Google Flights Scraper & Price Monitor API 💰$2/1k

Keywords: google flights scraper, google flights api, flight price scraper, google flights price history, airfare monitor, cheapest flight finder api, flight price tracker, google flights scraper no login, fare alert feed, travel price data, flight itinerary scraper, google flights price insights, round trip flight prices

Scrape Google Flights itineraries and fares for any route and date grid: price, exact price to the cent, airlines, flight numbers, departure and arrival times, duration, stops, layovers, aircraft, legroom and CO2. Add a cheapest-fare snapshot per route-date with Google's price insights and about 60 days of daily price history, so your very first run already has two months of history.

Turn on monitoring mode and schedule it daily: the actor remembers every fare and outputs only flights that are new or re-priced since the last run, or skips route-dates whose cheapest fare did not move. You pay for changes, not for the same fares again (with onlyNew, set includeSummaryRow: false too, otherwise the per-route-date snapshot row is still output and charged on every run).

No browser, no Google login, no API key. One fast request per route-date (about 20-90 KB), so a 30-route x 60-date grid finishes in minutes.

Who it is for

• Fare-deal newsletters and alert channels: re-query a fixed route x date grid every morning and publish the drops (changeType, priceChangePct).
• Travel agencies, TMCs and OTA analytics teams: benchmark your fares against Google's "best" bucket per route and date.
• Airline revenue-management and consulting shops: track competitor fares and schedules on specific routes, daily.
• Researchers and data journalists: fares, CO2 per itinerary and route-level price history.

What you get

Two row types in one dataset (rowType):

Field	Description
rowType	itinerary (one per flight option) or fare-snapshot (one per route-date)
queryOrigin, queryDestination, queryDate, returnDate, tripType, cabin, adults, currency	What was searched
bucket	Google's best or other list
price	Display price (total for all passengers, rounded up by Google)
priceExact	Exact price decoded from Google's booking token, e.g. 294.50
airlineCodes, airlines	Carriers on the itinerary
departDate, departTime, arriveDate, arriveTime, durationMin, stops	Schedule (local times)
layovers[]	Airport, airport name, city, minutes
legs[]	Per flight: from/to, times, flightNumber, carrier, aircraft, legroom, CO2
flightKey	Flight numbers joined by +, the stable key for change detection
co2Grams, co2TypicalGrams, co2DiffPct	Google's emissions estimate vs typical for the route
bookingToken	Google's token for the offer
changeType, previousPrice, priceChangePct	Monitoring mode: new, changed (unchanged on snapshots)
resultCount	Itineraries Google returned for the route-date
insightLevel, insightCurrentLowest, insightTypicalLow, insightTypicalHigh	Google price insights (snapshot rows)
priceRangeMin, priceRangeMax	Cheapest and most expensive fare for the route-date
priceHistory	About 62 daily [YYYY-MM-DD, price] points (snapshot rows)
exitCountry, source, scrapedAt	Run metadata

Input

Field	Default	Description
origins / destinations	["JFK"] / ["LHR"]	IATA codes; every origin is paired with every destination. A Google city id such as /m/02_286 (New York) searches all its airports at once.
routes	[]	Alternative: explicit pairs like DXB-LHR, SFO>NRT
dates	[]	Explicit departure dates YYYY-MM-DD
daysAheadFrom / daysAheadTo / everyNDays	30 / 32 / 1	Rolling window used when dates is empty (best for schedules)
tripType	one-way	round-trip returns outbound options priced as a round trip
returnAfterDays / returnDate	7 / empty	Round-trip return date
cabin	economy	economy, premium_economy, business, first
adults	1	Prices are totals for all passengers
maxStops	empty	Keep itineraries with at most N stops
currency / language / country	USD / en / US	Forced in the URL, so the proxy exit country does not change the data
outputMode	itineraries	or cheapest (one snapshot row per route-date)
includeSummaryRow	true	Add the snapshot row in itineraries mode
maxItinerariesPerSearch	15	Keep the first N itineraries per route-date (Google returns about 20-30)
onlyNew	false	Monitoring: output only new or re-priced itineraries
onlyChanged	false	Monitoring: skip route-dates whose cheapest fare is unchanged
minPriceChangePct	0	Ignore smaller price changes in onlyNew mode
maxSearches / maxItems	20 / 50	Hard caps per run (route-date searches / rows and charges). Searches run route by route (all dates of the first route first), so a low maxSearches can leave later routes unsearched. Duplicate routes are searched once.
ssrFallback	true	Load the full search page if the fast call fails
proxyConfiguration	no proxy	Optional Apify Proxy; see FAQ

Example: daily grid of 3 routes x the next 60 days, cheapest fare only:

{
  "routes": ["JFK-LHR", "LAX-NRT", "DXB-LHR"],
  "daysAheadFrom": 1,
  "daysAheadTo": 60,
  "outputMode": "cheapest",
  "onlyChanged": true,
  "maxSearches": 200,
  "maxItems": 200
}

Output sample

Real rows from a test run on 2026-10-03 (JFK-LHR, 2026-11-02, one-way economy; truncated):

{
  "rowType": "itinerary",
  "queryOrigin": "JFK", "queryDestination": "LHR", "queryDate": "2026-11-02",
  "tripType": "one-way", "cabin": "economy", "adults": 1, "currency": "USD",
  "bucket": "best", "price": 299, "priceExact": 299,
  "airlineCodes": ["AA"], "airlines": ["American"],
  "origin": "JFK", "destination": "LHR",
  "departDate": "2026-11-02", "departTime": "17:29", "arriveDate": "2026-11-03", "arriveTime": "09:20",
  "durationMin": 651, "stops": 1,
  "layovers": [{ "durationMin": 63, "airport": "RDU", "airportName": "Raleigh-Durham International Airport", "city": "Raleigh" }],
  "legs": [
    { "from": "JFK", "to": "RDU", "departTime": "17:29", "arriveTime": "19:32", "durationMin": 123, "flightNumber": "AA4501", "aircraft": "Embraer 175", "legroom": "30 in", "co2Grams": 110704 },
    { "from": "RDU", "to": "LHR", "departTime": "20:35", "arriveTime": "09:20", "durationMin": 465, "flightNumber": "AA174", "aircraft": "Boeing 777", "legroom": "31 in", "co2Grams": 502495 }
  ],
  "flightKey": "AA4501+AA174",
  "co2Grams": 613000, "co2TypicalGrams": 431000, "co2DiffPct": 42,
  "bookingToken": "CjRIcjVkZ1lRb0xDd0VBSENnWXdCRy0tLS0tLS0t...",
  "changeType": null, "resultCount": 31, "source": "rpc", "exitCountry": "MY",
  "scrapedAt": "2026-10-03T15:53:50.554Z"
}

{
  "rowType": "fare-snapshot",
  "queryOrigin": "JFK", "queryDestination": "LHR", "queryDate": "2026-11-02", "currency": "USD",
  "price": 299, "priceExact": 299, "airlines": ["Delta", "Virgin Atlantic"], "flightKey": "DL5638+VS22", "stops": 1,
  "resultCount": 31,
  "insightLevel": 5, "insightCurrentLowest": 299, "insightTypicalLow": 170, "insightTypicalHigh": 295,
  "priceRangeMin": 299, "priceRangeMax": 2884,
  "priceHistory": [["2026-08-04", 295], ["2026-08-05", 295], ["2026-08-06", 295], "... about 62 points"]
}

Pricing (pay per event)

Event	What you pay for	Free plan	Starter (Bronze)	Scale (Silver)	Business (Gold)
itinerary	One itinerary row (only new or re-priced rows in onlyNew mode)	$0.002	$0.0018	$0.0016	$0.0014
fare-snapshot	One cheapest-fare row per route-date with insights and 60-day history	$0.005	$0.0045	$0.004	$0.0035

That is $2 per 1,000 itineraries and $5 per 1,000 route-date snapshots on the free plan, plus Apify's small actor-start fee. Route-dates with no flights (invalid route, or Google returned no itineraries) are output as a row with error set and are not charged. The default input costs under $0.11. Set a maximum cost per run in the Console and the actor stops cleanly when it is reached.

How to schedule daily (monitoring mode)

1. Create a task with your routes and a rolling window, e.g. daysAheadFrom: 7, daysAheadTo: 60, and set onlyNew: true (itineraries) or outputMode: "cheapest" with onlyChanged: true (one row per route-date that moved).
2. Add a schedule to the task: daily at 05:00 UTC (0 5 * * *). For departures within 14 days, a second task every 6 hours catches intraday moves.
3. Connect the dataset to Google Sheets, a webhook, Slack, Zapier, Make or n8n. Each run outputs only what changed, with previousPrice and priceChangePct.

The state lives in the named key-value store google-flights-fare-feed-state in your account (one record per route, date, cabin, passengers and currency). Delete it to start fresh. Keep adults, cabin and currency fixed per task so prices stay comparable.

Use it from the API, n8n, Make, Zapier or an AI agent

• API: start a run and get the results in one call:

  curl -X POST "https://api.apify.com/v2/acts/programmx~google-flights-fare-feed/run-sync-get-dataset-items?token=YOUR_APIFY_TOKEN" \
    -H "Content-Type: application/json" \
    -d '{"routes": ["JFK-LHR"], "daysAheadFrom": 30, "daysAheadTo": 32}'
  

  The response is the dataset as JSON; add &format=csv for CSV. This endpoint waits up to 5 minutes; for longer runs, start the run asynchronously and read the dataset when it finishes.
• n8n, Make and Zapier: use the Apify node or app, choose Run Actor, pick this actor and paste the input JSON.
• Google Sheets, Slack, email and webhooks: add an integration on the actor or task in Apify Console.
• AI agents: add this actor to Apify's MCP server (https://mcp.apify.com) and call it from any MCP client.

FAQ

Do I need a proxy?

Usually not: in our tests Google Flights answered without one. If you see blocks, select Apify Proxy (datacenter first, then residential). The actor checks proxy health before the run and retries each request up to 4 times with a fresh session.

Why is the price total for all passengers?

That is how Google shows it. 2 adults = about twice the one-adult fare.

What is priceExact?

Google rounds the display price up to a whole unit; the booking token carries the exact amount (e.g. 294.50), which makes change detection reliable.

Which currencies work?

Any code Google supports; USD, EUR, GBP, MXN and ZAR were verified. If Google rejects a currency, the actor retries the search in USD and labels the rows accordingly.

Does it return the return flights of a round trip?

Not yet: round-trip searches return the outbound options with the round-trip total price, like Google's first results page.

Why do some searches take a few seconds?

Google sometimes answers a valid search with a temporary error; the actor retries it automatically (counted in RUN_STATS.errorResponseRetries). An invalid route is reported after 5 tries (10 with a non-USD currency, because a rejected currency is retried in USD) and is not charged.

Can I get statistics for a run?

Yes, the RUN_STATS record in the run's default key-value store has requests, retries, rows per search, proxy mode, exit country and duration.

How much does daily fare monitoring cost?

Example on the Free plan: 5 routes × 30 departure dates in cheapest mode is 150 route-dates. If every one changed, a run would cost 150 × $0.005 = $0.75; with onlyChanged you pay only for the route-dates whose cheapest fare moved.

Can I export Google Flights data to CSV or Google Sheets?

Yes. Each run's dataset downloads as CSV, Excel or JSON, and the Google Sheets integration can append every run to a sheet.

Can an AI agent use this actor?

Yes. Apify's MCP server (https://mcp.apify.com) can expose this actor as a tool to Claude, Cursor, VS Code and other MCP clients, so an agent can fetch fresh fares on request.

Limitations

• One-way and round-trip (outbound options) only; multi-city is not supported. Return-leg selection is not implemented.
• Google's price-insight labels (insightLevel, typical low/high) are not documented by Google; the field meanings are our best reading.
• Prices are what Google shows to an anonymous user at the time of the run; they can differ from the airline's checkout.
• Google can change its internal response format at any time; the actor flags a suspected change in RUN_STATS.parseDriftSuspected and the maintainer fixes it.
• Children, infants and loyalty fares are not supported.