Data sources & verification

Where every piece of information on this site comes from.

Hotel identity

OpenStreetMap. Names, addresses, coordinates, phone numbers, brand tags and official website links. Used under the Open Database Licence and attributed on every hotel page and in the site footer.

OpenStreetMap is contributor-maintained, so some records are incomplete and some website links are out of date. Where a link is dead we record that the research pass found nothing, rather than inventing a substitute.

Coffee facts

Hotel and chain websites. The primary source, and the only Tier 1 one. Room descriptions, amenity lists and FAQs, read directly with a short excerpt stored as proof.

Travel media and dated guest reporting. Tier 2, used where it is more current than an official page.

Traveller submissions. Reports from people who actually stayed, reviewed by a moderator before they affect anything displayed. For hotels whose sites decline automated access, this is often the only route to a confirmed answer.

Reference data

Mains voltage, frequency and plug types by country are the published national standards. Brewing format and machine descriptions are general product knowledge, not claims about any specific hotel.

How a claim gets verified

  1. A source is fetched, subject to that site’s robots policy.
  2. The extractor records only what the page literally states, with the sentence.
  3. Claims from all sources are compared; disagreement is surfaced, not averaged.
  4. Confidence is scored from source tier, corroboration, freshness and confirmations.
  5. Anything conflicting or low-confidence is queued for human review.

The research methodology page covers each step in detail.

What we do not use

  • Scraped review-site content.
  • Copied hotel marketing descriptions.
  • Any source that declines automated access.
  • Inference from a hotel’s brand, class or price.

Current totals

7,307 sources recorded across 17,896 hotels in 27 countries. Generated live from the database.