Boy does Amazon have some trouble on its hands today. With EC2 clawing its way back into working order, and Hootsuite, Foursquare, and Reddit addicts still shaking uncontrollably from a day’s down time, the giant of affordable, flexible cloud computing is going to have to figure out how to restore confidence in a service that already struck many end users as, well, nebulous.
While Amazon’s EC2 customers — and probably Amazon’s engineers as well — would love to believe that redundancies would never allow a massive failure to happen, yesterday reminded us that sometimes the unthinkable happens, even to methodical planners. What separates success from failure at that point is how intelligently and quickly the problem is managed — and not just from a tech perspective. From a reputational standpoint, too.
The folks at Amazon are no doubt working like dogs right now to restore, resolve, and prevent a recurrence of the EBS re-mirroring overload that took their storage to capacity and beyond. But as knotty as the technical aspects of the fix may be, the cloud giant should also be taking this time to publicly reassure the world that cloud technology and Amazon itself are worthy of another go. This is no time for a company to, as the New York Times reported, “not respond to requests for comment.”
Crisis PR isn’t rocket (or computer) science — but you do have to take it seriously, and long before anything hits the fan. A few important concepts for crisis management:
- Planning – Analysis of potential “points of attack” on corporate reputation – the vulnerable salients where a company faces potential exposure and damage. Develop advance messaging for negative events, and assignments of “war room” responsibilities for when and if crisis strikes.
- Practice – Training and role-playing for the internal crisis PR team to ensure readiness. No one ever prepared for “the worst” just by reading a plan.
- Proactivity – Tools and skills for going on the offensive when a crisis strikes – and to avoid worsening the crisis through stonewalling.
Following that framework, what should we have seen from Amazon? Instant public comment and accessibility with a calm, apologetic, and down-to-business tone accepting responsibility and providing good reason to believe that all will be well. Quotes to every major interested media outlet, answering questions and providing up-to-the-minute information. What did we see instead? Silence from Amazon in media coverage; status info on the company’s own site; and — I kid you not — an online form for media/industry analyst queries. Not good.
Amazon will undoubtedly fix the problem in its data center, but it should spend as much time fixing its crisis response strategy for the next time the unthinkable happens. Confidence is a hard-won asset; you don’t just let it go up in a cloud of smoke.
