Lightspeed’s PR Week Column: 10 Lessons Learned from Microsoft’s Service Disruption

In Uncategorized by Ethan Rasiel

Lightspeed CEO Ethan Rasiel contributed a column for PR Week on lessons learned from Microsoft’s handling of a service outage.

10 lessons from the Microsoft Exchange Online outage about handling a service disruption

June 26, 2014 by Ethan Rasiel, Lightspeed PR

On Tuesday, Microsoft’s Exchange Online service went dark for the entire business day, preventing its users from sending or receiving email. As Computerworld put it, the service had become Exchange Offline.Notably, Exchange is not a free service like Gmail but rather a part of the paid Office 365 service, meant to deliver increased reliability for businesses that cannot afford to be disconnected.

Glitches will happen, and I don’t expect any service to be perfect. But as a communications professional, I do expect customers, journalists, and the general public to be kept aware of the situation in a timely fashion. By any measure, this wasn’t what happened here. So, what can we learn from this about what not to do?

Ten ways to mishandle a service outage:

  1. Use jargon instead of plain English
    Microsoft’s initial communications on the matter – and the only update for most of the day – was provided through a customer dashboard that read: “A portion of capacity responsible for facilitating connectivity to the Exchange Online service has entered into a degraded state. Engineers are actively working on a solution to remediate impact.” This statement is dense and inscrutable and creates an impression that Microsoft is less than transparent.
  2. Hide your CEO
    Microsoft has a new and unproven CEO, Satya Nadella, who hasbeen on the job less than six months. This was an opportunity for him to show that he cares about customers. Instead, he has said nothing.
  3. Promise updates and then fail to provide them
    As ZDNet reported, Microsoft promised an update on the situation at 3 pm EST. When 3 pm came around, the projected update time shifted to 5 pm.
  4. Ignore social media
    Microsoft’s primary Twitter account never once mentioned the issue, but did manage to send out several other tweets the day of the outage, promoting Bing, Surface, and other products. Microsoft eventually issued a single tweet via its Office 365 handle downplaying the issue and asking customers to “see the SHD for service status” – without any clue as to what an SHD is or how to find it. Meanwhile, hundreds of tweets per hour from angry customers were flowing across the Internet. (See Figure 1)
  5. Stiff-arm the media
    Many publications noted that Microsoft didn’t return phone calls requesting comment. As CRN reported, Microsoft provides service updates through a dashboard that only customers with administrative privileges could access, meaning journalists had no way to learn the status of the service. By comparison, Amazon and Google make this information available on public websites.
  6. Don’t acknowledge the problem
    Microsoft didn’t publicly announce the outage until late in the day, as Redmond Magazine reported. Instead, Twitter carried the burden of informing people.
  7. Downplay the issue
    Microsoft’s first public statement on the matter – at 2:30 pm EST – stated, “Some Exchange customers are experiencing email delays.” The outage was widely reported to have lasted from nine to 12 hours, and customers lost email entirely for the duration of the outage. This is not a “delay” – it’s a day. Microsoft has also declined to state how many people were affected by the outage.IDG reported that “judging by the volume of complaints posted in discussion forums and social media sites, it must have hit a substantial number of users.”
  8. Spin your apology
    Microsoft eventually issued an apology, as the Seattle Timesreported, after service was restored late in the day. The apology was followed with an attempt at optimism: “[We] continuously strive to improve our service and using these opportunities to drive even greater excellence in our service delivery.” Bad grammar aside, this feels incredibly empty. Customers would rather hear specifics, not vague promises to do better.
  9. Don’t explain yourself
    As of this writing (a full day after the outage began) Microsoft had failed to provide or even promise any explanation of the issue, its cause, and steps the company is taking to prevent similar situations in the future. Instead, The Register reported angry customers complaining about two-hour hold times trying to get an explanation or update.
  10. Don’t communicate with customers
    As a Microsoft customer whose business was affected by this situation, I can report that a full day after the outage, I have received no communication from Microsoft. A few years ago, I had trouble streaming a film on Netflix. I never complained, and it seemed to be a one-time glitch, but they noticed the problem anyway. The next day, Netflix emailed me an apology and a credit toward my account. Earlier this year when I had trouble placing an order on Seamless, I tweeted about it, and someone from the company responded to my Tweet personally with a solution. That’s how it’s done.

Microsoft could have taken steps to manage this situation effectively by being transparent and responsive with customers as well as journalists. Instead, it is now dealing with a continuing backlash and the emergence of a trending hashtag on Twitter: #Office364.

Asked for comment, a Microsoft spokesperson provided the following statement: “On Tuesday, June 24, 2014, at approximately 6:30 am EST, some North American customers experienced email delays with Exchange Online. The issue has since been resolved and the service is now functioning normally. We sincerely apologize to our customers for any inconvenience this incident may have caused and continuously strive to improve our service and use these opportunities to drive even greater excellence in our service delivery.”