How to Monitor a Hotel or Restaurant's Tripadvisor Reviews

“Tripadvisor reviews — why are they disappearing?” and “TripAdvisor gone bad, what’s the new alternative?” (373 upvotes) are both live threads with the same undertone: guests and owners alike have noticed reviews coming and going without explanation, and there’s no dashboard that tells either side what actually changed. If you manage a property’s reputation — or you’re benchmarking a competitor’s — that opacity is the actual problem, not a lack of review text to read.

This applies the same pattern as monitoring any other review platform for deletions, but Tripadvisor’s data shape is specific to hospitality: trip type, per-aspect sub-ratings, and a city ranking that moves independently of the star rating.

Pull the current review window

const API_KEY = process.env.FETCHLAYER_API_KEY;

async function fetchReviews(location) {
  const res = await fetch('https://api.fetchlayer.dev/tripadvisor/location-reviews', {
    method: 'POST',
    headers: { Authorization: `Bearer ${API_KEY}`, 'Content-Type': 'application/json' },
    body: JSON.stringify({ location, sortBy: 'recent', limit: 100 }),
  });
  return (await res.json()).reviews;
}

location takes a Tripadvisor location id or a page URL — resolve a name to an id once with search-locations and cache it. Reviews arrive 20 per page regardless of what limit asks for; pages is the actual depth control, so a limit of 100 without enough pages set will come back short. Sort by recent, not a relevance-style default, for the same reason it matters everywhere else you’re polling for change: a re-rankable default produces false “new review” signals when the ranking itself shifts between polls.

Catching reviews that disappear

Tripadvisor doesn’t expose a “removed” flag, but every review does carry a stable reviewId — “the dedupe key for incremental pulls” — so unlike some review platforms, there’s no need to fingerprint on content. Diff on id, and apply the same guard you’d use anywhere else you’re watching a paged listing shrink:

import { readFile, writeFile } from 'node:fs/promises';

const STATE_FILE = './tripadvisor-state.json';

async function run(location) {
  let seen;
  try {
    seen = new Set(JSON.parse(await readFile(STATE_FILE, 'utf8')));
  } catch {
    seen = new Set();
  }

  const reviews = await fetchReviews(location);
  const isFirstRun = seen.size === 0;
  const currentIds = new Set(reviews.map((r) => r.reviewId));

  const fresh = reviews.filter((r) => !seen.has(r.reviewId));
  const missing = [...seen].filter((id) => !currentIds.has(id));

  // If the poll window was fully replaced by newer reviews, a "missing"
  // review may have simply scrolled past it rather than been taken down.
  const confidentRemovals = fresh.length < reviews.length ? missing : [];

  reviews.forEach((r) => seen.add(r.reviewId));
  await writeFile(STATE_FILE, JSON.stringify([...seen]));

  if (isFirstRun) {
    console.log(`Seeded ${reviews.length} reviews.`);
    return;
  }

  console.log(`${fresh.length} new, ${confidentRemovals.length} likely removed.`);
}

Watching the rating trend, not just the reviews

A property’s overall bubble rating and its position in the local ranking move independently of any single review — location-profile carries both, plus a distribution across the five bubble ratings that shows whether a rating slide is broad or driven by a handful of very negative reviews:

async function checkProfile(location) {
  const res = await fetch('https://api.fetchlayer.dev/tripadvisor/location-profile', {
    method: 'POST',
    headers: { Authorization: `Bearer ${API_KEY}`, 'Content-Type': 'application/json' },
    body: JSON.stringify({ location }),
  });
  const { profile } = await res.json();
  return {
    rating: profile.rating,
    reviewCount: profile.reviewCount,
    ratingDistribution: profile.ratingDistribution,
    ranking: profile.ranking,   // rank + total among comparable places, e.g. "#118 of 526 hotels"
  };
}

Log rating and ranking.rank alongside your review diff on the same schedule. A rating that holds steady while ranking.rank slides means competitors are improving faster than you’re declining — a different problem than an actual drop in guest experience, and one the review text alone won’t tell you.

Segmenting by trip type

Hospitality reviews vary a lot by who’s writing them — a property can be excellent for couples and mediocre for families, and averaging across trip types hides that. tripTypes filters the pull itself rather than requiring you to filter client-side:

async function reviewsByTripType(location, tripType) {
  const res = await fetch('https://api.fetchlayer.dev/tripadvisor/location-reviews', {
    method: 'POST',
    headers: { Authorization: `Bearer ${API_KEY}`, 'Content-Type': 'application/json' },
    body: JSON.stringify({ location, tripTypes: [tripType], sortBy: 'recent', limit: 100 }),
  });
  return (await res.json()).reviews;
}

// business, couples, family, friends, solo
const familyReviews = await reviewsByTripType(location, 'family');

Run this once per trip type you care about rather than trying to reconstruct the split from an unfiltered pull — Tripadvisor’s own trip-type tagging is more consistent than pattern-matching review text for it yourself.

Practical notes

• translate is a translation target, not a language filter. Setting language: 'en', translate: true gets every review’s text translated to English — Tripadvisor still returns reviews originally written in other languages either way. Use reviewCountsByLanguage from location-profile to size a non-English segment before deciding whether translation is worth the pass.
• The city ranking label is hotel-only. ranking.label (“#118 of 526 hotels in New York City”) is published for hotels specifically; the underlying rank and total numbers exist for restaurants and attractions too, just without the pre-formatted sentence.
• isClosed is worth checking before alerting on a sudden review drop. A property Tripadvisor has marked permanently closed will naturally stop accumulating reviews — that’s not a monitoring failure, it’s the actual state of the listing.

Next Steps

• Tripadvisor Reviews API — full endpoint reference and pricing
• Get a free API key — no credit card required