There’s an old theory which says it is important to start checking for award seats at midnight 330 days out because that is when schedules are loaded into the system and the seats first become available. The early bird gets the worm. Or, in this case, the early booker gets the award seat. And that was commonly accepted as the gospel for some time, right up until the airlines started getting smarter about revenue management and award seat inventory.
Once the airlines started aggressively managing their revenue inventory they also were able to better control award seats. That manipulating and tweaking of their systems made it a bit more challenging to find award seats in many cases. No longer could a consumer bet on the flights being opened for award booking at 330 days out. It was harder to figure out the patterns. Unless, of course, there was a larger volume of data to analyze. With lots of queries and lots of information some patterns start to emerge.
Take, for instance, Asiana’s flight from JFK to Seoul. Just how often are first class seats released for booking 300+ days out? Turns out the answer is nearly always.
There are a couple obvious “blackout periods” in the availability as shown in the two vertical columns of white. The few weeks leading up to Christmas is a no-go period as is the six weeks between mid-May and early July. But other than that it seems that getting a first class award seat on Asiana out of New York City isn’t all that hard.
But what if you didn’t book 300+ days out? Can you still find award seats? And what about those “blackout” windows? Turns out that they aren’t quite absolute.
The holiday periods mostly remain unbookable, but there are sporadic times where the inventory shows up. And within 30ish days of departure it seems that inventory starts to appear again, though not 100% reliably.
And, just for fun, I checked the data against the Los Angeles – Seoul route as well. The chart looks quite similar.
Certainly no guarantees there, but the numbers are strongly favorable, assuming you are able to book in advance.
Moral of the story: If you are confident in your plans it is a good idea to start looking as early as possible. But that’s not the only time the seats might show up.
n.b. I’m pretty sure the gash in the middle of that block of blue is a failure of the queries, not a lack of award space, though I suppose either is possible.
Never miss another post: Sign up for email alerts and get only the content you want direct to your inbox.
Fascinating! Thank you for showing this.
Do you have other data we could see?
I have a lot more data, Jake. It takes some time to parse it and put together the trend info, but I’ll be releasing more stories like this going forward.
Fascinating data for sure – I’d pay to have access to data like this 🙂 (e.g. the ability to do queries on it).
Wow, very cool data.
Cool stuff! Do you download this data from ExpertFlyer or some other paid source? This is really impressive – not just the data collection, but also the parsing and the analysis
The data is aggregated from publicly available sources, not paid/subscription systems.
Please do! I would be VERY interested in AA as I alao run scripts but couldn’t identify a reliable pattern there when it comes to black outs
Of course, I’m assuming you’re talking about availability open to Star partners. OZ classifies the large chunks you mention as peak-season dates (generally around 16MAY-6JUL and 6DEC-23DEC) for flights departing the USA, where these dates are set as 50% extra mileage for members of OZ’s FFP and generally unavailable to other partners. Searching JFK-ICN on OZ’s website, for example, I find availability for almost all of June 2015 at the extra mileage cost and many of the dates blocked off in December 2014.
Nonetheless, interesting post. You’ve been prepping it for nearly a year? Impressive.
If you’re willing to pay enough points then anything is possible, I suppose. But, yes, I’m looking at lowest tier redemption levels in all this data. And in the case of OZ redemptions it is partner inventory.
What is the diagonal white stripe that stretches from up and to the right from (x,y)=(early April 2014, 0 days) to (early March 2015, 337 days)? Is that lack of availability significant, or is it simply missing data? (I would guess the latter.) Maybe for future plots it would be useful to distinguish between the “missing data case” and the “lack of availability” case.
I did distinguish about that. Read all the way to the end. 🙂
“n.b. I’m pretty sure the gash in the middle of that block of blue is a failure of the queries, not a lack of award space, though I suppose either is possible.”
OK, thanks, didn’t see that. Usually it’s stuff like credit card disclaimers in italics at the end of blog posts, and for my own sanity I have trained my eyes to automatically skip such stuff…
I found this example particularly interesting! We have 2 seats in F on Asiana from JFK to ICN. We booked the seats in mid-November-ish 2013 for departure at the end of Sept. 2014. Very excited!
Great post, thank you for this information, look forward to more posts like this!.
Cheers,
PedroNY
I’ve been looking at United’s availability (for my own travel, so not quite this extensively) on a couple of routes and see inventory show up shortly before travel date (<30 days) during peak seasons. I guess I am too young to rely on this myth here. I do wonder how more people being educated about it will effect dynamic availability in the future.
Seth – this data is awesome. Can’t wait for more posts like this!
awesome info as usual.