A variety of market research firms recently released current statistics and future predictions for the computer industry, and the warhorse Mac OS gets mixed-to-negative marks for the future - depending on who you ask and what you ask.
While most U.S. computer makers are expected to post fairly healthy profits for the fourth calendar quarter of 1996, all analysts agree that PC sales, especially in the U.S. and Europe, are lower than expected. IBM and Hewlett Packard are expected to do fine, being large, global, diversified companies, and Compaq is recovering nicely from last year's bad inventory management and weak first half. Apple, on the other hand, posted a $120 million loss for the same quarter, but says the shortfall is almost entirely due to lack of PowerBooks to sell and lack of U.S. buyers for Performas.
Regarding this news, an Oppenheimer & Co. analyst told The Wall Street Journal that Apple's operating system is "out of gas" and that people are picking Windows over the Macintosh because there are more Windows titles available - and further, Apple will begin to lose ground in the education market (with no data or reasons to back that assertion).
International Data Corporation (IDC) released early estimates for 1996 operating system shipments. The company pointed out that Windows 95 fell short of projected units, and that "many corporate users delayed migrating to the newer operating systems," meaning Windows 95 and Windows NT. Not that the gains weren't spectacular - Windows 95 was responsible for 63 percent of all worldwide OS units in 1996, but that was still 9.3 percent less than projected. Windows NT grew 303 percent, which was still 32 percent short of expectations. IDC's "market observations" said "Apple clones have yet to translate into increased market share," and "Apple must introduce a fully multitasking operating system which is highly compatible with its current Mac OS and recapture its technology leadership in order to improve its position. Apple's recent announcement of the acquisition of NeXT Software and its plans to incorporate NeXT technology in future operating systems does not fully address this need."
Given all this negativity in IDC's report, how bad were Mac OS sales actually? Down from 6.8 percent to 6.6 percent of all units shipped. Compare this to OS/2, which lost nearly half its market share in the same period.
Mac OS Market Share -- Part of the problem is conflicting definitions. Most of the doom-and-gloom reports in the media about falling market share refer specifically to Apple Computer. Since 1995, that doesn't tell the story of the Mac OS market, because companies other than Apple now sell Mac OS computers. But since that's a recent development, there's a real tendency to think that Macintosh market share and Mac OS market share are the same thing. They aren't.
Market share, for the uninitiated, is the percentage of all new sales in a given category that belong to one particular company. To be technically correct, market share must be measured in a given time period, but it's usually referred to in the present tense as an estimate of what a company's sales are compared to its competitors right now. IDC's report said that 6.6 percent of all personal computer operating system sales in 1996 were Mac OS purchases, down from 6.8 percent in 1995. It's interesting to note that this figure does not include upgrades, only new licenses to new computer owners - so it's a way of measuring the market share of the hardware capable of running Mac OS.
Two-tenths of a percent decline isn't very much - by IDC's numbers, we're talking about a drop of about 150,000 units on yearly volume of five million. In 1995, IDC says there were 4.5 million Mac OS licenses sold, so how can five million be a decline? The operating system market is growing, that's how! Mac OS sales grew too, but not as fast as the other market segments did. To stay at 6.6 percent of sales, Mac OS shipments should have reached 5.15 million. They didn't, hence the drop in market share concurrent with a growth in overall sales.
As long as IDC's statistical certainty is greater than 0.2 percent (and it probably is, although it's not a given), Mac OS sales were statistically flat in 1996. Yet press reports, and Apple Computer itself, continually refer to declining sales, not flat or slower-growth sales.
Send In the Clones -- Mac OS clone makers account for some of the gains, and most likely for some of the lost sales Apple experienced. A NEWS.COM story looked at the big four clone makers - Daystar Digital, Motorola, Power Computing, and UMAX - to find out what makes each of them tick. As part of the report, NEWS.COM says that IDC's competitor, Dataquest, estimates that Mac OS clones comprised 8.5 percent of U.S. Macintosh market share during the third quarter of 1996.
Dataquest's number isn't valid for the entire year, obviously, but it makes things more interesting. If Apple's U.S. market share in the fourth quarter was about 7.3 percent, as has been estimated by one source, then adding other Mac OS sales to the mix raises overall Mac OS market share to 7.9 percent. That's still far short of the 13.2 percent Apple had a year earlier, but few people can be blamed, in the current press climate, for being skittish about buying a Mac. The total market share is probably higher than 7.9 percent - I cheated and applied third-quarter clone numbers to fourth-quarter Apple numbers. Clone numbers were likely to be higher in the last quarter due to the arrival of machines from Motorola (and APS), which do not seem to have cannibalized other Mac OS purchases.
Motorola is known to have shipped at least 40,000 StarMax clones in its first eight weeks of production, and a Motorola marketing executive told NEWS.COM that the company "suspects" a quarter of the buyers are first-time Mac OS purchasers. Motorola is staying out of the retail market so far because they initially got a late start, but now they don't have the facilities to handle the volume of sales they expect a retail unit would generate.
UMAX, on the other hand, shipped 100,000 units in the last six months of 1996 - numbers for Power Computing aren't available but could, according to a previous Tim Bajarin estimate, top 500,000 units in 1996. DayStar Digital's numbers are smaller, in the 3,000-unit area, because they make high-end systems with fat profit margins - the four-processor 200 MHz 604e box is for a specialized market, especially at $10,000 apiece.
[Note: Apple just released some additional information on clone sales, giving Power Computing credit for more than 100,000 sales during its first year. -Adam]
Both UMAX and Motorola believe they can achieve 10 percent of the Mac OS market by the year 2000. This is good news for Apple if and only if they achieve this by expanding the Mac OS market. UMAX in particular is committed to this - with a parent company located in Taiwan, UMAX believes that Asian markets are right for Mac OS technology, and they're in on the ground floor. The company told NEWS.COM they expect to increase Mac OS sales in Taiwan by 300 percent, and by even more in areas like Southeast Asia and parts of China. That's exactly what Apple needs to hear, and if all clone makers pull off similar market expansions, it will have been worth weathering the early years of cloning when clone sales are eating away at Apple's own market share.
What Does It Mean? What no market research firm has yet released are continuing studies where Mac OS market share is tracked, on a hardware level, separately from Macintosh (Apple Computer's) market share. When OS sales are used as a benchmark, IDC's numbers show nearly no change in market share from 1995 to 1996, despite the absolute beating Apple took in the press and in consumer confidence, month after month. That's a reasonably worthy achievement. IDC phrases it as "Mac OS clones have yet to increase market share," but given that everyone believed Mac OS market share was falling, it's not bad at all.
It's also a good idea to remember that the non-Mac OS market is not unified. According to the eighth Computer Industry Almanac, there are about 25 million Macintosh machines out there as of Q3 1996 (Apple says 26 million), about 180 million DOS users, 130 million Windows 3.x users and 53 million users of Windows 95/NT. Each of these operating systems has a slightly different programmer interface - code written for Windows 3.x will run in a kind of emulation under Windows NT, but the reverse isn't true. If Apple's plans (depending on who you ask) to release Rhapsody for non-PowerPC hardware pan out, developers could find themselves with an easy way to write a Mac OS program and have it available on all these newer Windows machines as well. That would undoubtedly bring more developers to the Mac OS platform, and more software for Mac OS means more sales, according to IDC's theories.
Clearly Apple has to get their act in gear - without a good Mac OS to license, clone sales will eventually fall by the wayside no matter how aggressive the offerings are. Individual Mac OS clone makers are too small to show on the market research radar scope, so stories tend to focus on Apple - whose losses for the fourth quarter resulted from a combination of tepid marketing, less shelf space, overall lower-than-expected PC sales, and a crisis in consumer confidence. Yet the clone makers are here, and are doing well (Power Computing turned a profit in its first full quarter), and it shows in the sales numbers, although you sometimes must dig a little to find the not-bad news.
[This article is reprinted and updated with permission from MDJ, a daily Macintosh publication covering news, products, and events in the Macintosh world. If you can't get enough insightful Mac news, sign up for a trial subscription to MDJ. For TidBITS readers who want to subscribe, there's a special limited-time rate of $11.95 per month (20 percent off). For more information, visit the MDJ Web site.]<http://www.gcsf.com/tidbits.html>