Strange married segment experience
#1
Original Poster
Join Date: Dec 1999
Location: NorCal - 1K 2MM
Posts: 2,089
Strange married segment experience
Buying SJC-IAH-BWI. Inexplicably for years this flight has operated with a through flight number, but change of equipment and gate at IAH. Found a cheap L fare, but wanted to book as two separate segments to maximize the chance of a CPU on one leg if both not available. Multi-city booking offered me only a more expensive T fare, not the cheaper L fare offered in the through flight. Checked individual legs, and sure enough, segment one was L=4, but segment 2 L=0, even thought a through L fare was offered when booked without multi-segment. OK, I realize, married segment issue.
On a whim before buying the higher two segment fare, I called res and asked them to try to book it. Agent says, sorry, L only available on segment one, that's why you're getting the higher T fare. Same as I deduced. Some small voice in my head said "well, push a little farther". So, I asked, "please book it as two segments anyway, and see what it fares to." Surprise -- it fares to the cheap L fare, giving L on both segments, even though both res and I see L=0 on second segment.
What's happening here??
On a whim before buying the higher two segment fare, I called res and asked them to try to book it. Agent says, sorry, L only available on segment one, that's why you're getting the higher T fare. Same as I deduced. Some small voice in my head said "well, push a little farther". So, I asked, "please book it as two segments anyway, and see what it fares to." Surprise -- it fares to the cheap L fare, giving L on both segments, even though both res and I see L=0 on second segment.
What's happening here??
#2
Join Date: Jan 2009
Location: LHR (sometimes CLE, SFO, BOS, LAX, SEA)
Programs: UA 1K
Posts: 5,893
This hurts my brain but it sounds like what you are saying is that there's a "direct" flight UA1851 SJC-IAH-BWI. And when you look at availability:
(1) Availability shown as a solution for a one-way journey SJC-BWI:
UA1851 - SJC-BWI - has inventory F9 T9 L0
(2) Availability shown as a solution for a one-way journey SJC-IAH:
UA1851 - SJC-IAH - has inventory F9 T9 L4
(3) Availability shown as a solution for a one-way journey IAH-BWI:
UA1851 - IAH-BWI - has inventory F9 T9 L0
(4) Availability shown as a solution for a one-way journey SJC-BWI:
UA1851 - SJC-IAH - has inventory at least L1 (let's say it's F9 T9 L4)
UA1851 - IAH-BWI - has inventory at least L1 (let's say it's F9 T9 L4).
The situation described in #4 is totally reasonable (sort of) -- a flight can have "more" inventory when it's part of a connection to a longer journey than when it's searched nonstop. Classic example is UA79 NRT-ICN which in my experience has much better inventory if you are connecting onward like NRT-ICN-SFO vs. as the nonstop.
You know, on most flights, there is an earlier flight SJC-IAH. I **wonder** whether if you do a search SJC-BWI and take a look at inventory on a fare solution that includes a wide-open early-morning SJC-IAH plus UA1851 IAH-BWI, I wonder if you would see L4 (or whatever) on the second sector. I think probably you would.
This is a pretty cool corner case, congrats. I don't think technically there is any reason why this shouldn't happen. After all, the inventory for "the direct flight" and "the individual flights as part of a particular married-segment journey that happens to be the direct flight's city pairs" are different things, under the hood … I think. But the result is not intuitive.
(1) Availability shown as a solution for a one-way journey SJC-BWI:
UA1851 - SJC-BWI - has inventory F9 T9 L0
(2) Availability shown as a solution for a one-way journey SJC-IAH:
UA1851 - SJC-IAH - has inventory F9 T9 L4
(3) Availability shown as a solution for a one-way journey IAH-BWI:
UA1851 - IAH-BWI - has inventory F9 T9 L0
(4) Availability shown as a solution for a one-way journey SJC-BWI:
UA1851 - SJC-IAH - has inventory at least L1 (let's say it's F9 T9 L4)
UA1851 - IAH-BWI - has inventory at least L1 (let's say it's F9 T9 L4).
The situation described in #4 is totally reasonable (sort of) -- a flight can have "more" inventory when it's part of a connection to a longer journey than when it's searched nonstop. Classic example is UA79 NRT-ICN which in my experience has much better inventory if you are connecting onward like NRT-ICN-SFO vs. as the nonstop.
You know, on most flights, there is an earlier flight SJC-IAH. I **wonder** whether if you do a search SJC-BWI and take a look at inventory on a fare solution that includes a wide-open early-morning SJC-IAH plus UA1851 IAH-BWI, I wonder if you would see L4 (or whatever) on the second sector. I think probably you would.
This is a pretty cool corner case, congrats. I don't think technically there is any reason why this shouldn't happen. After all, the inventory for "the direct flight" and "the individual flights as part of a particular married-segment journey that happens to be the direct flight's city pairs" are different things, under the hood … I think. But the result is not intuitive.
#3
Original Poster
Join Date: Dec 1999
Location: NorCal - 1K 2MM
Posts: 2,089
Thanks for the interesting reply! The situation is that SJC-BWI direct shows L>4, but the individual segments have L>0 segment 1, L=0 segment 2, the classic married segment situation, so creating two segments will not yield the L fare.
Your proposed test is interesting but doesn't work out the way you predicted. If I book the earlier SJC-IAH segment 1, segment two still yields L=0. I can only conclude that the web booking engine is not quite as smart as the agent's engine, and she (inadvertently) forced a through L fare for the two segment trip, even though segment 2 showed L=0 for both her and me. Perhaps the availability function and the booking function are somewhat divorced, so that even though the availability is 0, the booking engine is more forgiving.
The moral of this experience seems to be that if you are foiled by a married segment problem, call and ask them to book and fare it anyway, and see what happens!
Your proposed test is interesting but doesn't work out the way you predicted. If I book the earlier SJC-IAH segment 1, segment two still yields L=0. I can only conclude that the web booking engine is not quite as smart as the agent's engine, and she (inadvertently) forced a through L fare for the two segment trip, even though segment 2 showed L=0 for both her and me. Perhaps the availability function and the booking function are somewhat divorced, so that even though the availability is 0, the booking engine is more forgiving.
The moral of this experience seems to be that if you are foiled by a married segment problem, call and ask them to book and fare it anyway, and see what happens!
Last edited by Starman; Sep 17, 2016 at 11:28 pm
#4
Join Date: Jan 2009
Location: LHR (sometimes CLE, SFO, BOS, LAX, SEA)
Programs: UA 1K
Posts: 5,893
SJC-BWI search: UA1851 SJC-BWI "direct" inventory T9 L4
SJC-IAH search: UA1851 SJC-IAH inventory T9 L4
IAH-BWI search: UA1851 IAH-BWI inventory T9 L0
Multicity search SJC-BWI+BWI-IAH: UA1851 SJC-BWI displays as T9 L0 and also UA1851 BWI-IAH displays at T9 L0.
Agent searches SJC-BWI and somehow "sells" SJC-IAH + IAH-BWI: Somehow you get L.
Wowzers. That makes more sense but makes the result even stranger, yeah!
Your proposed test is interesting but doesn't work out the way you predicted. If I book the earlier SJC-IAH segment 1, segment two still yields L=0. I can only conclude that the web booking engine is not quite as smart as the agent's engine, and she (inadvertently) forced a through L fare for the two segment trip, even though segment 2 showed L=0 for both her and me. Perhaps the availability function and the booking function are somewhat divorced, so that even though the availability is 0, the booking engine is more forgiving.
#5
Join Date: Jul 2007
Location: San Francisco/Sydney
Programs: UA 1K/MM, Hilton Diamond, Marriott Something, IHG Gold, Hertz PC, Avis PC
Posts: 8,157
Most likely what's confusing things here is the fact that the flight has the same flight number for both segments, which is NOT a requirement for a married segment.
UA1851 (SJC-IAH) connecting to UA1851 (IAH-BWI) should be able to access the same married segment inventory as the one-stop UA1851 (SJC-BWI), so what happened for the agent is the correct behavior, whilst what happened on the website is not.
The other option in this case is to book the single flight, and then call and ask for it to be split into two. Might take a few calls to get an agent that knows how to do it, but it can definitely be done.
#6
Join Date: Jun 2013
Programs: UA Global Services, BA Silver, Emirates Gold, Airfrance Silver, Hilton Diamond, Marriott Gold
Posts: 23
The fact that the agent saw L=0 for IAH-BWI is typically a symptom of the fact that when looking at availability for each individual flight segment (but not necessarily booking it as you go), the inventory management systems don't have full visibility of the entire journey.
When the complete journey is booked and priced, all info necessary for the computer to recognize the trip as SJC x/IAH BWI is present and an accurate availability calculation can take place.
Just looking at availability for each distinct segment will reflect local, point-to-point status and it sounds as though on that basis, the IAH-BWI segment alone is not offering L class for sale. If that's the case, the computer is blindly unaware that the request for IAH-BWI is the second segment of an SJC-BWI trip. As far as I know, UA's availability systems will not split a 'direct' flight with a change of gauge into two unique flight segments with the same number, so the agent would have needed to look at each segment independently to achieve what the OP had asked for.
What is strange though is that .com should have the same journey context available to it as part of its multi-city search.
When the complete journey is booked and priced, all info necessary for the computer to recognize the trip as SJC x/IAH BWI is present and an accurate availability calculation can take place.
Just looking at availability for each distinct segment will reflect local, point-to-point status and it sounds as though on that basis, the IAH-BWI segment alone is not offering L class for sale. If that's the case, the computer is blindly unaware that the request for IAH-BWI is the second segment of an SJC-BWI trip. As far as I know, UA's availability systems will not split a 'direct' flight with a change of gauge into two unique flight segments with the same number, so the agent would have needed to look at each segment independently to achieve what the OP had asked for.
What is strange though is that .com should have the same journey context available to it as part of its multi-city search.
#7
Original Poster
Join Date: Dec 1999
Location: NorCal - 1K 2MM
Posts: 2,089
It's gone on for so long on this one flight that I sometimes wonder if this is some weird pilot or FA pay issue, where the company is trying to thwart the union contract. Or possibly something competitive on a mail or freight contract; there are a lot of daily USPS charters out of SJC to the east coast.
#8
FlyerTalk Evangelist
Join Date: Nov 2014
Location: MSP
Programs: DL PM, UA Gold, WN, Global Entry; +others wherever miles/points are found
Posts: 14,414
UA1851 (SJC-IAH) connecting to UA1851 (IAH-BWI) should be able to access the same married segment inventory as the one-stop UA1851 (SJC-BWI), so what happened for the agent is the correct behavior, whilst what happened on the website is not.
The other option in this case is to book the single flight, and then call and ask for it to be split into two. Might take a few calls to get an agent that knows how to do it, but it can definitely be done.
The other option in this case is to book the single flight, and then call and ask for it to be split into two. Might take a few calls to get an agent that knows how to do it, but it can definitely be done.
(a) UA1851 SJC-BWI as a direct flight, with through inventory
(b) UA1851 SJC-IAH; [ghost fare break]; UA1851 IAH-BWI. This will construct using the broken fare inventory even after the segments are reassembled and it is fared though. This is a known bug (or feature ) of the UA website. The final itinerary re-assembles as UA1851 connecting to UA1851 with a through fare but the wrong inventory.
It sounds like the agent is (correctly) able to apply the through inventory to the connecting option. The application in (b) is technically illegal per inventory rules and should not be allowed to price.
#9
Join Date: Jan 2011
Location: San Antonio, TX
Programs: 1K
Posts: 784
Even if it is married and same flight number, the upgrades are processed separately. So, you will be eligible for a CPU on each segment (or that's what has happened to me in this scenario, which was very often SAT-IAH-DFW). So, you're good either way if your experience is consistent with mine.
#10
FlyerTalk Evangelist
Join Date: Oct 2003
Location: Floating around
Programs: UA 1K (1MM), DL Gold (1MM), Marriott LTT
Posts: 10,344
Even if it is married and same flight number, the upgrades are processed separately. So, you will be eligible for a CPU on each segment (or that's what has happened to me in this scenario, which was very often SAT-IAH-DFW). So, you're good either way if your experience is consistent with mine.
You are likely hitting the jackpot with SAT-IAH and IAH-DFW being available at the same time.
-RM