Skip to content
‹ Back to IndustriesHospitality

Shiji's Q3 benchmark finds hotel scores rising while guests say fewer good things

Shiji's Q3 2026 benchmark of 11.9 million hotel reviews shows its Global Review Index rising to 87.0%, while positive guest mentions fell to 69.8%. Hotels answer reviews in 3.1 days on average, yet nearly one in three goes unanswered. A score-only dashboard can miss slipping sentiment.

This story was produced through MarketScale. See how Hospitality teams put it to work with Executive Thought Leadership.

By MarketScale Newsroom · ShijiGuest Experience BenchmarkGlobal Review IndexHotel Reputation Management
Share
Shiji's Q3 benchmark finds hotel scores rising while guests say fewer good things

Key takeaways

01

The global hotel review score rose to 87.0% while positive guest mentions slipped to 69.8%, so a dashboard that tracks only the score can show progress guests are not describing.

02

Hotels cut average review reply time from 5.6 days to 3.1 days in two years, but the response rate sits at 68.1%. That is well past the roughly 40% level where earlier Cornell research found replying more stopped adding much value.

03

Post-stay surveys lose guests at the start, not the end: 9.7% respond, but 95.1% of those who begin a questionnaire finish it. Shortening the questions targets a step that already works.

Free workspace

Turn your Hospitality expertise into content.

Record interviews, organize footage, and write with AI on a free trial of the MarketScale platform for qualifying companies. No demo required, no credit card.

Try it Free

Shiji's latest benchmark covers 11.9 million hotel reviews and 45.7 million individual guest mentions from 10,200 hotels in seven regions, and its two headline numbers point in opposite directions. The Q3 2026 Guest Experience Benchmark puts the company's Global Review Index (GRI) at 87.0%, up 0.2 percentage points on a year earlier. Over the same period, the share of guest mentions that were positive fell from 70.5% to 69.8%.

Both moves are small. What matters is that they went in different directions. A director of guest experience who sends owners a monthly reputation number has a real choice here: report the figure that went up, or explain the one that went down.

Why the score and the comments split

For operators, the gap suggests that a steady or rising score can sit on top of written feedback that is becoming less positive.

The global review score rose to 87.0% while positive guest mentions fell to 69.8%, so a dashboard that tracks only the score can show progress guests are not describing. Shiji draws a similar conclusion, saying that headline ratings alone are an increasingly incomplete measure of the stay. Jessica Kurtz, Shiji's director of pre-sales and global enablement, put the quarter plainly: "positive sentiment can fall even while reputation scores rise." She said hotels get more from linking review sentiment, operational feedback and channel data than from watching one score.

What the score is worth, and how old that evidence is

Owners watch the score because it carries commercial weight, although the best-known evidence for that is now well over a decade old. According to ResortConcierge AI, Chris K. Anderson's November 2012 Cornell Hospitality Report compared ReviewPro's Global Review Index with monthly STR data covering January 2010 through June 2012 in 11 European and North American cities. For the full set of hotels, each 1% rise in the index went with a 0.96% rise in revenue per available room, alongside 0.80% higher average daily rate and 0.20% higher occupancy.

The link varied by segment, from 0.49% RevPAR for luxury hotels to 1.42% for midscale, the top of the range. The finding was a correlation, and the author cautioned that hotel performance depends on many factors the study did not measure. Treat it as dated context from a dataset that ends in mid-2012, not as confirmation that this quarter's 0.2-point rise is worth a measurable amount of RevPAR. Even so, it may explain why owners tend to treat a climbing index as good news.

Newer research points the same way, with a twist. A 2018 meta-analysis of 25 articles by Yang, Park and Hu put the average elasticity of hotel performance at 0.888 to review ratings and just 0.055 to review volume. Gregory Lewis of Microsoft Research and Georgios Zervas of Boston University followed 4,477 hotels in five western US states from 2005 to 2014. Their October 2019 working paper found that by 2014 a hotel rated one star higher across platforms had 27.8% higher demand, while the effect on price was "negligible."

For an owner reading the dashboard, that suggests a better score is more likely to show up in bookings than in rate, and that the score itself carries far more weight than the review count.

Where the negative comments cluster

It also matters where the negative comments cluster. Shiji reports that in several regions, a significant share of negative sentiment still comes from room-related issues, and its report calls room experience the biggest reputation risk. Shiji's view is that operational improvements in room quality often generate larger reputation gains than new amenities.

That could cost operators on price. According to ResortConcierge AI, Hu, Yang and Park published a meta-analysis in 2019 that pooled 163 estimates from 22 studies and put the median elasticity of room rates to ratings at 0.851. Its summary reports that rates were most affected by value and room ratings, and that ratings for staff and cleanliness were not significant. Shiji's complaint data and the academic pricing data therefore land on the same part of the stay.

There is a fair objection. By simple subtraction, the drop in positive mentions is seven-tenths of a point, measured across tens of millions of comments, and one quarter does not make a trend. The case for acting is strongest at properties where their own positive-mention share is falling faster than the global figure, or where room complaints are rising.

Replies got faster, coverage barely moved

Speed is the clearest operational change in the benchmark. In Q3 2026, hotels needed an average of 3.1 days to answer a review, down from 5.6 days in Q3 2024. Three-, four- and five-star properties all improved. Their averages now sit between 2.9 and 3.2 days, versus 4.7 to 5.4 days the previous year. Shiji suggests that wider use of AI-assisted reputation tools may explain some of the improvement.

Review response in Shiji's Q3 2026 benchmark

3.1 days
Average global time to respond to a review, Q3 2026
5.6 days
The same measure in Q3 2024, the baseline for the speed gain
68.1%
Share of reviews that got any reply, which leaves nearly one in three unanswered
About 40%
Response rate past which replying showed diminishing returns in 2016 Cornell research

Shiji Q3 2026 Guest Experience Benchmark; Cornell Center for Hospitality Research (2016) via ResortConcierge AI

Coverage tells a different story. According to ResortConcierge AI, hotels in one Shiji global benchmark of 12,000 hotels and more than 40 million reviews answered 68.8% of reviews in 2025, taking 3.5 days on average. The same report cites 67.2% for the second quarter of 2026. Those figures cover different reporting periods and may define response rate differently, so they are not a clean comparison with this quarter's 68.1%. Still, they suggest the industry has become much faster at answering reviews without answering many more of them.

That may matter less than it sounds. The ResortConcierge AI report cites 2016 Cornell research that found diminishing returns once a hotel answered about 40% of its reviews. The report argues that better replies beat more replies, and it questions the habit of answering everything as fast as possible. At 68.1%, the global average is well past that point.

What the research says about speed and content

Shiji's own view is that fast responses strengthen guest trust and improve reputation visibility across review platforms. The academic work is cooler on the clock. In a 2015 study by Min, Lim and Magnini, 176 university students rated hypothetical replies to a negative review: an empathy statement and a paraphrase of the complaint improved ratings of the reply, while speed "did not influence" them. Tripadvisor adds that management responses are not factored into its Popularity Ranking.

What a reply says carries more weight. Hotels that start responding saw a 0.12-star rise in ratings and 12% more reviews, according to Proserpio and Zervas in Marketing Science in 2017. Wu, Ye and Chu found response effects weaken among firms whose replies contain highly similar content, and Google's help page steers owners away from sending the same "thank you" to everyone. That is the catch for AI drafting: a tool that cuts response time by producing near-identical replies could deliver the speed without the effect.

Our reading, not a reported finding: for a reputation manager deciding where AI drafting time goes, the evidence points toward choosing which reviews get answered and how well they are answered, ahead of shaving another day off the clock. The exception is a property still answering fewer than 40% of its reviews. There, coverage is the first fix.

Ctrip and Expedia take bigger slices

The second shift concerns where guests post their reviews. Ctrip's share of hotel reviews worldwide doubled, moving from 3% to 6%. Its share in Asia climbed from 12% to 19%. In North America, Expedia's share increased from 41% to 45%. Shiji argues that every market now shapes a hotel's reputation through its own mix of platforms. Knowing the source of feedback therefore matters as much as the overall score.

Review volume is still concentrated on Booking and Google, Shiji's report notes, and it warns that hotels focusing only on those two risk missing a growing share of guest feedback. The Asian jump is where that risk is sharpest. For a hotel group that sets response rules and staffing centrally around a few platforms, Ctrip's larger slice there is feedback the workflow could miss. The first check is whether the reputation tool pulls in every channel that matters in each market, and whether someone on staff can answer in the language guests use there.

Asia leads on scores, three-star properties send mixed signals

Asia's 90.1% leads Oceania's 84.3% by 5.8 points, by simple subtraction. Global Net Promoter Score, which comes from guests' answers to a survey question about whether they would recommend the hotel, rose from 55.74 to 56.26. Latin America improved most, from 51.41 to 54.47, with gains in all three hotel classes.

The class breakdown repeats the quarter's main pattern. Five-star hotels recorded a global NPS of 62.29 and still post the highest reputation scores. Three- and four-star hotels narrowed that reputation gap, yet three-star NPS fell to 50.89. For three-star hotels, public review scores are catching up while guests' willingness to recommend them is falling.

Shiji's own summary says a strong reputation score does not automatically translate into recommendations, with luxury hotels increasing advocacy while some lower-tier categories weakened. It credits the narrowing score gap partly to value, suggesting strong value may be becoming as important as premium experiences. That fits the 2019 meta-analysis, where value sat alongside rooms as a main driver of rates, and it suggests three-star operators would do well to track recommendation intent next to the public score.

One number stands out: five-star NPS in North America fell from 57.39 to 39.85. Shiji itself says the sample size and property mix need checking before anyone draws broader conclusions, so treat it as a flag to investigate, not a finding.

Most survey dropout happens before the first question

Globally, post-stay survey emails had a 57.1% open rate and a 9.7% response rate. Of the guests who started a questionnaire, 95.1% finished it. Most guests are lost before they start. Once someone begins, the length of the questionnaire is rarely what stops them.

Teams often shorten the survey when response rates disappoint. Shiji's numbers suggest that fixes a step that already works. The invitation is the likelier target: when it arrives, how it's worded, and how many steps sit between the email and the first question.

Before cutting questions, split your own funnel into three rates: opened, started and completed. If your losses look like Shiji's global pattern, test the invitation first.

Platform rules also limit efforts to raise participation. According to ResortConcierge AI, a court can impose up to $53,088 per violation under the Federal Trade Commission rule on consumer reviews, and Google and Tripadvisor take a stricter line by banning incentives for any review. Teams that reward guests for completing private surveys should therefore keep those rewards apart from any request for a public review.

The next round of data could show which way this goes. If the positive-mention share falls again while the index holds steady, the gap starts to look like a pattern rather than a one-quarter blip. The North America five-star figure will either recover or hold once the sample is checked.

Featured companies

Your experts belong here

Every story in MarketScale Hospitality starts with a company putting its general managers, operations leads, and brand teams on the record. Buyers are already reading this topic. The only question is whose experts they find.

Owners and operators buy from people who understand the property, and your teams prove it in their own words.

Book DemoSee how it works15 minutes, straight to a calendar.

About the author

MarketScale Newsroom
MarketScale NewsroomEditorial Team, MarketScale

The MarketScale Newsroom reports on the companies, technologies, and trends shaping 16 B2B industries. It turns primary sources and expert commentary into clear, useful coverage for the people doing the work.

B2B Weekly

The week in Hospitality, and sixteen other industries, every Monday.

Ten stories, one-line takes, five minutes. Free.

Hospitality: are you visible to AI?

Before they reach out, Hospitality buyers ask AI engines which vendors to trust. Explore how your experts, customers, and partners can become useful content for buyers and AI search.

Free Trial

You just read one Hospitality expert. Your company is full of them.

This article was produced through MarketScale. The same platform turns your general managers, operations leads, and brand teams into the articles, video, and social content Hospitality buyers are searching for. Start a free trial and see it with your own people. For qualifying companies, no credit card, no demo required.

NPS +73 · 1,000+ creators · 38+ countries

What your free trial includes

Hands-on access to the MarketScale platform
Media requests to your crowd, remote recording, AI writing tools
No demo required. No credit card.
For qualifying companies. Company confirmation required.

More Hospitality Insights

TrustYou's new AI Report Builder targets the hotel owner who never logs in

TrustYou's new AI Report Builder targets the hotel owner who never logs in

Hotel groups using TrustYou can describe a report once and schedule it for delivery as a PDF or Excel in the owner's language. It's designed to ease the workload of quality teams managing multiple properties, with the trustworthiness hinging on data provenance checks.

  • 01For owners and boards who never log in, a scheduled PDF is the entire AI product, so the review data and date ranges behind it deserve the closest scrutiny.
  • 02An AHLA/Actabl survey found that 84% of respondents see AI's greatest value in operational efficiency, while only 15% find it beneficial for frontline work.
  • 03A useful pilot test is whether one sentence-built report holds up across a portfolio's real mix of KPIs, comparison periods and languages without an analyst cleaning it up.

Nov 17, 2026

Why hotel operators at Otelier's summit want daily numbers before AI agents

Why hotel operators at Otelier's summit want daily numbers before AI agents

Real hotel GOPPAR is still 20% below 2019, and costs are rising faster than revenue. Leading finance teams are turning month-end cleanup into a daily early warning, and operators want alerts in place of dashboards. AI comes after the data is trusted.

  • 01A day that balances can still be wrong, and an AI agent reading that day will act on it with full confidence, so trusted data and a named reviewer come before automation.
  • 02For 2027 budgeting, Sigala points to a lower starting point for rate assumptions—roughly 2.9%—while noting that undistributed expenses are growing slightly faster than revenue.
  • 03Hotels that budget once a year can start reforecasting weekly; operators already expect slower group pace in Q1 2027, which makes the first quarter the earliest test.

Nov 17, 2026

Vingcard says its first MENA wallet keys skip the hotel app

Vingcard says its first MENA wallet keys skip the hotel app

Orascom Hotels Management has rolled out Vingcard Wallet Key at three El Gouna, Egypt resorts via an integration with Hudini, covering more than 500 guestrooms. Vingcard calls it the first Wallet Key rollout in the Middle East and North Africa. The hardest part is keeping the tap-to-enter experience working through real stay changes—room moves, late checkouts and front-desk check-ins that must sync to the locks.

  • 01A wallet key earns its keep on the stay that goes off-script: El Gouna wired guest wallets, door locks, the property management system and check-in together so room changes, late departures and mid-stay requests don't push guests back to plastic cards.
  • 02Offering two ways to get a key, online pre-arrival check-in through a web app and a staff-generated QR code at the desk, keeps wallet keys open to guests who never check in online.
  • 03Watch where Orascom goes next. The group runs 34 hotels in four countries, and the El Gouna rollout so far covers 500+ guestrooms at three resorts.

Nov 15, 2026

Explore More Hospitality Insights

Read more expert perspectives from across Hospitality.

Browse Hospitality Hub

About the Expert

MarketScale Newsroom
MarketScale Newsroom

Editorial Team

MarketScale

The MarketScale Newsroom reports on the companies, technologies, and trends shaping 16 B2B industries. It turns primary sources and expert commentary into clear, useful coverage for the people doing the work.

For B2B teams

Your experts could be publishing here

Stories like this one run on content MarketScale captures from real practitioners. See how your team's expertise becomes coverage in Hospitality and beyond.

Book a Demo

Or call us. No forms required. We pick up. 214-945-2512