Think for just a minute about the amount of complexity involved in pricing out any given flight itinerary. You’re dealing with thousands upon thousands of flights, a few thousand airports and hundreds of airlines, all with their data in a massive system. Fortunately the interfaces passengers work with simplify things dramatically, but that doesn’t mean it is actually simple. I just booked a mileage run for the first weekend in November to wrap up my qualifications for the year. And given just how complex it is to book, much less try to explain, I figured a real world example would be useful. Here goes!
I decided on a trip to HNL based on the fact that it is about the number of miles I need and it is relatively cheap. At $532 round trip this will be, by more than $100, the most I’ve paid for flights to Hawaii. And at 5cpm it is not quite great as a mileage run. Alas, I’m running out of time in the year and longer stays (for cheaper fares) or more circuitous routings aren’t viable. So I start with a generic search on ITA between NYC and HNL:
I get a calendar of options and the November 3 departures look reasonable (again, no more weekday travel for me this year so I’m stuck paying the premium):
The Hawaiian service is pretty awesome and if I was actually looking for a vacation I’d be booking that. But I’m not.
I like the United Airlines option so I start to drill down:
Those are the only two-stop options shown and the price on them is not at all competitive. Ouch.
Don’t give up hope at this point. ITA has an internal timeout on their systems. If you give it too generic a search (like I did here) it will go for the easy options rather than the more interesting ones. We have to give it a bit of encouragement.
Here’s what my next search looks like:
Note the UA UA UA+ on the second line. This tells the system to search for trips which have at least three flights marketed by UA (including codeshares) and the + at the end means anything else after that is acceptable. I also put in specific dates as I’m pretty sure that’s the right weekend for me to be gone. This search gives a ton of results the initial search missed, including the fact that 2-stop options really can be priced right:
Getting closer now, but still not exactly there. For the outbound I want to fly from LaGuardia. It is a bit cheaper to get to and the timing on the LGA-IAH flights means I can actually sleep in more on Saturday morning. On the return, I want to force the HNL-IAH flight, both because it means more miles and because it lets me connect to the IAH-EWR flight scheduled to be on the 787 on November 5th. And since HNL-IAH-EWR has R inventory open for upgrades I definitely want that route. To mix LGA and EWR I switch to multi-city. I also add IAH in the return routing details which forces that as a connection point:
I get a number of results for the outbound:
Choosing the 9am departure (yes, tight connections, but I’ll take the sleep and deal with a misconnect if it happens; I’m not at all worried since I don’t really have to get to Hawaii at any particular time) I get the following options for the return:
Choose the return I want (hooray, Dreamliner!) and I get the final itinerary:
As complex as it was to find the flights I want, making the booking is almost more difficult. It turns out that LGA-IAH-SFO-HNL is not a logical routing and, with all the other options for the same or less money and way less travel time involved, most online booking sites aren’t all that interested in offering it as bookable. Now it becomes a game of tricking the systems to offer the trip I want.
Once again, multi-city search comes to the rescue. I started with LGA-SFO, SFO-HNL, HNL-EWR. I could get the LGA-IAH-SFO flights I wanted but it wouldn’t add the non-stop SFO-HNL from there. And the price was coming out a couple hundred dollars high. I switched the query to LGA-IAH, IAH-HNL, HNL-EWR and the flights I wanted magically bubbled to the surface. A few clicks later and I now have a confirmation email from United. And a few clicks after that, confirmed seats up front on the return with an RPU debited from my account. Not too shabby.
I should note that I also use a couple different tools to compare the offered routings and figure out how many miles I’ll earn. In some cases the 6am flight might be worth many extra miles, making it more viable. Or adding an extra connection. In this case, however, adding the extra connections didn’t help enough to add the extra time to my trips. One tool is the one I built for comparing Star Alliance programs. If you have specific fare classes and operating carriers it can give you pretty good details on the earning rates. The other tool I use is the Great Circle Mapper. I find it easier and faster for quickly comparing routing options on the fly. Knowing that I would earn an extra 99 miles with the SFO connection rather than the LAX connection was a good nugget of data.
As I mentioned at the beginning of the post I think that this one is a bit pricey for a mileage run. That said, I do get to fly a 757, 767, 777 and 787 and the return can be confirmed in long-haul Business seats (777 is the old UA but still good enough for sleeping on the redeye) and, more importantly, it is the right number of miles on the right dates. And I get the 787 on day 2 of service, in BF, no less. All in all, not such a bad deal. Adding the ORD connection on the return would have meant 260 more miles but also getting home later than I want and not the aircraft I want. I’m going to pass.
It sounds complicated in terms of building the routing out with ITA but with a bit of practice it does get easier. Give it a try and if you run in to trouble feel free to ask.