Delta Air Lines
AI dynamic pricing (yield management)
In 2025, Delta Air Lines sets about 3% of its domestic fares with Fetcherr's AI engine and targets 20% by the end of 2025, according to its President on the earnings call.
Key points
- Fare-class grids replaced by AI-computed prices, route by route.
- Fetcherr pricing engine (large market model) integrated into revenue management.
- About 3% of domestic fares set by AI in Q2 2025, target 20% by end of 2025.
- Evidence level A, status mixed signals.
Objective
Set each fare at the level the market accepts, route by route, instead of relying on fixed fare classes, to capture more revenue per seat.
The deployment
Delta is progressively replacing its fare-class grids with prices computed by Fetcherr's engine, described as a large market model. The system ingests consumer searches, competitor prices, and demand signals such as the weather, then sets the fare. In the second quarter of 2025, about 3 percent of domestic network fares were set by AI, up from 1 percent at the end of 2024, with an announced target of 20 percent by the end of 2025. Delta says it is still in a test and ramp-up phase.
Results Proof A
Figures and framing stated by Delta's President on the second-quarter 2025 earnings call, corroborated by several established press outlets that report the same values (3% current, 20% targeted).
How it works
Inferred typical approachThe internal detail is not public. Here is a proven approach that leads to the same result, to adapt to your stack.
The stack in detail
- plateforme Fetcherr AI pricing engine (large market model) that ingests searches, competitor prices, and demand signals, then sets the price per route
- infra Systemes de revenue management Delta in-house systems the engine is integrated into and that keep supervision of the AI-managed scope
- infra Canaux de distribution Delta (site, app, GDS) surfaces where the computed fare is published and sold
How it runs, concretely
For ops teams-
1Ingesting market data AI
Consumer searches, competitor prices, and demand signals such as the weather are aggregated continuously.
-
2Demand prediction AI
The large market model estimates demand and willingness to pay per route and date.
-
3Setting the fare AI
The engine sets a price per route, replacing the fixed fare classes.
-
4Supervision and ramp-up revenue management team
The revenue management team progressively extends the scope of AI-managed fares and monitors the results.
The demand predicted per route and date, cross-referenced with competitor prices and consumer searches. Without a reliable market data flow, the proposed price drifts and unit revenue degrades.
How your customers perceive this type of use
Sourced studiesLe pricing algorithmique est le terrain le plus inflammable : 68% des consommateurs disent se sentir leses quand les marques utilisent le pricing dynamique et 80% jugent plus dignes de confiance les marques aux prix constants (Gartner, 2024). L'equite percue varie selon le secteur : le pricing dynamique n'est juge juste que par 33% a 40% des repondants selon qu'il s'agit de concerts ou de cinemas (YouGov, 17 marches). Le prix personnalise par les donnees individuelles est le plus rejete : 47% des Americains s'y opposent fermement (Consumer Reports, 2024).
Acceptance conditions
- La constance des prix comme signal de confiance : 80% jugent plus fiables les marques aux prix stables (Gartner 2024)
- Le secteur conditionne l'equite percue : le pricing dynamique est mieux tolere pour les cinemas (40% le jugent juste) que pour les concerts (33%) (YouGov 2024)
Red lines
- Le pricing dynamique percu comme abus : 68% se sentent leses (Gartner 2024)
- Le prix individualise a partir des donnees personnelles : 47% d'opposition ferme (Consumer Reports 2024)
- Les frais caches et hausses imprevues, vecus par 79% des consommateurs sur un an et associes a la perte de confiance (Gartner 2024)
Sources: Gartner 2024 · YouGov 2024 · Consumer Reports 2024
How to replicate
Inference, not sourcedData prerequisites
- booking and fare history
- real-time competitor prices
- demand signals (events, weather, searches)
Org prerequisites
- a mature revenue management team
- integration of the pricing engine into the distribution channels
- governance on price transparency and fairness
Possible stack
- vendor AI pricing engine (Fetcherr type)
- in-house demand models
- existing revenue management system
The plan, step by step
- Step 1Audit the data (booking and fare history, competitor prices, demand signals) and isolate a pilot scope of routes.Deliverable: Pilot scope defined and data flows wired
- Step 2Integrate the pricing engine into the revenue management and distribution systems.Deliverable: AI-computed prices available in pre-production
- Step 3Run the engine in shadow mode: compare its prices to the current grid without showing them.Deliverable: Gap report and estimate of incremental revenue
- Step 4Switch a small percentage of fares into production with safeguards (floors, ceilings, human supervision).Deliverable: 1-3% of fares managed by AI with unit revenue monitoring
- Step 5Ramp up progressively and set the governance on price transparency and fairness.Deliverable: Extended scope with an oversight committee and reporting
First step: Isolate a subset of domestic routes and test the pricing engine there against the current fare grid.
The same pattern, proven in other industries
full-funnel AI ad suite (automated ROAS bidding, new-customer prioritization, budgets and targeting managed by the algorithm)
ML-optimized ad platform (automated campaigns + ranking + genAI formats)
Shopee
AI automation of seller ad campaigns
Sources
- S1 Delta starting to scale its AI-driven dynamic pricing system Secondary archive pending
- S2 Delta Is Using AI to Determine Some Ticket Prices. What Does That Mean for Travelers? Established press archive pending
- S3 Senators Demand Answers as Delta Plans to Price More Tickets Based on What AI Thinks You'll Pay Established press archive pending
An error, newer info, a source?
This page lives on its accuracy. If a figure has moved, if the deployment has changed, or if you have a higher-quality source, tell us. Every sourced correction is verified before publication.