Southwest Airlines faced another potential crisis just two weeks after its holiday meltdown as the FAA shut down all flights in the U.S. for nearly two hours last Wednesday.
Southwest canceled more than 400 flights and delayed more than half of its remaining flights that day. But instead of the same crew scheduling malfunctions that led to the cascading cancellation crisis during the peak holiday travel period, Southwest was back and functioning normally the next day.
At the center of this has been CEO Bob Jordan, who was promoted to lead the Dallas-based company a year ago. He had already said that upgrading technology was a priority, but that came too late for the fiasco in December.
Jordan talked with The Dallas Morning News this week about what he learned after the holiday meltdown and what his company is doing to make sure it doesn’t happen again.
This interview has been edited and condensed for brevity and clarity.
Q: Was Wednesday’s FAA notification outage and ground stop a stress test for Southwest after the issues over the holidays?
A: We deal with irregular operations all the time. We don’t deal with events like a two-hour ground stop. So yeah, it was absolutely a test. We put in a lot of things in December that helped us to get us back to the full network this time. We created a group of new, trained crew schedulers. I think it was about 100 that we can pull in at any time. We actually activated part of that group in case we had a large set of crew scheduling transactions. It worked really well. We’ve also got procedures now to see more early warning detection around irregular things in the network. We are just calling it operational hypercare – more leadership and eyes on the operation. We had a lot of folks that were here all night.
Q: With what Southwest went through and the FAA issues, is there a technology problem across the industry?
A: I can’t speak for the FAA, but I think no. Every company has a lot of technology. We spend $1 billion a year on technology. We’ll talk a lot about the crew-solving software. We actually have done eight new releases of that in the last year. So you’re always working on stuff and then you have stuff that is older. What you don’t want are single points of failure where an issue can take down a critical system.
Q: Crew scheduling and SkySolver have been discussion points for a while. Did you know it was this vulnerable?
A: I don’t think I would use the word vulnerable. I think what happened is we had a historic storm. I’m not excusing what happened. I mean, we messed up for our customers. We messed up for our people.
We’ve done eight releases of SkySolver and we’re working on a release with (GE) right now that would automatically solve those old problems.
Q: Can you talk about the moment on Dec. 26 when you decided you would have to cancel two-thirds of flights to get back to normal?
A: We were watching obviously. There’s a lot of vigilance every day, and more vigilance operationally when you have a storm like this that’s going to hit the network. Then the situation got worse, things like frozen deicing fluid and more problems day to day. Each day though until probably the 24th, we were roughly aligned with the industry in terms of cancellation rates. We were seeing the needed signs of improvement.
It was really the 24th that it looked like we were diverging from the industry recovery. Now, we had a good plan to go solve that overnight. The same thing happened on the 25th. We had a good plan to start the day reasonably well on the 26th. When we woke up, it wasn’t possible to get all those past problems cleared. We were in worse shape than expected.
That morning we had a lot of discussion about what we needed to do to get the network recovered. Basically, I’ll just call it a reset. A lot of smart people were working on it and the only way to do that was to take the level of flying to a point you knew you had plenty of crew coverage. They took the network down about 60%. You also had to leave enough of the network in place so that crews could continue to come in from crew bases.
In every discussion that we had, it started with safety. And we were willing to do anything including stopping the operation to ensure that we stayed safe.
Q: What else is Southwest doing short-term to make sure this doesn’t happen again?
A: We’ve always had operational indicators or a dashboard to understand what kind of shape the network is in. We’ve done a lot of enhancing.
Supplemental staffing is a big deal because until we have technical solutions, the only way out is a manual effort with enough people to work the transactions. It’s over 100 folks now who can do the basics of crew scheduling like reroutes and notifications. We put in some additional crew communication tools where you can electronically send information. In the past, it has taken a phone call.
I talked about the work that we’ve got going on. I think it’ll be pretty quick to have a technical solution to this past problem. There are also some things that we learned operationally and a lot of those were already on the list.
In a case like this when it’s so cold, (crew) had to be rotated in and out more frequently because of the risk of frostbite. We had to stop the deicing and bring in people and change them out. We’ll be looking at our deicing top to bottom.
For long term, we’re working with Oliver Wyman and they’re industry experts and consultants in aviation. We’re doing a top-to-bottom review of what happened. We’ve already got really good operational plans that I’m sure we will learn things that we need to invest in more. We need to, maybe, change priorities. It will not take us months.
The other thing is our board’s really involved. The board has a duty of oversight and governance and the board has been involved since early on. They formed an operations review committee to provide oversight and assistance to management.
Q: How long will it take to overhaul the crew rescheduling system?
A: The tool that we use, it’s not used by everybody, but it’s an industry-standard tool used by a lot of airlines. We’re not on a real old release of that, I think we’re maybe one release behind. There’s been a little misconception that we have unique archaic tools or something like that. We don’t.
We’re going to learn things that we need to do to make the technology better, to make the processes better. We’ll learn all kinds of things in the crew area and it was already a big priority. In the last year, we added a senior leader over that area. I think we’ve hired 120 crew schedulers.
Q: You’ve spent a lot of time in technology at Southwest. Has it been hard to make technology a priority?
A: That’s an absolute misconception again. We have a great technology department. We had a big technology plan for operations underway. Technology is always an ongoing journey. You always have things that you’re working on.
Will we make changes? I am sure we will. There will be areas that we need to invest in even more, there will be things that we need to accelerate. I guarantee there’ll be changes to our plan to put even more focus on operations because, and I can’t say it enough, this just can’t happen again.
Q: There’s also a perception that Southwest has skimped on technology spending all the way back to co-founder Herb Kelleher. Is that accurate?
A: Not at all. If you go back, I started in 1988 as a programmer here. We were a much, much smaller company. We’ve always invested in technology. No, we have not skimped on technology. I promise you that.
Q: Do you have to slow growth plans until the technology can catch up?
A: I’m not worried about our growth rate at this point. I’m committed to doing anything we need to do to mitigate the risk of this happening again. It was a very unique set of events. And it didn’t start as a technology issue and, in my mind, it didn’t end as a technology issue. Again, the software worked as it needed to work and was designed to work.
Q: Is the point-to-point flight network that Southwest has too complex to put back together after these operational difficulties?
A: It’s been a topic of discussion for years and years and not just because of this storm. We’ve got a 50-plus-year history that shows our model serves our customers really well. On the operation side, the point-to-point network has actually been a plus in terms of recovery because we have fewer customers that rely on connections. All that said, everything that we need to do to mitigate the risk is on the table. Do I think there’ll be big changes to our network? No, I don’t. But if there are things that come out of the study that are tweaks? Absolutely.
Q: Where do you stand on getting bags returned and refunds and reimbursements to customers?
A: We’ve done 90% or so in terms of refunds completed and we’re handling those in roughly three days. The Department of Transportation requirement is seven business days. We actually wrote some automation very early on so there’s a bot running that automatically handles the request on simpler cases. This bot is processing the majority of refunds. We’ve got 98% or 99% of bags returned to customers. The 1% are the ones that are really hard – no name tag or bag tag.
The third area is reimbursements and we’re absolutely leaning into our customers on alternative flights, rental cars, hotel rooms, all the things that you would expect. They’re more complicated because you have receipts.
We mobilized an army here at Southwest as well as a third party. We have upwards of 3,000 people working on reimbursements. We processed over 30,000 Thursday. We’ll move through all those reimbursement requests in a few weeks.
Q: Will customers continue to be loyal?
A: I could talk all day about how unique the event was, but at the end of the day, we messed up for our customers and we messed up for our employees. And it’s a big deal. We need to do everything we can to mitigate the risk of this ever happening again.
We have a lot of goodwill in the bank with our customers, but we have no intention of using that up. You get to an event like this and we know we used up goodwill and we have work to do to earn our customers’ trust back. At Southwest, we’re transparent. We love on our customers and our employees and we know we have work to do to restore that relationship.
