Now is the time for start-ups and corporate nanoelectronics R&D programmes to stop talking about how cool the technology is, forget the hackneyed quips about the next big thing being very, very small and come up with some real business plans. That’s the scary part. Understanding value chains, building business models ... it’s all a very inexact science. Mistakes can cost millions. As an industry analyst covering the photonics sector during the rise and fall of the telecoms industry, I am convinced that the critical input to business plans for products based on novel technologies is realistic market forecasting. During the optical boom, I saw firms swarming with brilliant minds talk themselves into believing that the markets they were chasing were worth billions of dollars. They built business models based on this assumption, and then they crashed and burned.
Had they built those business models around the truth - that their addressable markets were actually worth hundreds or even tens of millions of dollars - many of these firms would be with us now. Let’s hope that those working with today’s hot nanoelectronics technologies can avoid this kind of hubris. All they need to do is answer a few simple questions: Where will those cool nano-enabled features really matter to customers? How much will they matter? Can these features be used to create entirely new species of product? And how are the answers to all these questions likely to change over time?
Answering these questions is the first step in coming up with a useful market forecast. Typically it will reveal the hard truth that the addressable markets for a new nano-enabled electronics or semiconductor product will be niche-like. MRAM chips, for example, are inexpensive enough to serve as replacements for battery backed-up static random access memory (SRAM) or to replace both an SRAM and Flash chip where both are used together. Yet they are likely to remain too expensive to attack the mainstream Flash and RAM markets for some time to come.
It’s also important to get timeframes right. On the demand side, for example, some of the advances in high-speed processing using carbon nanotubes or molecular nanostructures have been impressive, but they will not be needed for a few more years. On the supply side, OLEDs promise displays for mobile phones and personal digital assistants (PDAs) that consume only small amounts of power yet provide high quality video. If this promise can be kept then it will instantly solve the most difficult problem that mobile computing and communications faces - having the power to support the features required without running down the battery in a short amount of time. But it is still just a promise - today OLEDs use about the same amount of power as alternative technologies.
None of this should be taken to imply that the first high-tech giants to be born in the 21st century will not be nanoelectronics firms. Perhaps some spintronics company will strike it rich with a low-cost quantum computing platform and launch an entirely novel direction for computing. Someone might come up with a nano-enabled fuel cell or battery with the performance capable of solving the mobile power problem mentioned above. Indeed, while there is always a danger that some nanotech firms will follow the same overoptimistic path to ruin as many of the photonics firms before them, I also sense that others are underestimating just how revolutionary some of this stuff really is and how big it could be in terms of revenues.
As a result, caught between the overoptimistic and the overpessimistic, forecasting of nanotech markets can be frustrating. Only this week, I was told by one well-informed industry insider that my projections for field emission displays were preposterously high and by another equally well-informed industry insider that they were unconscionably low.
Forecasting of nano-enabled products is a business that doesn’t get you much love. But it’s a necessary one.