The semiconductor industry survived the recent rough recession. In fact, the industry finally learned from past mistakes by turning off the supply spigot faster and more aggressively than in previous downturns as demand plummeted.
This was the good news. The bad news is that many companies in the industry continue to struggle with a wide range of issues including excessively narrow focus on engineering chips rather than deeply understanding end markets now and going forward, inefficient global operations, and insufficient innovation.
To overcome these hurdles, capitalize on the market upturn and build momentum and growth, what should industry leaders do now? This article tackles that question head on. Drawing upon several of Accenture's industry research initiatives and associated analysis and insights, as well as deep industry experience meeting with executives throughout the industry constantly, this report recommends nine growth opportunities that will help semiconductor companies accelerate towards high performance.
Seven growth opportunities
One: Focus on end markets
In traditional semiconductor end markets, growth depends on winning the "fight for sockets" and gaining a greater share of the silicon in specific end market devices. Critical to winning this battle is having a much stronger understanding than now resides in most semiconductor companies of customers' reference designs including innovation channels, consumer behavior, and usability and form factor revolutionary modifications for future new products. Chip firms need to use this understanding to provide chips that meet future needs. This, in turn, depends on having effective relationships with direct customers, as well as distributors to gain better customer intelligence and a rich understanding of where innovation is headed in a three-to-five-year time frame.
In this pursuit, chip firms need to gather insights into the end market to understand the specific economics, the cyclical nature of demand, and the unique nuances of the market. Two areas are critical. First is understanding end market usage models via data analytics. This means grasping how usage and movement of technology, such as smartphones and e-tablets as a convergence of key forms and functions, is progressing within a specific market. Second is comprehending end market value chains, meaning what additional insights and expertise are required to succeed in developed and emerging end markets that result in added-value to direct and indirect customers. The goal is for chip firms to become value-add design partners versus volume component chip manufacturers.
David Patterson, known for his pioneering research that led to RAID, clusters and more, is part of a team at UC Berkeley that recently made its RISC-V processor architecture an open source hardware offering. We talk with Patterson and one of his colleagues behind the effort about the opportunities they see, what new kinds of designs they hope to enable and what it means for today’s commercial processor giants such as Intel, ARM and Imagination Technologies.