Esther Dyson’s Flight School

I just visited DayJet to chat with Ed Iacobucci, who is going to be a speaker at Flight School this June. It was an enjoyable trip. (Disclosure:) I got a free sample, which made the case. I had just flown in from Germany, almost a 10-hour flight. My ride to the Opa-Locka general aviation airport (DayJet flies only from smaller airports) took about 35 minutes. My flight to Boca Raton, where DayJet is headquartered, took 20 minutes (and that was with an unusually long route because of ATC vectoring). My ride from the Boca Raton airport to my hotel took 5 minutes.

I was a good example of what DayJet expected and what is turning out to be true: The vast majority of its trips are replacing car trips, not air trips (either commercial or charter). Had I not been able to use DayJet, I would have taken a taxi.

 (Just for the record: Last year I took a 90-minute paid air-taxi trip on SATSair, from Orlando to St. Simon’s Island. That would normally be a 300-mile, four-to-five-hour drive, but for me, a non-driver, it would have been from $400 to $900 and five or six hours in coach via Atlanta or Charlotte, before TSA and waiting time, and to Savannah, still about an hour away from St. Simons Island by car. SATSair charged about $800, as I recall. DayJet currently quotes from $1338.89 (for a two-hour window)  to $500 (for a six-hour-window) for a one-way trip from Orlando to Savannah.  It serves Brunswick, which is much closer to Savannah, but it doesn’t serve the Brunswick-Orlando city pair.)

I won’t explain the whole concept here, but let’s just say it reminds me of Federal Express – even though its concept is precisely the opposite. Way back in the mid-70s when FedEx started, it created an entirely new concept. Packages don’t care where they go on the way to their destination, so FedEx would send them all through Memphis – no exceptions. Manhattan to Brooklyn – through Memphis. Even Manhattan to Manhattan – through Memphis. (Although that has changed since.) The idea was that the costs and glitches of complexity outweighed those of transportation, so it would create a reliable, routine service – one that revolutionized the courier-service industry and created the overnight-deliver business.

DayJet’s premise is the opposite. People do care where they go, and the costs of complexity are way down given today’s computer systems and software, so Iacobucci created a company where there is no routine: Every trip is an exception, scheduled for specific passengers… but by computers, not by people. It’s no surprise this comes from the man who created the Citrix operating system, which is basically software for allocating scarce computer resources in real-time. DayJet allocates scarce aircraft and crew resources in near-real-time. (You can find out more at www.DayJet.com.)

But just as the courier-service people didn’t quite understand the point of FedEx, so do today’s charter operators not understand the implications of DayJet.

New markets, new metrics

“We’ve had to create new performance indicators,” says Iacobucci. “Our load factor right now is about 1.5 [of 3 seats on a plane]. But our effective load factor is lower, because we often fly people extra miles to get them where they are going – and they pay for the originally requested mileage. On the other hand, the company does very well when it fills all three seats – and as it scales up that becomes easier and easier.” Another metric is “quality of load,” which is determined by the customers’ time windows: Did customers book within a broad time window, spending less money, or did they demand tighter scheduling (and receive it)?

For example, there are two flights, from A to B and from B to C. Both have two passengers, so the apparent load factor is 2/3. But there are three passengers, one going from A to B, one from B to C, and the third from A to C, so the *effective* load factor was 3 trips out of 6 available, or ½. And the quality of load was even tougher to calculate, depending on the specifics of the customers’ time windows and the distances flown rather than the number of flight segments.…

Overall, Iacobucci says, the company is pretty much tracking expectations…after a late start last October mostly due to late delivery of the aircraft.

One specific surprise was that Tallahassee was so popular – a triumph of viral marketing among people in a community that revolves around the state government.

Training the customers

Another one is simply the time it takes for people to understand the service. (I remember the early FedEx commercials. First they explained the service. Then they focused on how even small businesses could use FedEx.) DayJet has the benefit of the Web, so that they can get price quotes for themselves and experience (even if a first they don’t understand) the trade-offs. For starters, you don’t get a discount for booking in advance: Since the schedules are fixed only the night before each day’s flights, being early doesn’t help. (Unlike most charter operators, DayJet owns its own aircraft, so it doesn’t have to allocate aircraft early or juggle them late in response to a private aircraft owner’s whims.)

Instead, you can get a better price by stating a bigger time window: I want to leave after 10 and arrive before 3, vs. I want to leave at 10 and get there by 11. The flight for the first is priced on the assumption that the capacity can be shared with at least one other passenger, while the second in essence pays for the whole plane. (That of course will be adjusted over time, in response to demand, capacity, competition and other factors. But it’s the essence of the DayJet per-seat concept.) You can also pay extra to be guaranteed a nonstop flight. “Every constraint has a cost,” says Iacobucci. “And every loosening of a constraint [such as a broad time window] makes things cheaper.”

So people are beginning to learn how it works… And Iacobucci can watch that happening through his software, which has one of the greatest sets of data and visualizations I have ever seen. New customers’ initial queries are wide off the mark: They enter constraints that result in too-high prices or they look for low prices and find flights that don’t suit them. But over time, each customer tends to make fewer queries before booking, and the initial pricing is closer to what the customer ends up paying. And experienced customers don’t book in advance.

Perhaps the most interesting is the metric that indicates how people value their time – which was estimated in one NASA study at $34 an hour for a generic business traveler. You can determine that, of course, by how much extra people will pay to save an hour of time. (Of course, even for any one person that will vary over a day and across days…) DayJet’s customer base, of course, is hardly average – and we’re talking mostly about people spending their employers’ money, not their own. The day I was there I saw figures ranging from $12 an hour to $300…

And how is it trending? If people become more productive through DayJet, over time they may be willing to pay more for the service. And indeed, says Iacobucci, “We’re seeing that the estimated value of time is higher for repeat customers.”