Why does Microsoft make the decisions it makes? If you ever wonder that, you're asking the wrong questions, because Microsoft isn't actually a monolith. It's a series of fiefdoms; quarrelling fiefdoms, as one of my favourite cartoons of the year illustrates beautifully.
The real org charts, by Manu Cornet
'Big Windows' argues with Windows CE and Xbox, the Windows kernel team argues with Windows architects, the LightSwitch team face off against the Jupiter team - not for personal reasons but because they all have different visions. I think that - apart from the politics you'll find in any business with more than three employees - Microsoft mostly treats this as a Darwinian process, letting the customer decide what they want. That doesn't help when the argument is about something inside Windows; say, who gets to control the timer - the kernel team or the power management team? It can lead to competing products; the tablet solution is Windows 7, no Windows Embedded, no Windows 8…
But the Microsoft executives in charge of teams and divisions tend to have a pretty big incentive to produce what they think the customer is going to want, because Microsoft emphasises customer satisfaction where it has the most impact - in the executive paypacket.
Discussing the upcoming System Center Advisor tool (you may remember it as Project Atlanta) with Brad Anderson, the corporate vice president of the Management and Security Division, we asked if it had been easy to decide that a tool that makes a product you've already paid for run more reliably should be free (at least if you have Software Assurance). After all, it's based on analysing the systems that are on customer sites to see what works well, as well as picking the brains of the teams running SQL Server in house at Microsoft IT. On the other hand, many other enterprise software vendors would charge a premium for it… In the course of explaining the decision, Anderson said something interesting.
"We debated a lot what the right way to bring Advisor to market was; we made a conscious decision to err on the side of getting it deployed in as many customer environments as possible to reduce any and all friction. Because the impact of eliminating 'crit sit' [critical situations] and downtime is the biggest lever for our business because customer satisfaction is higher. " (Incidentally, Anderson claims that Advisor will reduce SQL Server downtime by 50%, so it's well worth trying; and versions for other servers are on the way."
"One of the primary ways all the senior leaders at Microsoft are paid is, a lot of our compensation comes down to the reports that come back on customer satisfaction. So it encourages a long term perspective and a long tem investment in driving up customer satisfaction. We know customer satisfaction is first driven by the quality of products; one of the other top priorities is 'Is it up, is it delivering on what I need, how much caring and feeding is it needing?'. "
"What the Advisors are going to do is take all that learning, all that knowledge that comes from around the world, and deliver that down proactively in this continuous manner. The only impact that will have is to drive customer satisfaction. What is this going to cost Microsoft? Well, there's an awful lot of storage it's going to take up - there are real costs associated with this but that focus on customer satisfaction is the driving force. "
The Windows 7 and Windows Phone executives must be doing well, given the high customer satisfaction scores both products get from the majority of users. And if you're not happy with a Microsoft product, it's an incentive to find a way to get your feedback to the company directly - although the developer division's 'We Share Your Pain' initiative seems to have been discontinued (you can still see the project on YouTube - and yes, it is filmed at Microsoft UK and yes, it is a joke).