Traveloka Hotel & Accommodation Scraper

Pulls hotels, villas and apartments from Traveloka as structured JSON: the full property profile, room inventory, and a whole month of nightly rates per property with an automatic weekday-versus-weekend comparison.

It talks to Traveloka's own JSON API rather than rendering pages in a browser, so a property costs a handful of HTTP requests instead of a headless Chrome session.

What you get per property

Group	Fields
Identity	hotelId, name, displayName, url, accommodationType (Hotel / Villa / Apartment), starRating
Location	address, city, province, country, postalCode, latitude, longitude, areaName, areaGeoId
Property	checkInTime, checkOutTime, facilityCodes + facilityNames (100+ per property), supportedRateTypes, images
Rooms	rooms[] — name, description, bedding, room size, occupancy, view, smoking policy, amenities, photos
Pricing	priceCalendar[] — one entry per night: date, dayOfWeek, isWeekendNight, price, availabilityStatus, isCheapestOfMonth
Pricing summary	priceSummary — minPrice/maxPrice with their dates, averagePrice, weekdayAveragePrice, weekendAveragePrice, weekendPremiumPercent, nightsUnavailable
Reviews	reviewSummary — rating, ratingScale, reviewCount, ratingLabel

weekendPremiumPercent is the number most revenue managers actually want: how much more a competitor charges for a Friday or Saturday night than for a weekday.

Choosing properties

Four ways in, and you can mix them:

• Property URLs — paste Traveloka hotel pages. Only the numeric ID at the end of the URL matters, so shortened or re-slugged links still work.
• Property IDs — bare numeric IDs from an earlier run.
• Property name searches — matches property names, not cities.

  Kempinski Jakarta
  returns that hotel; hotels in Jakarta returns nothing useful. This is the right tool for assembling a named competitor set.
• Sitemap sweep — samples property IDs from Traveloka's public sitemaps for broad market analysis. Use sitemapSkipLeaves so repeat runs sample different properties instead of the same first page.

What this actor cannot do, and why

There is no "every hotel in city X, with prices" mode. Traveloka's search-list endpoint, and the dynamic per-room rate endpoint behind it, are protected by AWS WAF with the captcha action — a visual challenge that no headless client can answer. This was verified from clean residential Indonesian IPs as well as from datacenter IPs, on the first request of a fresh session with a valid WAF challenge token: the answer is captcha every time. It is a structural rule, not an IP-reputation problem, so proxies do not change it.

What that rules out, concretely:

• listing all properties in a city or around a map area;
• per-room nightly rates and per-room discount/strikethrough prices.

What replaces it: the price calendar, which is not WAF-guarded and returns the property's nightly lead-in rate for every date in a month. For tracking a competitor set's pricing over time this is strictly more useful than a single search snapshot — one request gives you 31 data points instead of one.

starRating is the official star classification. reviewSummary.rating is the guest score out of 10. availabilityStatus of NO_ALLOTMENT means sold out for that date, which is itself a useful demand signal.

Cost and tuning

Requests per property: 1 profile + 1 review summary + 1 room list + 1 per calendar month. A 12-month pull is 15 requests. Turn off includeRooms or includeReviewSummary for leaner runs, and keep calendarMonths at 1 unless you need the forward curve.

Traveloka's JSON API answers datacenter IPs normally, so the default Apify proxy is fine. Residential Indonesian proxies do not unlock anything extra here.

Output record types

Every record carries a recordType so failures are visible rather than silent:

• PROPERTY — scraped successfully.
• NOT_FOUND — the ID does not exist. Traveloka does not 404 for these; it echoes the ID back with every other field null, which this actor detects.
• GATED — an endpoint answered with an AWS WAF challenge.
• ERROR — a request failed; see error.

Partial failures do not lose the record: anything that could not be fetched is listed in warnings and the rest of the property is still returned.

A note on robots.txt

Traveloka's robots.txt disallows /api/, which is where this actor reads from. It allows /*/sitemap/**, which is what the sitemap sweep uses. Decide for yourself whether that fits your use of the data, your jurisdiction and your agreement with Traveloka. Keep request rates modest and scrape only what you need.