behind Kyoto (+20%) and Tokyo (+17%)