Monday, August 22, 2016

TechSavvy: Delta’s Digital Black Swan

MIT Sloan Management Review, August 22, 2016

by Theodore Kinni

Black Swan
Condolences if you were flying — or more accurately, not flying — on Delta last week. As the tally of cancelled and delayed flights climbed into the thousands, Nick Taleb came to mind — you know, the Black Swan guy. Taleb wrote that black-swan events have three characteristics: “rarity, extreme impact, and retrospective (but not prospective) predictability.”

I don’t know if the power failure at Delta — and the chain of unexpected events that followed it — qualifies as a black swan by Taleb’s standards, but it must have felt that way to CEO Ed Bastian. The day after the failure, he apologized for the second time and ruefully explained that over the past three years, Delta has invested “hundreds of millions of dollars in technology infrastructure upgrades and systems, including back-up systems, to prevent what happened yesterday from occurring.”

Delta isn’t the only airline whose systems have crashed recently, and with more and more companies integrating their systems, it seems like a good bet that digital black swans may become more and more common. In an article for CIO Dive, associate editor Naomi Eide offers some lessons from Delta for companies hoping to avoid similar events. First, she says, beware centralized control points, because when they go down, everything goes down. Regionally dispersed control centers are more expensive, but more robust. Second, ensure redundancy measures are in place and test them regularly. Third, practice recovery plans and responses to worst-case scenarios.

All of this still may not be enough to save your company from a true black-swan event; Taleb made a pretty strong case that they will be with us always. But it may be enough to avoid the growing numbers of gray ones. Read the rest here.

No comments: