Methodology

Airfare intelligence requires transparency

We distinguish live offers, temporary operational data, approved historical observations, and rebuildable aggregates.

01

What observed fare means

An observed fare is a normalized provider result that FareSearchEngine was permitted to retain at a specific time. It does not represent every fare in the market and is never presented as current inventory.

02

How history is collected

Only eligible, nonpersonal, license-approved observations are included in analytics. Repeated identical offers are combined within a configurable window to reduce refresh bias.

03

Comparable fares

Initial comparisons require compatible direction, cabin, trip type, booking window, trip length, season, stop category, and currency. Broader fallbacks reduce confidence.

04

Percentiles

A percentile describes where a fare sits within comparable observed history. It does not predict whether prices will rise or fall.

05

True Fare

Known mandatory charges, baggage fees, seat fees, and payment fees are separated from estimates and unknown costs. An unknown cost is never treated as zero.

06

Fare Intelligence Score

Version 1 is a deterministic weighted model covering historical price, True Fare, itinerary, flexibility, baggage, connections, and confidence. Missing components are excluded and confidence is reduced.

07

Book or Watch

Recommendations use conservative, structured rules rather than price predictions. Insufficient or stale history produces no recommendation.

08

Confidence

Confidence reflects sample size, distinct days, freshness, provider coverage, specificity, completeness, and stability.

09

Why fares change

Airline inventory is dynamic. Live offers expire and must be revalidated; historical observations cannot be booked.

10

Limitations

Observed history reflects FareSearchEngine’s provider coverage and user searches. Sparse routes, seasonal gaps, and provider bias remain important limitations.