SureFeed Hotel Rate Monitor

Compare observed Google Hotels prices for your chosen stays, see how a hotel compares with a selected competitor sample, and keep your own price history. Export the analysis as CSV, JSON or Excel.

Choose your starting point: collect fresh prices with the small live example below, or reuse your existing SureFeed datasets to avoid a new collection charge. The output shows observed prices, available comparisons and missing-data flags. See the first-run cost example.

Initial release. This is an analysis tool built on surefeed/google-hotels-scraper. It does not estimate occupancy, bookings, demand or revenue. Prices do not imply comparable room types, meals or cancellation terms.

Quick start: live prices

Copy this input into the Actor's JSON input and click Start. It requests up to 5 hotels in Rome for 2 adults, in EUR, for a two-night stay starting 14 days from today (UTC). Empty searches selects these moving dates, so there are no dates to edit for a first test. Live collection and monitor analysis are billed separately.

{
  "mode": "live",
  "datasetIds": [],
  "searches": [],
  "currency": "EUR",
  "country": "IT",
  "adults": 2,
  "maxHotels": 5,
  "sourceBudgetUsd": 0.05,
  "minPeers": 3
}

Before starting, add a historyName if you want to preserve this observation; reuse the same name in later runs. After the run, inspect the output dataset and export the results before increasing the sample.

For repeat price monitoring, use fixed stay dates. Set explicit query, checkIn and checkOut values in searches, then repeat the same search with the same guest count, currency and search country. Keep the same historyName. The moving default dates change each day and therefore do not track the same stay over time. A first observation establishes a baseline; a later compatible observation is needed to measure a price change.

First-run cost example

Live mode collects basic hotel results with optional hotel details disabled. At the published base rates, the example below applies if the run returns 5 charged source hotel results and produces 5 analysis rows, with the monitor running at 512 MB:

Component	Calculation	Cost
Google Hotels source collection	5 hotel results × $0.003	$0.015
Monitor analysis	5 analysis rows × $0.001	$0.005
Monitor start	1 start × $0.00005	$0.00005
Combined example total		$0.02005 — about 2 US cents

This is a worked example, not a fixed fee or a guarantee of 5 results. A two-night stay does not double the analysis count: each row describes a hotel observation for the whole stay. Source results and analysis rows are separate billing units; use the actual charged counts. Missing-price observations also count as analysis rows. Rates are in USD, before plan discounts; the requested hotel-price currency (EUR here) does not change the billing currency. Platform usage is included under the currently published pricing of both Actors. Check both Pricing tabs for your plan's current rates.

With mode: "datasets", there is no new source collection: analyzing 5 existing observations costs $0.00505 for the monitor under the same assumptions. Earlier collection charges are separate, and reanalyzing the same data is a new charged analysis.

The sourceBudgetUsd: 0.05 in the input is a source-event charge cap, not the expected charge and not an all-in spending limit. Set the monitor's maximum run charge separately. See full pricing and limits.

Start with existing data

Set mode to datasets and supply your own source dataset IDs in datasetIds. The source package also contains examples/input-datasets.json. The included example refers to 20 observations in Rome collected on 29 October 2026 for 20–22 November 2026, 2 adults, EUR, Italy. They are historical observations, not current offers. The dataset ID is an account-specific example; other accounts should use their own source datasets.

datasetIds accepts up to 10 datasets produced by the SureFeed Google Hotels scraper. Records must include entityId, name, query, dates, adults, currency, country, type and observation timestamp. Invalid records or conflicting duplicate prices stop the analysis instead of biasing results. Up to 5000 rows per source dataset and 10000 total per run.

Collect fresh prices

The Store example starts in live mode. Leave searches empty for a two-night Rome stay starting 14 days from today (UTC). For your own destinations, switch mode to live, clear datasetIds, and set searches: each object has query, checkIn, checkOut. Up to 10 searches and 100 hotels per search. examples/input-live.json compares two weekends. Update past dates before running.

Live mode starts the existing SureFeed Google Hotels Actor once per search, with a 300-second timeout per source run. The combined sourceBudgetUsd cap is divided evenly across searches. This is a source Actor event-charge cap, not a guarantee on total platform/compute/storage/proxy costs. Monitor platform usage is included in the monitor price shown below; the source Actor has separate billing. A source failure or wrong returned date/currency stops the analysis. Source run IDs remain in SOURCE_RUNS. A forcibly terminated parent can leave a source running until its own 300-second timeout; no automatic POST retries are used.

Comparisons

• price: total stay price divided by nights (preferred), or the nightly headline if no total is present. Rounded to 2 decimals. No currency conversion.
• Groups must match query, check-in/out, adults, currency, search country, tax basis, property type and observation timestamp. Price basis must also match. Unknown taxes suppress comparisons.
• competitorIds: exact entityId values. Empty means all compatible observed peers. The current hotel and optional targetHotelId are excluded from its peer benchmark.
• peerMedianPrice, vsPeerMedianPct, sampleRank require at least minPeers priced peers (default 3). Tied prices share the same rank. Rank refers only to the observed sample.
• priceChange, changePct compare the same hotel and stay context to its immediately previous observation. A missing previous price does not produce a fabricated change. Different dates or occupancy never share a time series.
• Missing price means price_not_observed, not sold out. Missing hotels are not inferred to be unavailable. Search samples may omit target hotels or competitors.
• roomConditionsVerified is always false: headline offers can differ in room, meal and cancellation terms.

History and exports

Supply historyName to append source observations to an account-specific named dataset. The same account and monitor name reuse that archive. The last 5000 appended rows are loaded for comparisons; historyWindowTruncated explicitly identifies older data outside the comparison window. The full archive remains available as a dataset. Existing observations are not appended again within the loaded window. Avoid overlapping runs for one monitor: concurrent runs can append duplicate observations, but subsequent analysis deduplicates them.

The default output dataset contains flat analysis records, exportable through Apify as CSV, JSON or Excel. SUMMARY in the default key-value store contains counts, history dataset ID and source dataset/run references. No email, messaging or recurring schedule is activated by this Actor. A first run establishes a baseline; real changes require subsequent observations for the same stays.

Local development

Node.js 22 or later. No third-party npm packages or installation step required.

npm test
npm run demo

Demo writes output/demo-analysis.json using the included observed data without network calls or charges. Tests use synthetic fixtures for change scenarios and mock API responses for orchestration.

Deploy the source ZIP to a new Apify Actor (ZIP source or import into its source editor), then Build with a fresh base image. Docker uses apify/actor-node:22; rebuild without cache when Apify publishes security updates. No API token is included in the code or examples. On Apify the runtime supplies APIFY_TOKEN and default storage IDs. The account token must permit source Actor runs and dataset storage; scoped tokens may require appropriate permissions.

Validated on Apify with 20 archived observations, an idempotent repeat, a 5-result live search and a 60-result multi-search run across Rome, Milan and two weekends. These are functional tests, not a guarantee of future availability or full market coverage.

Technical references

• https://docs.apify.com/api/v2/actors-runs-post
• https://docs.apify.com/api/v2/dataset-items-get
• https://apify.com/surefeed/google-hotels-scraper

Developed for SureFeed. No promise of increased revenue or full market coverage.

Permissions

Full permissions are required to read your source datasets, run the SureFeed source Actor and maintain your named history dataset. Do not enter your API token into the input; the Apify runtime supplies it. History is stored in your account. No messages or schedules are created.

Pricing

Base monitor rate: $1 per 1000 analysis rows ($0.001 each), plus the automatic Actor start event ($0.00005 at up to 1 GB). Monitor platform usage is included. Check the Pricing tab for your plan’s current rate; the Store’s “from” price may reflect a plan discount. Missing-price observations are also analysis rows and count toward billing. In live mode, the SureFeed Google Hotels source Actor is billed separately under its current pricing; sourceBudgetUsd limits its event charges only. Example: 20 analyzed observations cost $0.02005 for this monitor at 512 MB, plus source collection if requested. Reanalyzing an existing dataset is a new charged analysis. History archival itself is not an additional monitor event. Set the Apify maximum run charge as well as the source budget; these are separate limits.