Please feel free to forward this newsletter to as many friends as possible. I have a personal goal of having one million readers. I can achieve it with your help.

the Software View: Software market dynamics (Part I)

Welcome back, gentle reader. Frederic P. Brooks, Jr., in his book, The Mythical Man-Month, writes, "The programmer, like the poet, works only slightly removed from pure thought-stuff. He builds his castles in the air, from air, creating by exertion of the imagination. Few media of creation are so flexible, so easy to polish and rework, so readily capable of realizing grand conceptual structures ...

Yet the program construct, unlike the poet's words, is real in the sense that it moves and works, producing visible outputs separate from the construct itself. It prints results, draws pictures, produces sounds, moves arms. The magic of myth and legend has come true in our time. One types the correct incantation on a keyboard, and a display screen comes to life, showing things that never were nor could be."

For those of you with Internet and World Wide Web access and a Netscape Navigator web browser, please point your browser to:
http://www.softwareview.com/
Scroll down the page and you will spy a link entitled, "Current views web log (fresh news daily! Click reload, clear the browser cache)" The daily news page is also known as a "web log". It is en vogue and the fashion of these days to call it that. Click on the link, click "reload" on your browser or clear your browser cache to ensure that you always receive the freshest, hottest daily news concerning Java, Linux, XML, and the software industry! The link never changes, but I will be updating the HTML file page behind it every day. Please, do take a gander at it every day.

Also, some important news, gentle readers. the Software View is now an Associate Internet World Wide Web site of Amazon.com. I would like to extend my sincere, heartfelt gratitude and thanks for your patronage. I am offering links to books, et cetera that you can purchase from my web site. I would greatly appreciate it if you would purchase software industry books from my web site. Help support my newsletter and web site by purchasing items from Amazon.com from my web site. Here is the URL (Uniform Resource Locator):
Click here

Now, dear readers, on with this week's episode of the Software View!

With this issue, I begin a multi-part exploration of the dynamics of software markets.

PRICING AND DOMINANT MARKET SHARE

J. William Gurley writes, "From an economic perspective the software industry is quite unique. Never in the history of the industrialized world has a single industry offered more powerful economies of scale. The majority of the product costs are fixed (research and development). Therefore, per unit costs fall quickly and dramatically as unit volumes increase. In this world of potentially unlimited scale, market share is critically important. If a company can achieve greater than 50 percent market share, and is willing to lower price, it can become somewhat invincible. Microsoft prices its products assuming it will take greater than a 50% share. Even though they have a great brand, and even though they make great products, they still use price as a weapon. And it does not stop there. In addition to using price to gain share, Microsoft uses price to protect share. If a software company is willing to lower price far enough, it will discourage others from entering the market. The required research and development costs will simply be too high relative to the potential pay-off. Price can be both the means and the end.

There are two types of vulnerability. In the first category are those companies that mistakenly find comfort in fragmented product sectors. Second, are those companies that have majority market share, but fail to price aggressively enough to deter market entry.

We speculate that intentionally lowering price is a particularly unsettling strategy for most executives. This is understandable. Using price as a weapon in industries with more typical variable costs as a percentage of sales can be remarkably dangerous. Take for instance the personal computer industry. With 20 percent gross margins, a 5 percent reduction in price can lead to a 25 percent reduction in gross profits. However, with a software gross margin of 90 percent, a 5 percent reduction in price only leads to a 5.5 percent reduction in gross profits. The PC vendor needs an unrealistically high price elasticity of 6.7, just to make up for the price erosion. However, the software vendor can break-even with a price elasticity slightly above 1.0 - a much more realistic assumption.

We should note that there are two software companies outside of Microsoft who seem to have a more thorough appreciation for software economics. Both Intuit and Netscape have majority market share, and have used price aggressively to obtain and maintain their current positions. Of course, Microsoft has responded to this situation by lowering price to the unsustainable level of free. This may be an early signal of impending doom for the software industry. If multiple players recognize that the lowest price and the greatest market share wins, standard prices could deteriorate to uncomfortably low levels. Switching costs, true product differentiation, and capital availability will be the only things that limit the destruction brought about by continuous product giveaways.

We leave you today with a general warning for software industry executives. If you are in the software industry and do not have dominant market share in your category, you should be nervous. If you are in the software industry, do not have dominant market share, and Microsoft is entering your category, you should call a board meeting and consider drastic alterations to your current business plan."

THE SOFTWARE ECONOMY

Ted Lewis writes, "Although the thought of boosting production to increase supply, drive down costs, and mainstream a product wouldn't play to Keynesians, this scenario may be a high-tech company's salvation in the Information Age. Driven by the Internet, the new economy requires a drastic change in business models for those who want to be major players.

The amorphous Information Age, where bits reign over atoms, information superhighways replace asphalt, and software standards mean more to Wall Street than price/earnings ratios. Already, it is causing high-tech companies to clamor for real estate in cyberspace. It has boosted the Nasdaq into the stratosphere, and has driven the popular press into a lemming-like frenzy.

Some valuable lessons in how to do business in the Internet Age. It's part of a new business reality that I call "software economics." While not yet formalized by Harvard economists, an understanding of these new economic principles is the secret to success, both for the current Information Age and for the future Internet Age. It's the lubricating force of the coming software economy.

Software economics does adhere to certain principles: driving prices to commodity levels, setting standards at nanosecond intervals, targeting special interest groups, identifying shooting ranges (and shooting), and appealing to tribalism. These techniques are already being successfully used by such companies as Intel Corporation, Netscape Communications Corporation and Qualcomm Incorporated.

As the Internet barges into society, all businesses will have to learn the principles of frictionless capitalism. We are moving beyond the Industrial Age, even beyond the Post-industrial Age, into the Information Age, where the rules of economics are stood on their head. In the age of the Internet, software rules and inverse economics make products continually better, yet cheaper. The time is ripe for a new model that companies can use to guide their strategies.

The characteristics of this model include: applying learning curve theory to mainstream a product, using the New Lanchester Strategy to plan and execute strategies and following "market-of-one" principles. Companies planning to play in the coming software economy -- I call them FutureBusinesses -- won't survive without mastery of these concepts.

MONOPOLY GAMESMANSHIP

Microsoft's great success in the desktop personal computer world started with a simple idea: the more market share you have, the more you get. In other words, the rich get richer. The trick is to get started before everyone else. The result is a positive feedback mechanism that accelerates revenue growth. For example, Gates explains how Microsoft beat out two other operating systems when the IBM personal computer first appeared. Microsoft ignored the cost of development, packaging and marketing, and emphasized the accelerated revenue growth of a large market. Thus, Gates adopted "achieve a monopoly" gamesmanship, and priced MS-DOS as if it had already captured the majority of IBM PC buyers. Gates' competitors, by contrast, were aiming for "pay-as-you-go" marketing and charged higher prices.

Now for the positive feedback mechanism. As MS-DOS became widespread, it became more entrenched. Microsoft added coal to the engine when it built applications on top of MS-DOS. Money made from licensing it was poured back into enhancing its value, allowing Microsoft to move more rapidly down the learning curve and cement its hold on the industry. The more market share Microsoft got, the cheaper its products became, and the more people wanted them. The result of this powerful feedback force was the mainstreaming of MS-DOS."

I call this concept and particular scenario, situation a "virtuous cycle". The more users an operating system possesses, the more it compels independent software vendors to create applications for that operating system. The larger the quantity of compelling software applications an operating system possesses, the more users are attracted to it.

Ted Lewis writes, "This is a classic example of a learning curve. The idea originated in the 1930's to explain efficiencies in airplane manufacturing. Aerospace engineers noticed a 30 percent decline in production cost when production volume doubled. Even in those days mass-production technology was good, but it wasn't good enough to explain those results.

The slide-rule gang soon realized the generality of this phenomenon. While some improvement comes from pouring in more resources and working longer and harder, working smarter also has its rewards. Ingenuity, experience and training -- in other words, learning -- also contribute to a decline in production costs. Learning curves measure ingenuity.

Today, the learning curves theory is used in the semiconductor industry with resounding success. It has been quantified by Gordon Moore, Chairman Emeritus of Intel. Moore's Law predicts that the semiconductor industry learns enough to double processor performance every 18 months. Or, if you hold processor performance constant, Moore's Law becomes a declining-price model. Here, the industry is learning how to reduce chip prices by 48 percent per year.

Compare that to the learning rate of the software industry. There, the rate of declining production costs lags that of the hardware industry by a significant amount. Software learning rates can be expressed as a declining cost per function point (a measure of the number of inputs, outputs and internal and external files manipulated by the program). With that measure, we find that software producers learn at the significantly more sluggish rate of approximately 4.5 percent per year.

MAINSTREAMING

A forecasting technique that uses learning curve data to model the effects of business strategies on technology-driven businesses. This model makes the assumption that slipping down a learning curve results in greater market share, which in turn feeds even greater gains in market share -- in other words, positive feedback. I call this the "mainstreaming effect," which leads to market domination by a mainstream product, service or standard. This is also known as a "virtuous cycle".

In a nutshell, mainstream curves are shaped by a formula that is roughly the inverse of the learning rate of a learning curve. A higher learning rate results in faster mainstreaming. Thus, a particular product or service reaches market penetration rates approaching 100 percent more quickly when the learning rate is higher. Prices drop more rapidly, which leads to more consumption, which leads to greater market penetration.

In general, we can forecast the growth of markets and predict when they will mature by plugging learning curves into the mainstream model. Compare the "progress" of hardware and software technologies using this mainstreaming model. Hardware undergoes a rapid maturation (as determined by market penetration) in the 30-year span between the years 1990 and 2020. Software will not begin its rapid ascent until the last half of the next century. That is, not unless it achieves some breakthrough comparable to that of VLSI technology. Some have suggested that techniques such as object-oriented programming, componentware and visual programming technologies could dramatically improve the learning curve of software technology.

The lessons of the powerful positive feedback of frictionless capitalism, and the corresponding mainstream effect, dictate that Information Age companies learn at a higher rate than their competitors. Success in the new economy depends on how fast a company slides down various learning curves, thereby moving more quickly toward mainstreaming than their competitors. It is the basis of the coming software economy.

Intel propels itself down the slippery slopes of Moore's Law by religiously applying Davidow's Law (be the first to render your own products obsolete) to its microprocessor business. Netscape evolves Web standards while the standards committee ponders the next version of HTML. Other companies, such as Adobe Systems Inc., Mountain View, Calif., and Symantec Corp., Cupertino, Calif., stay out front through fleet-of-foot mergers and acquisitions. These are variants of learning curves. In practical terms, the results catapult a product, service or standard into the mainstream. If we believe in the positive feedback forces of frictionless capitalism, once a product achieves mainstream status, it is nearly impossible to dislodge.

GIVE IT AWAY, NOW

In case you haven't noticed, the software economy is distinctly nonlinear, non-Newtonian, and particularly non-Keynesian. Subscribers to non-Keynesian market forces -- which I call FutureBusinesses -- quickly rise to positions of leadership within their segments of the new economy.

These companies make idealistic assumptions that might seem absurd to a Stanford MBA. For example, what business school taught Netscape and Qualcomm how to build a business on the basis of free software products? Netscape Navigator freeware propelled the company to 70 percent market share. Eudora, Qualcomm's e-mail client, is arguably the most popular freeware package in existence. Sun Microsystems Inc., Mountain View, allows individual, private software developers to freely download its most important technological contribution of the decade, Java^TM. By giving away the Java Development Kit, Sun is trying to make a play for ubiquity, to build market share, and to mainstream Java.

How can Netscape and Sun get away with this? In the FutureBusiness model, demand effortlessly follows production. In a sense, the non-Keynesian software economics assumes zero production and distribution costs, and seeks to mainstream rather than to balance supply and demand. If production and distribution are free, why conserve supply? This leads to a radically different result. A company either mainstreams or dies.

These strategies bounce Keynesian economics on its head. In essence, Keynesian economics has production dogging demand. The manufacturer initially estimates demand, sets a price, and then produces the product or service. At some later time, market feedback is used to adjust demand estimates. If the warehouse is overstocked, the perception is that demand has diminished, so the price is lowered to clear out the inventory, and production is curtailed to decrease the supply. Conversely, if demand is great, manufacturing and prices are stepped up to balance supply with demand.

This approach works, but it takes time. Manufacturers must wait for demand to rise or fall before adjusting production. This Machine Age delay introduces uncertainty into old-style markets.

The FutureBusiness, on the other hand, optimistically applies the learning curve rule. Production is boosted to double supply and lower the price. That drives up demand (assuming demand varies inversely with price). There is no looking back: Production is constantly forced upward until market saturation and mainstreaming is achieved.

While demand and price work together in the old model, they work in opposite directions in the new model. This is called inverse economics, and is a principle of the Information Age. The goal is to apply inverse economics as fast as you can, produce as cheaply as you can, and mainstream by winning or buying market share at light speed.

Netscape applied this technique in the extreme by giving away its Web browser to buy mainstream status. "Free" is a very compelling price. By plunging the Web browser learning curve into the basement, Netscape Navigator hit the rooftops overnight.

How can Sun Microsystems, which gives away its core product, the Java Development Kit, ever hope to make a profit on Java? Well, the answer is that it probably won't ever make a huge profit on Java per se. One unyielding characteristic of the upcoming Internet software economy is that consumers expect everything to be free. Leaders of the upcoming Internet software economy must give away their core product (software, content, etc.) and then try to recoup revenue via peripheral products (like the official Sun series of Java books) or services (like official Sun certified Java training courses).

Now, all FutureBusinesses are stampeding to the Internet to give away their products! A recent op-ed piece in an industry publication suggested that the next step is to pay consumers to use a new product. Once hooked, of course, the consumer is levied upgrade fees.

But commodity pricing is just one of several principles that FutureBusinesses will have to follow to compete in the software economy. They will have to use information technology to drive out cost and follow the learning curve model. Some companies can also help maintain market share through branding (such as the "Intel Inside" campaign).

Next is the principle of inverse economics. Future businesses must follow Davidow's Law. Companies that set the pace of obsolescence also lead the pack.

Another principle is that of the "market-of-one." FutureBusinesses must try to narrowcast, rather than broadcast, their marketing messages to special interest groups. The mass society is dead. Mass production, mass marketing and mass media fail when they do not recognize niche markets.

Finally, we will see the principle of tribalism. Macintosh users, for example, belong to a different tribe than Windows users. Products that achieve a certain cult status also achieve market dominance.

These principles are on the loose across the Internet. The obvious idea of using the Internet to pummel prices is already out of Pandora's Box. The abundance of free software, free product information, and other kinds of free content are clear indicators that mainstreaming is alive and well on the Internet.

But price pummeling is easy. Refining the other principles will be more difficult. Finding market-of-one opportunities and using narrowcasting techniques to serve those highly individualized and personalized markets, for example, are still immature art forms. Furthermore, Netheads hold a deep-seated desire to return to agrarianism, tribalism or both, and these urges will be exploited by astute FutureBusinesses. But these new companies are only just now beginning to learn to use chat rooms (virtual communities), avatars, electronic barter and targeted marketing to appeal to innate tribalism."

To be continued ...

Sincerely,
Mark Kuharich

Join my free e-mail newsletter called the Software View by clicking here or sending mail to thesoftwareview-owner@west-point.org