# Hedonic Intelligence > Due diligence on the guests, not just the building. Before you buy, back, or reposition a hotel, Hedonic Intelligence reads every public guest review it has and tells you why its guests stay or leave, and what to fix. A finished report in 14 days, with no cooperation from the hotel. Hedonic Intelligence reads the full public review record of any hotel and returns the loyalty, value-verdict and friction patterns underneath its published rating, as one decision-grade PDF delivered in 14 days. It is built for hotel investors, asset managers, transaction advisors, and operators assessing rivals. How it sits next to the tools buyers already know: STR covers the market; reputation-management platforms (the category that includes ReviewPro and TrustYou) cover the operator's own rating from the inside; Hedonic Intelligence reads the guests themselves, from the outside, for the people deciding on the hotel. It works from public reviews only, so any property can be read, competitors included. Evidence base: 259,647 public guest reviews across 117 hotels in 8 countries, luxury to midscale, 25 to 26 languages. ## Key findings ### What does the published hotel rating miss about guest churn? The published rating does not separate guests who felt a stay was overpriced from guests who felt it was worth it. In a cohort of 259,647 reviews across 117 hotels, guests who wrote "overpriced" signal non-return at 57.0% (n=13,061), versus 2.7% (n=13,745) for guests who wrote "worth it", behind the same published score. This reads from what guests write, not from reservation data: it signals intent, never confirms a booking. Among satisfied (8 or higher) guests specifically, overpriced perception predicts 6.3x higher churn (18.8% versus 3.0%), and among the churners who name a specific item, 60.2% cite an ancillary charge (breakfast, parking, minibar), not the room rate itself. The fix is usually a line-item conversation with the operator, not a rate-card conversation with revenue management. Source: [Study 01, Chapter 1: A score is a settlement, not a verdict](https://www.hedonicintelligence.com/benchmark.html#value-verdict). ### Can two hotels with the same rating have very different guest loyalty? Yes. At an identical 8 out of 10, Luxury-tier guests signal non-return at 16.2% against Midscale guests at 4.0% (n=24,005 of 47,666 rating-8 reviews with a churn signal), a spread hiding inside a number a due-diligence memo would normally treat as fixed. This is a signal, not a booking record. The published rating is a trailing average; it conceals who is quietly leaving, and it conceals the most at the top of the market, where the room rate and the reputational stakes are both highest. Source: [Study 01, Chapter 3: Luxury does not buy immunity](https://www.hedonicintelligence.com/benchmark.html#tier-concealment). ### Is the gap between what a hotel's rating says and what its guests actually do a one-off, or a pattern? It is a pattern, not an outlier. Across the 117-hotel cohort, the mean per-hotel gap between overpriced-verdict churn and worth-it-verdict churn runs 53.8 percentage points (95% CI 51.3 to 56.4pp, median 54.1pp), and 86% of the 81 qualifying hotels (those with enough data in both cells to test) clear a 40-point gap. These are signals, not booking records. This reframes a due-diligence question from "is this specific asset's gap real" to "assume the gap is real until this asset's own data says otherwise." Source: [Study 01, Chapter 8: The DD Gap, across the whole cohort](https://www.hedonicintelligence.com/benchmark.html#dd-gap). ### What is The Portugal 500? The Portugal 500 is the first national index of Portugal's celebrated hotels' guest experience, seeded from recognition programs (Michelin Keys, Forbes Travel Guide, Leading Hotels of the World, Relais and Chateaux, SLH), read across four review platforms (Booking, Expedia, TripAdvisor, Google), from 1,596,417 extracted public reviews of 1,665,388 collected. Read the same way as the 117-hotel cohort. First release published 2026-07-10, 500 hotels; findings refresh on a periodic re-read cadence, not continuously. See the findings at [The Portugal 500](https://www.hedonicintelligence.com/portugal-500.html). ### How is Hedonic Intelligence different from ReviewPro, TrustYou, or STR? STR covers the market. Reputation-management platforms (ReviewPro, TrustYou) are bought by a hotel to manage its own rating from the inside. Neither is built for the person who has not bought the hotel yet, or the advisor who cannot get inside the operator's dashboard on a pre-acquisition target. Hedonic Intelligence reads any hotel from the outside, from public reviews only, and delivers a finished brief rather than a dashboard. ## Pages - [Home](https://www.hedonicintelligence.com/): positioning, the deliverable, method and perimeter, and FAQ. - [Study 01: The 117-Hotel Cohort](https://www.hedonicintelligence.com/benchmark.html): the full published study, eight chapters, one page, chapter by chapter, from the value-verdict finding through the cohort-wide DD Gap. - [Studies](https://www.hedonicintelligence.com/studies.html): hub for both investigations, plus a pointer to the sample read. - [The Portugal 500](https://www.hedonicintelligence.com/portugal-500.html): national-scale investigation of Portugal's celebrated hotels across four platforms; first release published 2026-07-10, 500 hotels. - [Sample read](https://www.hedonicintelligence.com/sample-read.html): a walk through the shape of a finished, paid, one-hotel deliverable, illustrated with cohort data on an anonymized property. A product demo, not a study. - [Privacy](https://www.hedonicintelligence.com/privacy.html): data handling and GDPR posture. ## Contact - [Book a 20-minute read intro](https://cal.com/rui-andrade-uo8qa8/20-min-call)