You probably arrived looking for a number: the figure that says whether your ancillary performance is good. This page does not have one, and the reason is not modesty. Almost every ancillary benchmark in circulation is a number with its definition stripped off, and a number without its definition is not evidence. It is decoration.
In short
There is no single ancillary revenue benchmark worth quoting. Published figures use different denominators, different category mixes, and different channels, so two of them rarely measure the same thing. Read industry sources for direction and definition, then benchmark yourself against your own baseline, measured the same way every period.
Why do widely quoted ancillary revenue figures become unreliable?
Take any widely repeated ancillary figure and trace it backwards. It usually began in a real study, was quoted in an article, was quoted again from the article, and by the fourth repetition the qualifying sentence has fallen off. The surviving number carries no denominator, no period, and no scope.
That matters because ancillary revenue is not one metric. It is a family of metrics that share a name, and the choices between them move the answer by more than any operational improvement ever will.
What makes two ancillary revenue figures incomparable?
| The choice | Common variants | Effect on the number |
|---|---|---|
| Denominator | Per passenger, per booking, per segment, per departure | A multi sector booking counted per segment produces a much smaller figure than the same booking counted once |
| Category scope | Seats and bags only, or the wider catalogue including insurance, lounges, and meals | Broad scope raises the total and dilutes the per category detail |
| Channel | Direct only, indirect only, or blended | Direct and indirect behave differently enough that a blended figure describes neither |
| Gross or net | Face value, or value after commissions, refunds, and chargebacks | Net figures are the honest ones and are almost never the quoted ones |
| Taxes and fees | Included, excluded, or unstated | Carrier imposed charges and government taxes can dominate a small ancillary sale |
| Haul and market | Short haul, long haul, domestic, international | Willingness to pay differs by hours in the air more than by anything else |
| Period and currency | Financial year, calendar year, a single quarter, one reporting currency | A strong quarter and a moved exchange rate look identical in the output |
Any two figures that differ on even three of these rows are measuring different things. Most published comparisons differ on all seven and do not say so.
Where can you find reliable ancillary revenue figures, and how should you read them?
Genuine, checkable ancillary data does exist. It is less convenient than the numbers on a slide, which is why it circulates less.
Airline annual reports and investor presentations
The most reliable source, because the figures carry legal weight. Find the segment note rather than the highlights page and read the definition in its footnote. Many carriers fold ancillary into other operating income, which tells you the figure you wanted was never disclosed.
Regulatory filings
Listed carriers file into their home market regulator, and some aviation regulators collect baggage and change fee reporting directly. Drier and far more precise than anything downstream, usually free to read, and narrow in scope, so treat them as a floor rather than a full picture.
Industry associations and specialist consultancies
Several publish annual ancillary studies covering many carriers on a consistent methodology, which is exactly what makes them worth reading. Go to the source document and read the method section first. If the method is not published, treat the headline as an opinion.
The trade press
Useful for finding where a figure originated, and unreliable as the figure itself. Follow every stated number back to its primary document before repeating it. If you cannot find the primary document, that is your answer about the number.
One habit is worth more than the whole list: read the definition before the result. If you cannot describe how a figure was calculated after reading its source, you have learned something useful about the figure.
What a properly published figure looks like
The discipline above is easier to picture with a real example. The series below is from the IdeaWorksCompany press release of 29 October 2024, published with CarTrawler, and it demonstrates every habit this page argues for: the publisher is named, the date is fixed, and the definition travels with the number. These are a la carte fees only, the part travellers pay directly for bags, seats and buy on board, expressed per passenger and adjusted into 2024 dollars. Each of those clauses changes what the figure means, which is exactly the point.
| Category | A la carte revenue per passenger |
|---|---|
| 2015 | $13.74 |
| 2016 | $15.54 |
| 2017 | $18.29 |
| 2018 | $19.08 |
| 2019 | $20.67 |
| 2020 | $21.61 |
| 2021 | $21.69 |
| 2022 | $28.28 |
| 2023 | $24.87 |
| 2024 | $24.97 |
Notice what the definition rules out. This is not total ancillary revenue per passenger, because it excludes loyalty programmes and commissions. It is not comparable to a single airline's disclosure, because it is a global estimate across every model of carrier. Read this way, one well published series teaches more about method than any number of averages traded on slides.
Build a baseline you can defend
The benchmark that changes decisions is your own, measured consistently. It is also the only one you can fully see the inside of.
Once that definition is fixed, the worked ancillary revenue arithmetic shows how to carry your own volume, denominator, attach assumption, and value through the sum without borrowing an external result.
Then hold four disciplines.
Segment before you aggregate. A single blended figure across long haul and short haul, direct and indirect, entry brands and flexible brands, hides everything worth acting on. The interesting comparisons are internal: this route family against that one, this fare brand against that one.
Fix the denominator and leave it alone. Changing from per booking to per passenger midway through a year does not improve the metric. It ends the time series.
Recompute history whenever the definition changes. Change it backwards as well, and keep a dated note of what changed and why. A metric without a change log becomes unusable within about two people.
Track the shape, not the level. Direction over several periods carries real information. A single period against an external figure carries almost none.
Two other pages shape how you read your own numbers. What the fare already included is covered in unbundling and branded fares, and whether the price was fixed or constructed is covered in dynamic offer pricing.
Frequently asked questions
What is a good ancillary revenue per passenger figure?
The question cannot be answered without a definition. Per passenger over which passengers, counting which categories, gross or net, and in which channel? Two answers built on different choices are not comparable, so the target has to come from your own history.
Where can I find real ancillary revenue figures?
Airline annual reports and investor presentations, regulatory filings by listed carriers, the annual studies published by industry associations and specialist consultancies, and the trade press. Always read the primary document rather than the article summarising it.
Why do published ancillary benchmarks vary so widely?
Because the denominator, the category mix, the channel, and the treatment of taxes and commissions all vary between publishers, and most figures do not state which choices were made.
Should an agency benchmark itself against airlines?
Only for direction. An airline counts inventory it owns and prices itself. An intermediary counts commissions and margin on someone else's inventory, so the two figures describe different businesses.
How long before an internal baseline is useful?
Long enough to cover the seasonality of your own booking mix. Until you have seen a full cycle, a movement in the number is as likely to be the calendar as it is to be anything you did.