Microsoft cloud failure hurts its credibility and prospects for Office365
Outages aren't the problem; silence and a lack of options are the problems
The days-long SAAS outage Microsoft only admitted after the fact won't do much to encourage customers disenchanted with the bloated, expensive Office suite to stick with Office by signing up with the cloud-based Office365 instead.
Starting Thursday, email and some other services were unavailable on Microsoft's Business Productivity Online Services (BPOS) service. BPOS is a SAAS bundle of Exchange Online, SharePoint Online, Communications Online and Live Meeting, all hosted by Microsoft.
Customer forums were filled with complaints about having to go three days wihtout email, and intermittent problems going all the way back to January.
The problem cropped up the form of "malformed email traffic," according to Microsoft, which resulted in email delays of up to six hours. Later problems with DNS failures, messages backups and a second wave of malformed email caused the ongoing problems, according to Microsoft's explanations.
The outages affected BPOS, not the new Office365 service, which was opened for beta testing last month, Microsoft was careful to point out.
Unfortunately, the Dashboards and constant updates Microsoft promised BPOS users didn't give much information about the outage or how to work around it until two days after the initial failure.
That delay is an example of "the cavalier way Microsoft's treating its paying customers. Again." according to InfoWorld's Woody Leonhard, who is testing Office365.
Microsoft also offered no updates, notices or warning during a series of email outages August and September of 2010, Leonhard wrote.
Occasional outages – as with Amazon – have to be expected in SAAS or cloud services. They're a problem, as they would be with any IT failure.
They're a manageable failure if the customers have recovery tools, workarounds, alternate service-provision points or other options in case the service goes down – but only if the service provider is quick to admit the failure and give customers whatever information it has on the problem.
Both failover resources and immediate, detailed updates on service levels are routine parts of most outsourcing contracts, but are vanishingly rare in the U.S.-based SAAS and cloud markets.
That lack of contingency planning – more than security, cost or ignorance – is the biggest holdup in the widespread adoption of cloud-based resources by enterprises.
Business-unit managers with credit cards keep pumping money into the market. Behind the scenes IT is either unprepared or fuming, because they're not being allowed to do their jobs by providing backups and failsafes that can pull business managers out of a tough place when some new IT service they hired goes belly up, taking corporate data or wasting hours of time for end users while everyone stands around waiting for the cloud to come back online.
Analysts say that gap is because of the immaturity of the cloud market – which is ramping up in both reliability and backup options as it becomes more established and more credible.
Microsoft's BPOS outage – especially frequent outages – don't help.
Not admitting the problem or paying customers the courtesy of explaining what it was until long after it's fixed show an immaturity in the vendor, not the cloud technology.
No matter how close the experience is to your current office suite, it's not a good idea to go with an immature service provider and immature service as a replacement for anything as critical to the hour-to-hour process of conducting business as email.
Even though the outage hit BPOS, not Microsoft's newer service, the ultimate hit in credibility won't be to BPOS, it will be to Microsoft itself and, especially, Office365.