When "the next big thing" isn't

We're in an industry that is defined by change. We've learned to live with it and profit from it, even if we don't like it, it unsettles us, or it changes our business model.

One of the most significant changes that we have seen is the increasing role of consumer electronics in our industry—from ICs through end products and software. This trend is a distinct shift from the industrial, instrumentation, commercial, and military/aerospace applications that traditionally dominated and defined our products and applications. Now, we need extremely high volumes to support the incredibly expensive fabs that implement the progression of Moore's Law. Or, is it the other way around, and we need the costly fabs to provide the volume of components? It's hard to say which comes first, because both factors—chip demand and fab capacity—tend to drive each other in a recursive, positive-feedback, reinforcing manner.

How much of the electronics business depends on mass-market consumer products? It's hard to say, because there are so many legitimate ways to define and measure it, as well as inherent data inaccuracies. You can use component dollars, final-product dollars, number of units of either of those, allocation of R&D efforts, and more. The numbers I have seen put the consumer part of our industry at 30 to 50% of the total business, up from around 10 to 20% just 20 years ago.

With this change in industry focus comes a change in business strategy, for both established vendors and newer players. Everyone is looking for the next big thing, to ride the wave of success that will drive volume, yield enormous returns on investment, lock up a market for a few years, and magically solve all sorts of problems for the winners. Being on the right side, at the right time, of this "killer app" is the objective.

There are short- and long-term downsides to this quest. It is impossible for 100 companies to each get 10% of the market, and, when you bet big, you can lose big, too. This situation contrasts sharply with the climate just a few years ago, when companies would boast about diverse customer bases and note that no single customer or application accounted for more than a tiny fraction of their businesses. Now, they proudly note the opposite: that relatively few big customers are significant portions of their revenue. Of course, that's true only until a few of those key accounts go elsewhere. When the next big thing doesn't happen or it happens with someone else, a company's fortunes may change dramatically.

There's another side effect that should concern you. The application-engineering departments at most vendors are usually key resources in helping users successfully implement and complete solid, respectable designs. As vendors focus on that handful of key customers representing the next big thing, they have less incentive to provide support to the individually smaller, but broader, potential-user base. In fact, for many application-specific ICs, there are only a few viable customers. (After all, how many OEMs specialize in integrated disk-drive controllers and read/write channels?) So, some application departments are really auxiliary design teams for key customers. This approach might make business sense, but it could hinder the long-term development of industry engineering talent, depth, and versatility.

Contact me at bschweber@edn.com.