Airline offer management that shows every bundle on every channel's shelf before you say yes
$1.50 more per passenger from bundles, each one checked in every channel first.
See one case, screen by screen ↓Airline offer management is how an airline's commercial team builds, prices and governs the offers and bundles it sells — fare family merchandising, ancillaries and dynamic bundles — from the fare-family shelf to NDC and GDS agencies. Airline Offer Management is an offer management system run by eight agents: they propose bundles from counted shopping and sales, simulate take-up and buy-down, show the shelf as a shopper sees it in every channel and check the disclosure and refund rules. The Head of Offer Management approves each offer channel by channel, with a reason code.
How to choose AI for airlines: a buyer's guide to agentic AI solutions →
Updated
In airline offer management today, fares are sold through filed fare families while seats, bags and changes are sold one by one, priced by separate teams in separate systems. A new bundle starts as a pattern in a monthly report. An analyst sizes it in a spreadsheet with a guessed take-up, and nobody checks the buy-down it causes from the higher fare family. Then it is built in the airline merchandising tool, reviewed for full-fare advertising, baggage fee disclosure and ancillary fee refunds, and launched in every channel at once.
The trouble shows up after launch. Dynamic offers and bundles are not filed branded fares, so a GDS agency terminal sells a dynamic bundle as plain Main: the ticket says 0 bags, the traveler pays at the airport and is owed a refund. Results are read weeks later, often without a control group, and a bundle that stops selling lingers for years because retiring it means servicing orders already sold.
Estimated split for one new bundle on one market, by hand and with the solution. Totals are about 40 working days (6–10 weeks) against the 5-day target.
Eight agents propose bundles, simulate take-up, show every channel's shelf and keep the rules; people approve.
Corvane Airways flies Chicago–Las Vegas, where a rival sells a seat and a bag together for $71. Overnight the agents propose a Weekender on Main. Here is how the Head of Offer Management works with them, screen by screen, in the working solution.
Renee's Today screen opens with what the agents did overnight: 412,600 shopping sessions read and all 12 live and draft bundles checked against 11 rules. Eight decisions need her, est. +$329,300 a season on those ready to sign. First is the Weekender on ORD–LAS, due today, because GDS agencies would lose the bag. The rival shelf shop is 30 hours old, so the agents did not use it for prices.
Bundle designer: “Start with the Weekender on Chicago–Las Vegas: it is due today, and the GDS view loses the checked bag.”
On Main, a preferred seat, the first checked bag and one same-day change for $64, against $132 à la carte, on ORD–LAS Thursday, Friday and Sunday flights from Dec 3 to Mar 28: 136 flight-dates. Last winter 23% of Main passengers here bought a seat and a bag separately, and 38% of Main shoppers opened the bag fee page before booking. Each reason links to its source.
The shelf shows the fare families for CV 1427, Fri Dec 4, as a shopper sees them on Corvane.com, in the app, at an NDC agency and on a GDS agency terminal.
The simulator projects 9.0% attach (range 7.1%–10.6%), 1,371 Weekenders sold and est. +$47,300 for the season after the à la carte sales it replaces: $3.10 more per Main passenger. The Checker re-runs it from the raw data without seeing that result: 8.4% attach and est. +$43,000, a 9% gap inside the 15% tolerance. It sees more buy-down from Main Flex, 2.1% against 1.4%, because Main plus the bundle is $228 against $239 for Main Flex.
Both buy-down figures are under the 3% stop rule, so the test watches it.
Corvane.com, the app and NDC agencies show all three items; NDC agencies show them as text only. On a GDS agency terminal the Weekender is not a filed branded fare, so it sells as Main with the seat and the change as optional services, and the ticket's bag allowance shows 0. The traveler would be charged $40 at the airport and be owed a refund under Part 260.
Channel renderer: “Policy D-4 withholds an offer from a channel that cannot show all of it. I recommend web, app and NDC only.”
Full fare with taxes on every channel (14 CFR 399.84) and bag fees disclosed beside the shelf (399.85). Each item carries a value so a service not provided is refunded pro rata under Part 260: seat $11, bag $17, change $36. Gold members already have a free bag, so they see the Weekender without it at $47. In the offer editor the price sits at 48% of à la carte, inside rule R10's 40–85%.
The Rules keeper re-checks every save in the offer editor.
The Take-up simulator answers with its sources: revenue for the season peaks at $64. At $59 more people buy, 10.2%, but the season earns est. +$44,900 and buy-down rises to 2.0%. Above $64 shoppers compare with the seat and bag alone at $57, and at $69 it earns est. +$40,600. Renee says “Try it at $59”, sees the shelf re-run, and goes back to $64.
Take-up simulator: “Re-ran it at $59: 10.2% attach, 1,554 sold, est. +$44,900 for the season (−$2,400 against $64).”
The decision dialog lists each channel: Corvane.com, the app and NDC agencies ready; GDS agencies withheld because they lose the checked bag. Renee picks a 15% test cell and reason code CH, “Channel cannot show it all”, and confirms she has read the agents' reasons and sources. Under $150,000 a season, no co-sign is needed from Victor Hale, the CCO.
“I have read the agents' reasons and sources. This is signed under my name, Renee Calloway, and logged.”
The shelf now shows the Weekender approved for web, app and NDC, and withheld from GDS agencies. A filed version can be requested in one click; it goes to Marcus Bell, the Distribution Manager, to file a Weekender branded fare, due Nov 2. The evidence pack holds the shopping and sales passages, the simulation, the Checker's re-simulation, the four channel renders, the rule checks, the test plan and the decision.
T-55 opens Monday Oct 19 at 06:00 for 15% of shopping sessions in the web, app and NDC channels, assigned by session, never by person, for 28 days. Control keeps today's shelf. The Test designer checks three stop rules every morning: attach under 4% after 7 days, Main Flex buy-down above 3%, complaints above 2 per 1,000 sold.
Basic Plus has sold at 1.1% attach for 90 days across 7 markets, under the 2% floor. The Results reviewer plans the stop channel by channel: Corvane.com and the app on Oct 19, a 14-day notice to 41 NDC agency contacts, NDC agencies on Nov 2. All 214 sold orders are serviced as sold through Feb 26, 2027, and a service not provided is refunded at its value.
Results reviewer: “Nothing changes until you approve.”
The same Weekender, seen by the five people who carry it — what their weeks looked like, and what they look like now.
Eight agents propose bundles, simulate take-up, show every channel's shelf and keep the rules; people approve.
Reads counted shopping sessions, bookings, ancillary sales and rival shelves from public booking paths, and links each signal to a market.
Proposes bundles — items, name, price, markets and channels — and writes why, with numbered sources a person can open.
Projects attach, buy-down from the higher fare family and net revenue across a price range, after the à la carte sales the bundle replaces.
Re-simulates every proposal from the raw data with its own model, without seeing the first result, and compares the two.
Renders the fare-family shelf as a shopper sees it on the website, the app, NDC agencies and GDS agency terminals, and lists what each channel loses.
Checks every offer against 11 eligibility, disclosure and refund rules, including 14 CFR 399.84, 399.85, Part 260 and Part 382, and blocks a breach with the rule and the route forward.
Writes the test cell, control and stop rules for every approved offer, and checks running tests every morning.
Reads test against control by market, drafts the weekly report, and proposes retirements with a stop date per channel, agency notice and servicing of sold orders.
Edit, approve per channel with a reason code; the CCO co-signs above $150,000. The agents propose; a named person decides.
Ask about an offer, a channel or a rule, or tell the agents what to try. Every answer cites its sources, and you decide.
Why $64 and not $59?
Revenue for the season peaks at $64. At $59 more people buy (10.2%), but the season earns est. +$44,900, and buy-down rises to 2.0%. Above $64 shoppers compare with the seat and bag alone at $57: at $69 it earns est. +$40,600.
What does the GDS lose on the Weekender?
The checked bag. The Weekender is not a filed fare, so an agency terminal sells it as Main with the seat and change as services, and the ticket shows 0 bags. Policy D-4 withholds it from GDS agencies until a filed version exists.
Which bundles should we retire?
Basic Plus: 1.1% attach over 90 days on 7 markets, under the 2% floor. The plan stops it on Oct 19 on web and app and Nov 2 in NDC agencies, and keeps all 214 sold orders.
What is due today?
The Weekender on ORD–LAS by 17:00. Long-haul Comfort by Friday 12:00, then Victor Hale co-signs. The Family Pack is blocked by the family seating rule, and four low-risk changes can go in one sign-off.
13 screens from the working solution, on its sample data. Pick one to see it large.
The limits the agents work inside are settings. Saved threshold changes are logged and apply from the next agent run.
| Setting | Default | Choose from |
|---|---|---|
| CCO co-signs above | $150,000 a season | Any amount |
| Checker tolerance | 15% | Any percentage; a larger gap goes to a person with both workings |
| Main Flex buy-down stop | 3% | Any percentage; also a stop rule in every test |
| Retirement floor | 2% attach over 90 days | Any percentage of attach over 90 days |
| Largest test cell | 20% of sessions | Any percentage; each offer picks 5%, 10%, 15% or 20% of sessions |
| Longest test | 42 days | Any number of days |
| Stop using a feed older than | 24 hours | Any number of hours |
| Escalate to the CCO after | 24 hours overdue | Any number of hours; a reminder before each decide-by time goes by email and chat, by email, or in the app only |
Every figure is labelled: a target the solution is built to, an estimate, a typical published result, or a proven one.
“demo” = seen in the working solution, on its sample data · “target” = the design goal, measured in the pilot · “estimated” = our estimate · “typical” = published figures (IdeaWorksCompany, 2025: $157B global ancillary revenue, 15.7% of airline revenue; airline distribution industry research: about 55% of airline revenue through indirect channels). $1.50 × 14.2M leisure-market passengers is about $21M a year, our arithmetic. Regulations named: 14 CFR 399.84, 399.85, Part 260 and Part 382. Corvane Airways, its rival, and every person named on this page are fictional — characters and sample data in the working solution.
Airline offer management is how an airline's commercial team builds, prices and governs the offers and bundles it sells — fare family merchandising, ancillaries and dynamic bundles — from the fare-family shelf to NDC and GDS agencies. Airline Offer Management is an offer management system run by eight agents: they propose bundles from counted shopping and sales, simulate take-up and buy-down, show the shelf as a shopper sees it in every channel and check the disclosure and refund rules. The Head of Offer Management approves each offer channel by channel, with a reason code.
Dynamic bundling only pays if the bundle does not pull buyers down from the next fare family. The Take-up simulator projects attach, net revenue and buy-down from the higher fare family across a price range, and the Checker re-runs it with its own model. In the demo the Weekender peaks at $64; at $59 more people buy but the season earns less and buy-down rises. A buy-down stop, 3% by default, is a company rule and a stop rule in every test.
A dynamic bundle is not a filed branded fare, so a GDS agency terminal can sell the trip as the plain fare with some items as optional services. The Channel renderer shows exactly what each channel loses — in the demo, the ticket's bag allowance shows 0 — and under channel parity the offer is withheld from that channel until a filed version exists.
Yes. For NDC offer management, NDC agencies are one of the four channels on the shelf, rendered through the NDC 21.3 schema alongside the website, the app and GDS agencies, and an offer can be approved for NDC agencies on its own.
The Rules keeper checks every offer for the full price with taxes (14 CFR 399.84) and bag fees disclosed beside the shelf (14 CFR 399.85). Each bundle item carries a value on file, so a service not provided is refunded at its value under Part 260. These rules are locked; only Consumer Protection Counsel can change them.
Yes. The agents propose, simulate, check and plan; they never publish, file, retire or refund. Nothing reaches a channel without a person's signed yes for that channel and a reason code, the Chief Commercial Officer co-signs above $150,000 a season, and every approved offer starts in a test cell with stop rules.
Yes. The Results reviewer proposes retiring a bundle under the attach floor — 2% over 90 days by default — with a stop date per channel, a 14-day notice to NDC agencies and a plan to service every sold order as sold. It suggests; a person approves the retirement.
The Agentic Solution Engine builds and deploys it from your requirements and documents — your fare families, distribution policy, customer service plan and a sample of shopping and sales — and it goes live once every quality gate has passed. We will walk you through it on your own markets first.
Every memo carries its proof, and every dispute gets a cited reply.
1-min demoAgency Incentive ManagementEvery agency paid exactly what its contract earns, with each excluded sale explained.
1-min demoAirline Group Sales ManagementGroup quotes the same day, priced on the seats they displace, with every deadline watched.
1-min demoWe'll run Airline Offer Management on a sample of your own markets, fare families and channels.