Yesterday, I pointed out that it's absurd to consider Microsoft a declining company when in fact it's been making an average of $15B a year in profit and has increased the value of its balance sheet by billions of dollars every year for the past three years, during the worst recession of the past 80 years.

Today, to answer the critics, I will point out what's gone well and what's gone wrong with Microsoft. I'll even make suggestions about what the company should change. I believe the company is coming to a crucial point in its history, and CEO Steve Ballmer needs to make some radical changes.

First off: Microsoft's core businesses are as healthy as ever. Now that it fixed the Windows Vista debacle and Windows 7 is shipping, Microsofts' Windows Division and its Business Division are firing on all cylinders. Consider this: in fiscal year 2009, which ended June 30, 2009, the Windows division made $10B in profit. It has already equalled that in only 3 quarters of business this year. That means when Microsoft announces its earnings later this month it's likely to report year-on-year profit growth of more than 20% in the Windows division.

Of course, that's largely due to the benefit of a new product. In the business division, Microsoft will not only beneift from the new shipment of Windows 7, but also the upgrade in business applications as it shifts to a new version of Office. Profits are likely to grow there as well.

Now, what about what's wrong with Microsoft? Well, as has been well documented, the Entertainment and Devices division has been a bust and was recently reorganized, and the Online Services division continues to lose money. In the fiscal year ending June 30, 2009, the Online Services division lost learly $1.6B. It's on track to have similiar losses. Microsoft has a terrible track record trying to lead in mobile devices.

What should Microsoft do? Focus. Clearly Microsoft's strength is in creating operating systems and applications for enterprises and consumers. It has never been very good at mobile phones, online services, or media. It should pull out of these businesses entirely. Although the mobile smartphone market is growing very fast and has many temptations, the fact is that it is packed with competition.

Microsoft can make plenty of hay in PC operating systems, business applications, and consumer applications such as gaming. Will beancounters at Boeing be calculating their profit-and-loss statements on an iPad or an iPhone any time soon? I don't think so. The enterprise market will continue to be ruled by PCs and Microsoft business applications for some time. Microsoft makes lots of business softaware, and it should make more of it.

In order to focus on business and consumer applications, Microsoft should cut out products and divisions that don't work. You can see Steve Ballmer is starting to do that by nixing the management in the Entertainment Division. But he's got another problem: Online Services. Microsoft should scrap the entire division by selling it, spinning it out keeping a minority stake, or winding it down.

If you think about it, Microsoft has never been especially adept at online services or media, and this division still loses a ton of money. Microsoft should take any of those assets or cash it would receive from a sale or winding down of the division and reinvest it some of its newer successes. Let me give you an example of newer product areas where Microsoft is having some success and can succeed:

• Communications and collboration products -- Sharepoint has been one of the company's fastest growing products, and enterprise collaboration applications are clearly an area in which it can grow. It can also buy startups to link newer communications and collaboration features into this product.
• Xbox -- The product has crossed over to profits and success. Microsoft can add to this by adding functionality, more applications and services, and developing a games franchise.
• Cloud services and applications -- The threat to Microsoft is overrated. So far, pure cloud computing, in which companies outsource their IT infrastructure and applications to the “cloud,” is for startups and smaller companies. Bigger companies still want to run IT inhouse. And Microsoft's cloud platform, Azure, is being taking seriously as a cloud services platform within itself. Microsoft can focus on building an array of cloud services and applications to sell.

Microsoft has proven that when it gets elements right in its core business (Windows 7, servers, business applications, Xbox), it can still scale to new levels of profit. In addition, it has significant opportunities in places that are adjacent to its core products, in markets such as gaming, cloud services, servers and development tools. But the company has allowed many failed projects, such as Windows Mobile and online, to go on for too long, and it's time for it to focus on areas where it is having success through some major reorganization.

And to reiterate: I believe the company will have a banner profit year and the negativity on the stock is somewhat misplaced. This company has some of the strongest fundamentals in the world, and the stock is cheap. You can now buy one of the strongest companies in history at a forward P/E of 10.

(Disclosure: Long Microsoft).