Where Microsoft's Profits Come From

derrida writes "Microsoft is the largest, most profitable software company in the world. In case you had any doubts about where Microsoft's profit comes from, there's nothing better than a graph to make all those numbers clear. As you may have guessed, the desktop division is quite profitable, while the online division is a money pit."

Re:The chart is mis-labeled

[biryokumaru]

Isn't 2007 the one with the ribbon that no one can use? Doesn't that make it a new product, the fact that no one knows how to use it anymore?

Re:Interesting graph!

[Anonymous Coward]

Some actually are buying retail copies - but not all that many. Most of it is people buying computers near the holidays. This time MS actually had the OS ready for those people buying near Christmas (they had missed that date with Vista and of course Vista's reputation kept many away too). Lots of people had put off purchasing a new box until it was going to come with Windows 7.

Re:Interesting graph!

[rjch]

What I would find interesting is to know what events occurred during the valleys and rapid climb moments indicated in the graph. Specifically, what happened in Dec '06 and Sep '09?

December 2006 was the release of Vista. (Well, November 30th, but close enough) September 2009 was the release of Windows 7.

Re:Interesting graph!

[xZgf6xHx2uhoAj9D]

If the different colours are perfectly parallel, then there is zero movement in the upper layers and they only look parallel due to how the data is presented (stacked). In order for them to be "synchronized" you'd have to see the layers diverging from one another, not parallel to one another. You can a little bit of this, but not much. For instance, between December 2006 and March 2007, Office sales diverge a wee bit from the layers underneath. The Servers and Tools seems to stay completely flat, maybe even shrinking a bit.

Really it's just Windows sales going up and down and the two layers on top of them not doing very much.

Re:Interesting graph!

[zippthorne]

Normally you avoid data distortions like this by using a better kind of chart.

The problem is that they're trying to visualize two different things in one chart (relative and total values), and the compromise you make doing that in a stacked chart pretty much sacrifices everything except the sum of the values.

Also, area-shaded line graphs make absolutely no sense if you've only got a few data points.

Re:The chart is mis-labeled

[DesScorp]

Can you still get winXP and office 2k? Maybe he had to buy a new computer and didn't want a legacy OS on it. Computers don't last forever you know.

You can still get brand new Windows 95 discs on Ebay. XP and 2K are no problem to acquire. And I'm of the school that says unless there's a real reason why you should upgrade, you shouldn't be forced to. Lots of people use older operating systems because it suits their needs. I'd say for 90 percent of businesses, Windows 2000 would quite ably suit their needs. The only reason many businesses upgrade is because "Microsoft tells us it's time to upgrade".

Re:interest income?

[DaveGod]

Finance income and charges are added/deducted after operating profits. Investors usually want to look at how a company manages it's finances differently to how they want to analyse operations. Wiki has an example income statement [wikipedia.org].

Re:The chart is mis-labeled

[TheRaven64]

I've had two books published that I typeset with LaTeX, so I'm biased to agree with you, but what you say is nonsense. Even MS Word For Windows 2.0 could properly support styles. I could - and did - write documents and then change the heading styles later. Unlike LaTeX, it provided inheritance between styles too, so you could define a heading style, then a heading1, heading2, and so on style for the different heading depths, and update all of the subheading types just by changing the heading style that they inherited from.