Google Flights Scraper: Fares, Cheapest Days, Price Drops

Get flight prices from Google Flights for any route and date. Each row is one flight option with its price, airlines, flight numbers, departure and arrival times, duration, stops, layovers and CO2. Run it on a schedule and every row shows how the price moved since your last run.

It works for any pair of airports Google Flights covers. It is tested on JFK-LHR, AMM-DXB, RUH-CAI, RUH-LHR, DXB-BOM, CAI-JED, DOH-IST, JED-AMM, SFO-NRT and LAX-SYD.

Who uses it

• Travel agencies: quote fares for client routes and dates in your currency.
• Corporate travel teams: watch the routes your staff fly and get told when a fare drops.
• Fare alert and deal sites: track many routes every day and publish the drops.
• Analysts: study fares by route, season, cabin and number of stops.

What you get per flight

• Route, trip type, departure and return dates, cabin, passengers
• Price and currency. The price is the total for all passengers. For a round trip it is Google's round-trip total.
• Airlines, flight numbers, departure and arrival time (local), duration, number of stops
• Layover airports, cities and minutes, with a flag when you must change airport
• Each segment: flight number, airline, airports, times, aircraft, legroom, CO2, codeshare numbers
• CO2 in kg, the typical CO2 for the route, and the difference in percent
• Google's flags: isBest (Google's "best flights" list) and isCheapest (lowest price in the search)
• Google's price level for the search ("low", "typical", "high") and the typical price range
• Price watch: priceStatus (new, dropped, rose, unchanged), previousPrice, priceChange, priceChangePercent
• A Google Flights link for the same search

Each search also gives one free status row. It says how many flights Google returned, the cheapest price, and how the cheapest price moved since the last run.

Input examples

One-way, fixed date

{
    "routes": ["JFK-LHR", "AMM-DXB"],
    "departureDate": "2026-12-04",
    "adults": 1,
    "currency": "USD"
}

Round trip, 2 adults, Jordanian dinar

{
    "routes": ["AMM-DXB", "RUH-CAI"],
    "departureDaysFromNow": 30,
    "tripLengthDays": 7,
    "adults": 2,
    "currency": "JOD",
    "country": "jo"
}

Business class, nonstop only, Saudi riyal

{
    "routes": ["RUH-LHR"],
    "departureDate": "2026-12-15",
    "cabin": "business",
    "stops": "nonstop",
    "currency": "SAR",
    "country": "sa"
}

Cheapest day to fly in a two-week window

The actor searches every departure day from departureDate to departureDateTo. With the default output you get one row per day: that day's cheapest flight.

{
    "routes": ["AMM-DXB", "RUH-LHR"],
    "departureDate": "2026-12-01",
    "departureDateTo": "2026-12-14",
    "tripLengthDays": 7,
    "currency": "SAR",
    "country": "sa"
}

Input tips

• Write a route as two IATA airport codes: JFK-LHR, amm dxb and CAI to JED all work.
• Leave the departure date empty and set Days from today for a schedule that always looks the same distance ahead.
• Stops sets the most stops allowed: any, nonstop, 1 stop or fewer, 2 stops or fewer.
• Max flights per search limits rows per route and date. Google's best flights come first.

Output example

This row comes from a real run on 2026-11-04 (JFK-LHR, 4 December 2026, one adult, economy). It was the second run of the same search, so the price watch fields are filled.

{
    "type": "flight",
    "route": "JFK-LHR",
    "tripType": "one_way",
    "departureDate": "2026-12-04",
    "returnDate": null,
    "cabin": "economy",
    "passengers": 1,
    "price": 298,
    "currency": "USD",
    "priceType": "one_way_total_all_passengers",
    "rank": 6,
    "isBest": false,
    "isCheapest": true,
    "airlines": "Aer Lingus",
    "flightNumbers": "EI 104, EI 152",
    "departureAirport": "JFK",
    "arrivalAirport": "LHR",
    "departureTime": "2026-12-04T17:00",
    "arrivalTime": "2026-12-05T08:05",
    "durationMinutes": 605,
    "duration": "10h 5m",
    "stops": 1,
    "layoverAirports": "DUB",
    "layoverMinutes": 125,
    "layovers": [
        { "airport": "DUB", "airportName": "Dublin Airport", "city": "Dublin", "minutes": 125, "changeOfAirport": false }
    ],
    "aircraft": "Airbus A330, Airbus A320neo",
    "co2Kg": 392,
    "co2TypicalKg": 434,
    "co2DiffPercent": -10,
    "priceLevel": "typical",
    "typicalPriceLow": 170,
    "typicalPriceHigh": 470,
    "priceStatus": "unchanged",
    "previousPrice": 298,
    "priceChange": 0,
    "priceChangePercent": 0,
    "previousCheckAt": "2026-11-04T10:55:40.100Z",
    "segments": [
        {
            "flightNumber": "EI 104", "airline": "Aer Lingus", "from": "JFK", "to": "DUB",
            "departure": "2026-12-04T17:00", "arrival": "2026-12-05T04:30", "durationMinutes": 390,
            "aircraft": "Airbus A330", "legroom": "31 in", "co2Kg": 340, "alsoSoldAs": []
        },
        {
            "flightNumber": "EI 152", "airline": "Aer Lingus", "from": "DUB", "to": "LHR",
            "departure": "2026-12-05T06:35", "arrival": "2026-12-05T08:05", "durationMinutes": 90,
            "aircraft": "Airbus A320neo", "legroom": "29 in", "co2Kg": 52, "alsoSoldAs": []
        }
    ],
    "googleFlightsUrl": "https://www.google.com/travel/flights/search?tfs=GhoSCjIwMjYtMTItMDRqBRIDSkZLcgUSA0xIUkABSAGYAQI&hl=en&gl=us&curr=USD",
    "scrapedAt": "2026-11-04T11:10:30.695Z"
}

The free status row for the same search:

{
    "type": "search_status",
    "route": "JFK-LHR",
    "departureDate": "2026-12-04",
    "status": "ok",
    "flightsFound": 21,
    "flightsWithPrice": 21,
    "flightsDelivered": 20,
    "cheapestPrice": 298,
    "previousCheapestPrice": 298,
    "cheapestPriceChange": 0,
    "typicalPriceLow": 170,
    "typicalPriceHigh": 470,
    "priceLevel": "typical",
    "charged": false
}

A date scan gives one free status row per route. This one comes from a real 14-day round-trip scan in SAR:

{
    "type": "scan_status",
    "route": "AMM-DXB",
    "status": "ok",
    "firstDepartureDate": "2026-12-01",
    "lastDepartureDate": "2026-12-14",
    "tripLengthDays": 7,
    "daysScanned": 14,
    "daysWithPrices": 14,
    "daysFailed": [],
    "cheapestDepartureDate": "2026-12-01",
    "cheapestReturnDate": "2026-12-08",
    "cheapestPrice": 1038,
    "charged": false
}

Status values: ok, no_results (Google lists no flights), no_prices (flights listed without a price), invalid_search (Google could not use an airport code), failed (Google kept refusing), partial_cost_limit and skipped_cost_limit.

Daily price tracking

1. Set fixed dates (departureDate, and returnDate for a round trip).
2. Give the watch a name in Watch list name, for example amm-dxb-december.
3. Schedule the actor in Apify (Schedules, then for example every day at 08:00).

Every run saves the prices it saw in the key-value store google-flights-price-watch. The next run with the same name compares each flight by its flight numbers:

• priceStatus: new, dropped, rose or unchanged
• previousPrice, priceChange and priceChangePercent
• In the status row: previousCheapestPrice and cheapestPriceChange for the whole search

Turn on only output new or changed prices to get changes only, or only output price drops for drop alerts. Rows that are left out are not charged. Saved prices for past dates are removed automatically.

With Days from today the date moves every day, so each run prices a new date. Use fixed dates for tracking.

Pricing

Pay per result. Platform usage is included in the price.

• $0.50 per 1,000 flights (event flight). One flight option with a price.
• $4.00 per 1,000 cheapest days (event cheapest-day). One row per scanned day with that day's cheapest flight. Each day is a full Google search, so it costs more than a flight row. Choose All flights for every day in a scan to get every flight at the flight price.
• Free: status rows, flights without a price, unchanged or risen flights left out by the price-watch filters, failed searches.

Example: 10 routes with 20 flights each cost $0.10. A 30-day cheapest-day scan of one route costs $0.12.

You can set a maximum cost per run in Apify. The actor stops when the next row would go over it.

Speed and limits

• About 1 to 10 seconds per search, depending on retries. 10 routes took 85 seconds in the Apify cloud. The default input (2 routes) finishes in about 5 seconds.
• Google lists about 10 to 35 flights per route and date. Up to 100 rows per search.
• Departure up to 330 days ahead. Up to 9 passengers (adults plus children). A date scan covers up to 62 days. One run does up to 500 searches.
• Infants are not supported. Google's public results page returns no flights for searches with infants.
• Round trips list the outbound flights with Google's round-trip total. The return flight is chosen on Google's next page and is out of scope for now.
• Booking options per seller, multi-city trips and airline filters are out of scope for now.
• Prices depend on the market. Set Country and Currency to match the traveller.
• Google sometimes lists a flight without a price, mostly for searches with children. Those flights are skipped and counted in the status row.

Reliability

• Requests start at least 1 second apart. Two searches run at a time by default.
• Google sometimes answers with a page that has no flight list, mostly to cloud IP addresses. The actor treats that as a refusal: it starts a new session, waits, and then moves up the proxy ladder: no proxy, Apify datacenter, Apify residential.
• The run always ends with status SUCCEEDED. Each search gets a status row, and failed searches are free.
• The run summary (key-value store record RUN_SUMMARY) shows requests, HTTP status codes, retries, the proxy tier used and bytes per search.

Data and privacy

The actor reads public Google Flights result pages. It does not log in, and it does not solve captchas. Rows hold flight and fare information only, with no personal data. Check the rules that apply to your use of the data, such as Google's terms and data protection law in your country.