There's a deep irony in last week's announcement that Microsoft's head of Windows marketing, Michael Sievert, has "left the company," corporate-speak for "been dumped." At least they didn't add the killer clause "to pursue other interests," corporate-speak for "dumped with enthusiasm."

This move is traditional for under-performing tech companies. Marketing gets blamed and the head of Marketing gets walked off the gangplank.

The irony is thinking that somehow different marketing would convince more customers to purchase a product, whereas the evidence is that marketing has minimal effect on adoption. In the tech world, if the product helps people do their job better, it can find a home. Absent a pressing need, no one will bother. The most important factor in making a product decision is ... the product!

It's tempting to think that a new spin will somehow make potential customers see the light. In a way, changing the marketing is almost like blaming the customer for just not being smart enough to understand why he/she/it should buy. "We'll tweak the messaging to highlight the items we haven't communicated well." In reality, it's never the communication, it's the product -- but it's easier (and cheaper) to change ads than it is to admit the product is inadequate and needs to be changed. Moreover, firing Marketing makes it look like the vendor is doing something to address its marketplace problems.

Without rehashing the myriad criticisms of Vista, Microsoft's main problem is that there is a perfectly acceptable alternative already in use with most potential customers: XP. And Vista does not provide sufficient improvement to justify the license cost and work required to migrate.

It wasn't like that for XP. People adopted it quite rapidly, because it was significantly better than the previous Microsoft version. Thinking that the reason Vista isn't being adopted is because it needs better marketing is a serious misperception about the realities of day-to-day technology use.

The countervailing example that reinforces this point is what is going to happen with Microsoft Server 2008. It's gotten very good reviews, and people are going to move to it with alacrity -- because it offers significant improvement over the existing product.

Of course, Microsoft expects that eventually people will upgrade to Vista, if for no other reason than new computers will come with it pre-installed. But will they?

With the rise of virtualization, I can certainly see companies choosing to run virtual instances of XP well past the "official" XP availability date. They'll strip off Vista, install a hypervisor, drop an XP instance on top, and go on their merry way. Even worse, from Microsoft's point of view, is the alternative: taking this migration as an opportunity to rethink commitment to Windows and consider moving to Apple or Linux (both of which, by the way, provide a very good virtualization solution to extend the use of XP).

The key, in my view, will be what software providers do regarding Vista. Will they migrate to Vista-only releases, or continue to release on XP as well? If software vendors stop supporting XP, companies will have to move to Vista. If, on the other hand, they continue supporting XP (and why wouldn't they if it continues to be a viable market?), the impetus to move to Vista is dimished.

The bottom line is that people (and companies) make technology decisions based on solving their technical issues. Shouting louder won't make them ignore their real-world experience.