The year 2010 has been both profitable and turbulent for the semiconductor industry. As companies across the retail, communications and consumer technology industries build new products with embedded technology and greater programmability, the software-defined nature of devices is changing the usefulness and product life-cycle expectations for semiconductor components. This change requires much deeper understanding by semiconductor designers and manufacturers of shifting customer needs. As such, the end user – the customer – has more power than ever before. To achieve high performance, semiconductor companies must lend a more attentive ear and more sharply focus their attention on their newly empowered bosses. Designing chips, and all the engineering that goes into that, is not nearly enough to achieve high performance in this industry.
In this intensified focus on customers, chip firms need to gather insights into the end market to understand the specific economics, cyclical nature of demand, and its unique nuances. Two areas are critical. First, is understanding end market usage models via channel and data analytics. This means grasping how usage and movement of technology, such as smart phones and e-tablets as a convergence of key forms and functions, are 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 markets that result in added-value to direct and indirect customers. The goal is for chip firms to become value-add design and supply partners versus volume component chip manufacturers.
Beyond becoming value add partners, semiconductor companies must move from being engineering-centric to product-centric. What does this mean? Whereas engineering-focused companies concentrate on a roadmap for technology and specific step-by-step improvement in technical performance, product-focused companies concern themselves more with creating platforms to match uses and interests of their ethnographically diverse customers. Leading firms incorporate this process within their strategic long range product roadmaps with a 3-5 year view.
The transformation from an engineering-centric model to an end-market model calls for a new set of skills. There are five key areas where significant change is required to create the necessary integration and responsiveness in the value chain:
• market and customer segmentation; • engineering; • product lifecycle and release management; • supply chain and supplier management; and, • sales operations and deal management. Market and Customer Segmentation
Semiconductor firms must develop a more pragmatic assessment of the market dynamics in which they operate and target for growth. This requires thoughtful segmentation of markets, customers and forecasted demand. The inaccuracy of demand forecasts causes havoc on the sales and operations planning of costly capacity. It requires understanding differences in geographic segments and end-product usage behaviors, and tracking micro-trends in various technologies with advanced market trend intelligence from the channel and emerging markets. Traditional insights must be augmented with knowledge of buyer attitudes, perceptions, needs and values by customer and geographic segment.
Guys, i agree with oldness of story but we shouldnt forget, author has right, now its not the case that R&D fellows at end companies sit in the lab and dig for best compatible product with help of booklets, technical spec. texts etc. If we just check out carrier pages of semi companies, there are many newly named positions else than regular ones(FAE,FSE), now we have PSM, PME etc. As a semi company you should know by heart where your product will be used, how u can make it different than others where design-ins are on. Even if the customer is unwilling to include your engineer for support, you should be able to understand needs perfectly.
For Sup.Cha. i totally agree with author, how many semi.companies are running their own sup.chain? We still have distis in between. Still in case of shortage times customers are getting in trouble cause of distis greediness and semi.cos unwillingness for punctual outcome, just check out what happenned at 2010.
There are 2 types of end-customers at todays electronic market! 1s, who are involved in design of their high-end chips as like apple, sony, samsung, high capacity of R&D and budget to run many projects simultaneously. This type of customers are directing semi co.s for their needs. There isnt such need for highlevel of product&design support. Others who has R&D but in need of gudiance and support while design ins are in place.
This is fact, future's brilliant semi.cos wont deal with production facilities, neither there wont be so many foundries to lead shrinkage of chips. JVs, Team-Ups will be more and more common. Perhaps we wont see lots of SoCs or product precific chips. Usage of same chip with different application is beyond dispute, just check inside of apple's new product family.
I have a guess about future business model of semicon. industry. The winner of ARM vs. Intel battle will shape the next. What do u think?
Hmmm, this is very old story, I agree with @VincePG...I worked in the IC design space for several years and it was always about end customers not chips...on some occasions we would even contact customers of our customers to make sure we understand final product use...and complex reference design, drivers etc was part of the sale process over 10 years ago...Kris
This was news about 15 years ago. Complex chips have been sold with ever more complex reference design and system level software requirements for quite some time. It has now become so complex that if you do not have a system level partner to work the system integration angle you will never get designed in.