Microsoft didn’t have to lose data cen­ter vir­tu­al­iza­tion to VMware. It wasn’t a ques­tion of infe­rior tech­nol­ogy or prod­uct func­tion­al­ity. They lost because their busi­ness model was wrong.

Microsoft’s Server and Tool Busi­ness (STB) has done extremely well by iden­ti­fy­ing mar­kets where com­plex soft­ware func­tion­al­ity can be engi­neered for sim­ple instal­la­tion and use, and then brought to mar­ket at a dis­rup­tively low price point. Time after time this model worked: Win­dows Server, Sys­tems Cen­ter, Exchange and Share­Point. Not sur­pris­ingly, when it came to vir­tu­al­iza­tion, this was the busi­ness model: offer lots of func­tion­al­ity at a dis­rup­tively low-price. Why did it fail so badly in this case?

It failed because vir­tu­al­iza­tion was dif­fer­ent. Vir­tu­al­iz­ing a set of servers isn’t like upgrad­ing the oper­at­ing sys­tem or Office. Vir­tu­al­iza­tion enables sig­nif­i­cantly improved eco­nom­ics but along the way it requires sig­nif­i­cant changes in how the data cen­ter infra­struc­ture is man­aged and how the data cen­ter is orga­nized. There­fore it isn’t sur­pris­ing that cus­tomers didn’t just jump into vir­tu­al­iza­tion. Instead they brought in con­sul­tants to help them plan and imple­ment, and along the way cre­ated a new chan­nel focused on the deliv­ery of vir­tu­al­iza­tion solutions.

VMware had an “expen­sive” solu­tion (although still a very smart invest­ment con­sid­ered by itself). Sell­ing the VMware solu­tion made a lot more sense for the chan­nel because they made a lot more money doing it, and at begin­ning the VMware solu­tion was domon­stra­bly “bet­ter” than the Microsoft solu­tion (more advanced func­tion­al­ity). So if a chan­nel part­ner could make a lot more money sell­ing VMware guess what they did?

As the Hyper-V solu­tion caught up (and with Sys­tem Cen­ter in many ways sur­passed VMware) guess what happened? More or less nothing, because the chan­nel had no incen­tive to switch brands. Yes, VMware has an “expen­sive” solu­tion, and cus­tomers would like a “cheaper” solu­tion, but that’s not enough given the part­ner economics.

I finally under­stand what peo­ple were try­ing to say when they noted that Paul Maritz knew the Microsoft play book (“he had writ­ten it”).  The genius wasn’t just to under­stand your big com­peti­tor well; it was to jujitsu Microsoft (use their own weight against them).  The big­ger ques­tion is why Microsoft didn’t see it com­ing. The ulti­mate loss is enor­mous. Win­dows Server has done extremely well in the data cen­ter. Sys­tem Cen­ter is remark­able man­age­ment sys­tem. But now flanked by VMware on both sides it’s going to be chal­leng­ing to get out of the pin­cer.  Ballmer has changed the STB man­age­ment. Microsoft is a smart and pow­er­ful com­pany. Next move is necessarily theirs.