SGI Revenues: GAAP v. non-GAAP

SGI (Silicon Graphics International) reported results for its quarter ending December 25, 2009 today, and the numbers look good to me. However, that interpretation requires breaking a general rule.

I like to use GAAP (Generally Accepted Accounting Principle) numbers unless there are good reasons to use non-GAAP numbers for putting a value on a company. Usually revenues are the same, or close, for GAAP and non-GAAP. If there are differences, they appear as non-GAAP net income and EPS. In the past some companies have pushed non-GAAP numbers that misled investors.

In this quarter, for SGI, GAAP revenues were $94.1 million, but non-GAAP revenues were $151.5 million. How can that be? In this case, apparently the higher figure represents payments received on equipment and services sold. But because the high-end computer equipment sold by SGI includes software and maintenance commitments, all of the revenue can't be booked under the accounting rules used.

SGI results from a combination of Rackable Systems and the old, bankrupt Silicon Graphics that took place in 2009. The old SGI was technologically formidable, but had a business culture of putting innovation above profits. I had some worries that the combination might end up with the wrong culture, but so far I have been pleasantly surprised. Rackable bought SGI's assets for a song, and brought a big pile of cash into the combined company. Having shed its debt and some not-so-brilliant business ideas, SGI looks pretty hot. It sells supercomputers and other high end products and services to government agencies and enterprises that need them. Rackable, of course, was known as the leading vendor for energy efficient server farm systems.

Still, there is a way to go before getting too bullish on the future of SGI. Non-GAAP net income was only $5.6 million, or $0.18 per share. While that might seem to justify today's stock price, there are a lot of moving parts involved.

I own some SGI stock (I kept my RACK stock), but would like to see at least a couple more quarters of good revenues and profits before adding significantly to it. It is a classic investor dilemma. If I wait and SGI does well, I'll certainly have to buy in at a higher price than today's. On the other hand, the potential for going back to a string of losses with just an occasional quarter of profits is all too real at this point.

For a detailed account, see my SGI Analyst Conference Summary for 2/2/2010. The new corporate web site is www.sgi.com.

So I had better ...

Keep diversified!