Three datasets that should not be confused
Shopping offers describe flight options returned for a search. A booking integration may let you reprice and reserve them. Booking records describe reservations or transactions you are authorized to access. Website observations record what a source displayed for a particular trip and collection time.
A fare observation is not evidence that someone bought a ticket. A flight-search API key does not automatically grant access to industry-wide booking history.
| Your task | Data to evaluate | Key question |
|---|---|---|
| Sell and service flights | GDS, NDC or another booking integration | Can it reprice, create orders and handle the servicing you need? |
| Report on an agency’s bookings | Authorized reservation and transaction feeds | Does it include the actual records and changes for that agency? |
| Observe an airline’s website price | Direct website fare observations | Was the same flight, party, fare context and time checked? |
| Analyze market trends | A suitable historical dataset or a repeated, comparable collection program | Are the history, sampling method and missing observations understood? |
Where GDS and booking APIs help
They can connect structured shopping data to pricing, booking and servicing workflows. For example, Amadeus Quick-Connect documents search, price confirmation and order creation. Duffel describes content from NDC, GDS and low-cost channels, alongside booking and ancillary services.
The actual content, negotiated fares and permissions depend on the product and agreement. It is inaccurate to assume that every GDS lacks low-cost airlines, or that an airline website always offers a lower price.
Where website observations can add another view
They can help a team inspect the direct website offer for a defined trip, compare it with another source or evaluate a coverage gap. That value depends on successful collection, field completeness and a matched comparison—not simply on using scraping.
Multiple sources can be complementary. OAG’s Airfare Data describes fare intelligence collected from airlines, GDSs and online travel agencies. Its historical data product is a different scope from a single on-demand observation.
What makes two prices comparable?
- Match the flights, dates, direction, passenger count and passenger types.
- Match cabin and fare family, baggage, inclusions and fare conditions when known.
- Keep original currency and verified market context visible. A requested locale is not proof of the effective point of sale.
- Record collection times and separate observed totals from calculated totals.
- Do not treat a later lower quote as proof that an earlier booking was overpriced.
Where Farelint fits
Farelint is an early-stage airline fare-observation API. It supplies normalized records for supported pilot workflows through HTTP and MCP. It does not provide ticketing, agency booking records or an established market-wide historical dataset.
If you already have a provider, start by identifying a specific missing input. Then compare a bounded sample for accuracy, usable result rate, freshness and cost before deciding whether another supplier helps.
Written by Farelint. Product scopes can change; the linked provider documentation and your agreement govern access.