Poor alignment strategy is a common failure mode in many businesses. Chip Making is a very interesting industry to study with respect to this, as much can be learned. Parts of it are very commoditized and other parts are very solution-oriented. Moreover, this has changed with time. In its early days, semiconductors were merely building blocks of systems. This is why they are often referred to as components. VLSI changed this dramatically. Very Large Scale Integration occurred at the point where one-hundred-thousand transistors could be integrated into a single chip. At this level, it was possible to build entire systems on a chip. When this transition occurred in the early eighties, some semiconductor companies recognized its impact and began to align sales and marketing, while others did not.
Intel is a great example of a company who did recognize the change. In aligning sales, marketing, and manufacturing; Andy Grove was able to execute brilliantly in the microprocessor market. Intel had lost this market to Zilog in the late seventies and it was alignment that allowed them to take it back. In contrast, Motorola did not recognize this need for change and would falter badly over the next two decades. My belief is that it was mostly the result of major management changes at the time. Jack Welty, a very dynamic leader, retired from heading Motorola’s components group. The group continued to execute to his vision for some time, instead of developing their own based on evolving needs. Texas Instruments is a company that has often been maligned for a company with bad marketing. Instead, I place the blame for its past problems on misalignments between sales and marketing, as well as failures to restructure sales for evolving market needs. Their marketing accurately targeted major opportunities, including surface mount packaging, DSP and its relevance to cell phones. In the case of surface mount, marketing put together great educational marketing material that guided customers through the complexity of this technology. Without these materials, surface mount would have been put behind by years, if not decades. But TI’s sales force remained in a commodity-based structure. They failed to align to marketing by not converting to a solution-based structure. The result was that sales never monetized the development. Learning from this experience, TI did align in the case of DSP, which allowed them to succeed. Another good example comes from National Semiconductor, which restructured from commodity selling to solution sales and marketing under Gil Amelio to address the emerging mixed signal market. They quickly shot to number one in this market in the early nineties as a result.
Another interesting thing about semiconductors is how process and product development do not have significant need for alignment. If you look at Intel or any well-organized chip maker, the people who develop process are more closely aligned to manufacturing than they are to product development. They approach process development from a background of clocking Moore’s Law. Their perspective is to consistently deliver nodes of the clock on time and in a very manufacturable state. Product development is closely aligned to manufacturing separately through Intel’s intense internal DFM effort. When Intel got into trouble in 2005, it was not because this process was broken as many believed. It was because its product development was not aligned through sales to the customer. It had the most leading edge process technology, but not the right architecture — leaving a hole for AMD to fill. The lesson to learn is that even the masters have misalignment problems. It is one of the most difficult things to execute on, while at the same time, it is one of the most overlooked.
