The Schedule Opened. The Front Cabin Did Not.

In the past week American and United put twenty new international routes for 2027 on sale, and United calls its own share of that the largest international expansion in its history. Both went on sale with every start date already inside the award booking window, which almost never happens. I swept all twenty against live award data on the evening American's went on sale, and then swept them again eighteen hours later. The two runs do not agree, and the disagreement may be the most useful thing in this issue.

A route announcement gets written up as a destination story. Vienna, Porto, Ibiza, Okinawa. The line that matters to anyone booking with miles is the one at the bottom about when tickets go on sale, because a schedule loading is the moment award inventory can first appear.

Can first appear. That is the whole of this issue. The schedule opening and the award opening are two different events, run by different departments on different timetables, and this week is an unusually clean chance to measure the gap between them.

What I did

Twenty new routes plus American's reintroduced fourth daily flight to London, so twenty one in total. For each one I searched from its first published operating day out to July 29, 2027, where the 331 day booking window ends today, and resolved every result down to the airline actually operating the nonstop segment. The question is not whether you can reach Barcelona. It is whether the new route loaded award space.

Then I did it again the next afternoon, which turned out to matter more than the first run did.

Award inventory is a process, not a state

On the first sweep, four of American's seven new routes did not exist in the data at all. Porto, Vienna, Barcelona, and Reykjavik returned nothing in any window I tried, including near term dates where connecting itineraries certainly exist. That is what a search looks like before the plumbing has caught up with the announcement.

Eighteen hours later all four were there, and the routes that had already been visible had gained space.

Two different things are mixed in that table. The four that went from nothing to something are mostly the aggregator building routes it had never needed before, so I cannot tell you American loaded that space overnight. Nice is the honest one. It was visible in both runs and went from five days of economy space to thirty five, which is inventory arriving on a route the search could already see, inside a single day.

Which is the practical lesson to carry into the numbers below. If you checked one of these routes on the on sale date and found nothing, you did not learn that there is nothing. You learned what had loaded by the hour you looked.

American: economy nearly everywhere, the front cabin almost nowhere

Here is the second sweep in full. Seven new routes, 679 route days between them.

348 days with economy award space. Two days with business. Zero with first.

Take the Tokyo line for a moment. Economy nonstop on 132 of 133 days, through AAdvantage at 35,000 and cheaper still through Qatar Airways Privilege Club, Alaska Mileage Plan, or Qantas Frequent Flyer. On a route American has not flown in about seven years, returning on a 787-9, the back of the aircraft is effectively wide open the week the schedule loads. The front of the aircraft has nothing at all.

That combination is the finding. An airline that had loaded no award inventory would show zeros in both cabins. This is a schedule that loaded generously in one cabin and held the other back.

The two seats that exist, and what they cost

I am trying to be exact rather than tidy here, because two is not zero. Business class appeared on Philadelphia to Porto, on April 5 and July 12, through AAdvantage, carrying $5.60 in taxes. They priced at 146,000 and 143,500 miles.

Set that against the control below, where American business on routes it already flies starts at 57,500. These are not saver seats that slipped out early. They are priced at roughly two and a half times the going rate, which is what an airline charges when it would rather not sell you the seat yet.

Premium economy tells the same story once you strip out the noise. Charlotte to Barcelona is the only new route with genuine premium economy, on seven days through Alaska Mileage Plan at 35,000. The premium economy that shows on Amsterdam and Nice is a single program pricing at 523,000 and 573,000 miles, which is not availability in any sense a reader should trust.

Why that near zero is a finding and not a glitch

A zero is the easiest number in this work to get wrong. It looks like a discovery and it is usually a plumbing failure, so before writing any of the above I ran the same search, in the same window, through the same pipe, against five routes American already flies.

169 days of American business class award space in summer 2027, all through AAdvantage, all on routes that already exist. The pipe works. Summer 2027 business inventory is visible where there is business inventory to see. It is simply not on the new routes.

The sweep took twenty minutes. Working out which half of it I was allowed to believe took the rest of the night, and that is probably the half that matters.

The one route with real business space is not a new route

New York JFK to London Heathrow returned business on 52 days from 57,500 miles, and first on 24 days from 107,000. It is the only route in the sweep with a front cabin worth booking.

American lists that flight separately from the seven, and correctly, because it is not one of them. Its own release calls it the reintroduction of a fourth daily flight on a route American already serves. Nothing in the data attributes that space to the new frequency rather than to the flights already operating, so I am not claiming it does.

One caution on that route, and on the three London routes in the control table above. The mileage price is only half of what a London award costs. The cheapest business seat in my search carried $5.60 in taxes, but others on the same route came back at $733.50, and the reason is where the flight departs rather than what it costs in miles. The United Kingdom levies its passenger duty on departures from the country, at a materially higher rate in premium cabins, so the westbound half of a round trip is where the cash lands. The tax is also scheduled to rise again on April 1, 2027, which falls inside the operating window of every one of these routes.

The practical version: a one way to London on miles is cheap in cash, and coming home from London on miles is not. Booking the return on a different program, or routing home through another European city, is worth pricing before committing. If London is a connection rather than a destination, a layover under twenty four hours generally avoids the charge altogether, though that is a rule with conditions and worth confirming against your specific itinerary rather than taken on faith.

So the pattern holds. Every genuinely new route came back with economy, a sliver of premium economy on one of them, and two expensive business seats across all seven. The only real front cabin space in the sweep sits on the route that was already there.

Where I cannot tell you anything

United business class cannot be assessed from this data, so I am leaving it alone. The control that cleared the American finding failed for United, and it failed hard enough to be worth showing.

Across twelve established United long haul routes in the same window, including Newark to London, Newark to Frankfurt, San Francisco to Tokyo, and Washington Dulles to Paris, nonstop United business award space returned zero. Not scarce. None, on any of them. On those same twelve city pairs the search returned 1,674 nonstop business records operated by other airlines, British Airways, Lufthansa, Air France, Swiss, Virgin Atlantic, Japan Airlines, and American among them.

So the search sees business class on those routes perfectly well. It simply never returns United's own metal. When a search shows you no business class on Newark to Frankfurt in peak summer, the search is what is broken, not Newark to Frankfurt. Had I skipped that step I would have published a clean, symmetrical, and completely false claim about both airlines. Half of it was true and I could not have told you which half.

What United did show

Economy, unevenly, with the same caveat that this is what a search surfaced rather than what exists.

Nonstop United economy loaded on five of the ten new cities: Luxembourg on 24 days from 30,000 miles, Okinawa on 16 days from 37,000, Toulouse on 14 days from 25,000, and then it thins badly, with Ljubljana on two days and Marseille on one. The other five, Olbia, Catania, Ibiza, Valencia, and Terceira, surfaced no nonstop United economy award space at all. Those five are indexed in the data, unlike American's four on the first sweep, so that is a real absence rather than a blind spot. United's new routes are also subject to government approval, which its own release states.

What to do with this

Monday's argument was that a route going on sale is an award availability event, and that the on sale date deserves action. I still think that is right. This week sharpened what the action is.

If your 2027 trip is in economy, this is the good news the announcements did not spell out. Chicago to Tokyo nonstop with space on 132 of the route's first 133 days is an unusual thing to be able to buy six months ahead, and it is available precisely because nobody has cleared an upgrade into that cabin yet or watched demand build. That space is the most perishable thing in this issue.

If your trip is in business, stop refreshing and start scheduling. Two seats across 679 route days, both priced like the airline would rather keep them, is not scarcity you can beat by looking harder. A schedule that loads with the front cabin held back is telling you the inventory decision has not been made, not that it has gone against you. New long haul routes are where an airline is least certain of demand, and that is exactly where it waits. Check in a few weeks, and again after the first booking curve builds.

The eighteen hour lesson is probably the cheapest edge in here. One look on announcement day is not a search. It is a timestamp.

What I would not do is move a balance this week on the strength of an announcement. Transfers into most of these programs are final, and a speculative transfer made against a business seat that is not there protects nothing. The discipline is the one this letter keeps coming back to: a deadline is a reason to finish a decision you had already reached, not a reason to start one.

A route going on sale tells you the airplane exists. It does not tell you the seat is for sale in miles. On the new routes I could actually measure this week, the gap between those two sentences was the entire front of the aircraft.