Jefferies & Co. said investors are uncertain about Microsoft Corp.'s (MSFT) future in consumer technology and that they have chosen to discount all parts of the business until it becomes clearer what Microsoft’s competitive position in the key areas of tablets, smartphones and online search will be. Microsoft

Software companies love to tout their innovations, but a recent study at the University of Pittsburgh shows that the most successful put mor emoney into actually running the company rather than research.

The study was a joint effort by Dr. Jennifer Shang, an associate professor of business management in the Joseph M. Katz Graduate School of Business at the University of Pittsburgh, along with colleagues Shanling Li of McGill University and Sandra Slaughter of the Georgia Institute of Technology. It looked at the software industry from 1995-2007.

We divided it into three categories: research and development, marketing and operations, Shang said, while emphasizing the last one. Software companies depend on innovation from research in order to differentiate themselves from others. Then once developed, obviously marketing is critical to the company. What we found out is a company's operating ability, how they manage their business, is even more important to their survival.

Operating capabilities often come down to one thing: managing finances. Shang said the most successful software companies were able to invest the right amount into each area of the company, and not put everything into research. Furthermore, they were able to handle their costs efficiently. While this is easier to do in a boom period, such as the late 1990s and early 2000s, it becomes significantly harder when there is a bust.

We call it the munificent times: everyone survives it well, the stock runs sky high and you get lots of money flowing around from venture capitalists. The real test comes when there is not that much capital, how do you survive it, Shang said.

According to Shang, the industry has a bankruptcy 15.9 percent, which is more than most other industries. The pharmaceutical industry, for instance, has a bankruptcy rate of 4.7 percent.

Shang and her colleagues collected data from 870 software firms over the 13 year period. Of the 870 companies, two stood out to Shang as perfect examples of the boom and bust nature of industry, and why operating expenses are important.

Peregrine Systems was a company that busted, while Epicor is one that has thrived. Peregrine from 1999-2001 grew so fast and acquired a lot of companies. They introduced a lot of new products. But they didn't invest in their management. They boomed so fast, they didn't have the operational capabilities to manage the company well. They proved expanding quickly does not always pay off, Shang said.

Eventually, after filing for Chapter 11 bankruptcy in 2002 and laying off half of its workforce, the company was bought up by Hewlett-Packard. Epicor, by contrast, an enterprise software firm, was able to grow steadily and invest equally in research and development as well as operations and marketing. They had the slow and steady approach, Shang said.

The study, which was titled, Why Do Software Firms Fail? Capabilities, Competitive Actions, and Firm Survival in the Software Industry From 1995 to 2007, was published in the journal Information Systems Research.