Scoring rubric · v2026.1

The Standard

Every property review on this site carries a Standard score. Five dimensions, scored 1 through 5, averaged to a property score with one decimal place. A property scoring 4.6 or above is At the Standard. Between 4.0 and 4.5 is Within Reach. Below 4.0 is Below the Standard — and yes, we publish those.

How a property gets reviewed

The desk pre-screens the property — opening date, ownership group, GM tenure, F&B operator, the architecture and interior credits, the published rate band. If any of those don't clear a basic check, the property does not go on the visit calendar.
An editor visits for at least two nights, paid by {LTS Editorial LLC}. We do not accept press-rate visits, host-fee discounts, or comped F&B. If a property's PR contact has facilitated a visit, the visit is held until a self-paid follow-up confirms the experience.
The editor scores each of the five dimensions on a 1-to-5 scale. No dimension is left blank. No dimension is scored below 3.5 without a written justification paragraph.
The piece is fact-checked against named entities, dates, suppliers, architects, GMs, F&B operators, and rate cards. Any single inconsistency the desk cannot resolve sends the piece back into hold.
The piece is filed against the editorial standard (currently v2026.1) and published with the dimension scores in the article frontmatter and the published Standard score in the header.

The five dimensions

Setting

Site, architecture, integration with place. Includes views, light quality, soundscape. A property at the Setting standard is one where the location does not feel borrowed — the building belongs to where it sits, not to a global property template.

Suites

Rooms, villas, residences. Includes bathroom, terrace, materiality, scale, in-room technology. A property at the Suite standard is one where the room you book is not a step down from the room in the press image.

Service

Observed staff-to-guest ratio, name recognition, recovery on a deliberately-introduced problem, language fluency, butler conduct. We test service deliberately: a minor billing query, a last-minute itinerary change, a reservation we ask the property to chase on our behalf. The recovery and the conduct around it are what score.

Table

F&B across all venues. Includes breakfast, signature restaurant, in-villa dining, wine list depth. Not every property needs a Michelin-track signature room. Every property at the Table standard has a breakfast that is composed, sourced, and timed properly.

The Detail

The single hardest dimension to fake. Turn-down, the welcome amenity, the un-asked-for, the operating temperature of the pool, the brand of soap, the weight of the towel, the gauge of the linen, the temperature of the espresso, the bias in the curtains. A property at the Detail standard is one where the operator has done the work nobody is watching for.

What disqualifies

Single-night reviews. We never publish based on fewer than two nights in the property.
Pre-opening reviews. We may publish opening reports based on tours of unfinished properties, but we never publish a scored review until the property has opened and our editor has slept in a paying-guest room.
Properties under management transition. If a property is in the first 90 days of a new GM or a new operator, the score is held until the operating team has bedded in.
Properties where the desk has a conflict. If the editor has a personal or financial relationship with the owner, brand, or operator, the piece is reassigned or held.

The verdict bands

At the Standard — 4.6 to 5.0: The property delivers on every dimension. We have stayed and would book again at full rate.
Within Reach — 4.0 to 4.5: The property is close but has at least one dimension that needs work. We have noted what.
Below the Standard — under 4.0: The property does not currently meet the bar. The review explains where and why. We re-visit when the operating team has addressed the issues.

For the broader editorial standard governing every piece — sourcing, fact-checking, anonymous-source policy, AI policy, what we will not publish — see /editorial-standards/. For the operational corrections log, see /corrections/.