Letting the cat out of the bag
Pre-announcing products is an easy way of keeping your market informed - but it all has pitfalls.
Apple chief Steve Jobs has a catchphrase he likes to use in his regular keynote speeches to developers and customers at its conferences. Everybody knows there will probably be "one more thing" at the end.
It's the announcement that everybody has been waiting for. It's the point at which the famously secretive company reveals a product that may not ship for a while, but something that Apple wants to be sure gets a lot of attention before it does go on sale.
Both models of the iPhone got their "one more thing" moment in a keynote, both well before the product went on sale. In the case of the original iPhone, Jobs said it was coming at the beginning of January 2007. It did not go on sale until the end of June, almost six months later. With the iPhone 3G, the gap was narrower - less than a month.
At the end of the 1990s, Jobs launched Mac OS X well ahead of its public availability. And the company's switch to Intel was just one more thing that would take months to complete.
Apple has a reputation for not signalling its plans to the world. For a good many products in the company's range, the first sign of a change in the line-up is when the online shop temporarily pulls down the shutters as it gets a refit. Very often, this is what happens for 'speed bumps', when the company upgrades the specification of a computer to take account of new chips. Before these products go online, Apple maintains tight secrecy about them, helping to seal its reputation as a company that does not pre-announce.
Yet, Jobs' carefully scripted off-the-cuff revelations - some well in advance, others made as the product rolls into shops - show Apple as a company that pays close attention to what works and what does not in product launches. Apple will cheerfully use pre-announcement if there is some advantage to be gained from it. The iPod Shuffle was "one more thing" but was on sale the same day.
Although common in the automotive business and in aerospace, no market sees more pre-announcements in terms of volume than computer and electronics technology.
According to research performed by Barry Bayus, Sanjay Jain and Ambar Rao for the Journal of Marketing Research in 2001, some 50 per cent of products are pre-launched in some way.
At Microsoft, the pre-launch can last for years. Long before it mutated into Vista, the world's number-one software company was talking about what was likely to go into Longhorn, the codename for the project. It was perhaps a foolhardy move, as many of the features that the company claimed it would have in the final shipping version fell off the list.
Researchers such as Jehoshuah Eliashberg of the Wharton School of the University of Pennsylvania, Meng Su of Beijing University and Cornell University's Vithala Rao regard company size and market dominance as indicators of conservatism in pre-announcement. Such companies do not have to use the tool and do not have to run the risk of being accused of trying to sell vapourware. But Microsoft had other reasons for being so early to promote Longhorn.
Originally, Longhorn was meant to ship by the end of 2005, just three years after the release of Windows XP to stores. It was pushed back to 2006. And then it slipped back even further to 2007. To placate PC makers, Microsoft issued vouchers that would let customers buy a PC for Christmas 2006 with XP installed but which let them upgrade once Vista finally shipped.
One of the Longhorn casualties was Windows Future Storage (WinFS), a thorough overhaul of Windows' file system. And it was perhaps the most pre-announced feature of all time, dating back to the early 1990s.
Those with long memories will recall Cairo, the code name for a version of Windows NT - at that point known as version 5.0, had the Object File System that would replace the clunky old filesystem inherited from MS-DOS, itself carried over from old CP/M conventions.
By the time NT 4.0's replacement, Windows 2000, had appeared, the Cairo filesystem had been stripped out and pushed back into the to-do pile. And it was the same for WinFS: dropped from Vista because it was proving too tricky to finish.
Was Microsoft wrong to trail Longhorn so early?
In a paper written in late 2005 for the Zyman Institute of Brand Science (ZIBS), Alina Sorescu, assistant professor of marketing at Texas A&M University, and colleagues cited an important reason why companies should pre-announce: wrong-foot the competition.
Pre-announcement was a technique that IBM used aggressively at its height in combination with FUD: fear, uncertainty and doubt. The company made much of its image as a safe bet - nobody ever got fired for buying IBM, the urban myth went. If a faster competitor had a type of product ready, IBM would pre-announce something that it was working on in the expectation that customers would wait that little bit longer to get their hands on the market leader's product.
Getting developers on board
Pre-announcement can work in favour of companies trying to build a position against an incumbent. Without the weapon of pre-announcement, an old technology might build up unstoppable momentum because customers are not aware that they may have an alternative before allowing themselves to be locked-in to a particular product or technology. In markets where the switching cost is high, such as electronic design tools, where a lot of information may be stored in proprietary file formats, it can make sense to pre-announce to offer the prospect of an alternative.
In the case of Vista, deterring competitors was probably a secondary benefit. With a commanding market share and the position of default operating systems on PC-class hardware - something that has only recently changed with the defection of Apple to Intel processors and the growth of a market for pre-installed Linux - Microsoft had little need in the mid-2000s to put customers off buying from competitors.
In technology, pre-announcement has a further benefit: getting developers onboard. In many respects, the biggest competition to Vista continues to come from the operating systems' predecessor XP.
The thing that Microsoft typically has on its side is its foresight in making sure external developers are fired up over the technologies in each new version of the operating system. The prevalence of Windows means that there are plenty of people out there who want to stay one step ahead of their competition and so are generally keen to keep an eye on what Microsoft plans to launch. Microsoft keeps them happy by supplying plenty of pre-release software nice and early to look at and play with. Much of the pre-launch information about Vista concentrated on the new capabilities to which the developers would have access. Those developers would help promote Vista by writing applications to show off the operating system.
This process of technology evangelism is not unusual to Microsoft. Apple is pretty good at the same thing. But it is an important factor when considering the real business impact of each operating system launch by Microsoft. The company can tell you this stuff is great and that your lives will be enriched by the experience. But having a bunch of nominally independent developers make the pitch is a lot more powerful and it is easy to get carried along with their enthusiasm for features that, in all honesty, are good for developers but have a lot less effect on how an end-user will see the system or what payback the business will see from using a new technology.
In the case of the Xbox, pre-announcement proved successful, according to Wybe Popma and colleagues from the Rotterdam School of Management at Erasmus University. In doing so, the company broke one of the rules of pre-announcement: that it is something reserved for market leaders. The console was launched 18 months ahead of its actual introduction in 2001. But there was a good reason for doing a pre-announcement, even though Microsoft had no track record in producing games consoles and was faced with stiff competition from Nintendo and Sony at that time.
"Early announcement was intended to ensure software support for the new console," wrote Popma and colleagues.
Software supporting represents a significant cost to developers and, although they are not strictly customers of a game console, it was important to have them onboard as close to actual launch as possible.
In general, Eliashberg and Wharton colleague Thomas Robertson found in the late-1980s that customer switching costs and learning requirements correlate with the likelihood of a company making a pre-announcement.
The Vista launch process helps demonstrates where things can go wrong: if a firm cannot deliver on what it promised, its reputation will suffer. For this reason, some companies construct an image of not pre-announcing. When he was CEO of storage company EMC, Michael Ruettgers told Harvard Business Review in 2001: "One way we have gained...credibility is in our refusal to follow the common practice of pre-announcing products months in advance of their release - the vapourware phenomenon, designed to get customers to hold off planned purchases of competing products."
Palm shows its hand
Declaring its hand too early has tripped up handheld computer maker Palm twice. The first time, recorded by Popma and colleagues, was in March 2001 when Palm was wrong-footed by the competition: "Handspring, Palm's main competitor in the personal digital assistant (PDA) market, announced the Handspring Visor Edge. Palm, as the market leader, feared for its innovative reputation and responded by pre-announcing (ahead of plan) the Palm m500 one week later, months before actual introduction. In the time between the announcement of the m500 and its market introduction, demand for the existing high-end model, the Palm Vx, dropped dramatically.
"Customers stopped buying the 'old' Palm Vx in anticipation of the new m500. Distributors were left with oversupplies of the old product. Prices of the Palm Vx had to be cut. This, in turn, resulted in a price war with Handspring... in combination with a general slowing of demand in the PDA market. Both firms lost over 80 per cent of their stock market value in the following six months."
The second time Palm fell foul of pre-announcement was when Palm and Handspring founder Jeff Hawkins stood on stage at the D conference in May 2007 and demonstrated what was then the forthcoming Foleo product. As he talked about the lightweight subnotebook computer, he declared: "Foleo represents a whole new product direction for both Palm and for mobile computing... in my opinion, the Foleo is going to be the most successful and the most significant product that Palm has done."
Just three months later, Palm killed the product off. It never made it to the stores. It was a product that went from concept to manufacture to termination before anyone had a chance to buy one.
Palm president and CEO Ed Colligan explained in a letter to developers and potential customers that a lack of resources to complete the Foleo doomed the project rather than a lukewarm reception - journalists and analysts were quick to declare the product a non-starter. The company would pour its development resources into a new generation of smartphones able to run Linux which likely would not ship before 2009.
The Foleo's cancellation represented a major blow to the company's image that was entirely avoidable given that, had Hawkins not pre-announced the product, the project could have been killed off privately long before a launch decision had to be made.
Broken promises seem to hit share price. A 1997 analysis performed by Kevin Hendricks and Vinod Singhal found that product launch delays cut, on average, the market price of a company by 5.25 per cent.
Is pre-announcement worth it? Sorescu and colleagues claim: "Despite the pitfalls of pre-announcements, academic research has, in general, viewed them as valuable signalling tools... Although these signals are theorised to convey favourable information about the firm's future prospects, there is little or no empirical evidence that they translate into financial returns for stockholders."
So Sorescu's team set about trying to work out whether pre-announcement strategy worked in terms of share price. The researchers collected a sample of 419 pre-announcements from from the period 1984 to 2000. Only 37 of them had formal introduction announcements; the bulk were revealed informally to journalists working on technology magazines. The average time from pre-announcement to actual introduction was close
to six months.
From their analysis, which attempted to control for long-term movements in price that result from momentum and to adjust for other shifts in valuation, the researchers concluded that pre-announcements led, on average, to a long-term uplift of 13 per cent in share price. The effect could last for a year in the case of pre-announcements followed by a long gap.
Sorescu's team conclude: "Our results also reveal that the more specific the content of a pre-announcement, the higher the stock returns in the short-term. Furthermore, updating investors after the pre-announcement leads to higher stock returns in the long-term.
"However, this finding does not mean that firms should provide updates at any cost, even with unreliable information, because doing so can damage their reputations and have a detrimental effect on future market values."
Pre-announcement 'reliability' comes into play: firms with a good track record saw a more positive reaction from shareholders. "Once a firm is known to have produced vapourware, positive financial returns to its future pre-announcements may not materialise. Therefore, while vapourware can deter entry in the short-run, it might be detrimental to firm value in the long-run," the A&M researchers claimed.
In short, don't use pre- announcement to dig yourself out of a hole if an earlier tactic did not pay off. Although pre-announcement has proved an effective weapon, it can just as easily go off in your hands.