A friend of mine just finished designing a system using the Raspberry Pi computer board. This is a pretty cool, low-cost system... one of several now on the market.
I was looking it over and observed that it includes a couple of analog I/O ports: audio and composite analog video (selectable PAL/NTSC). And... that's it. There are add-on boards available that include "general-purpose" analog I/O and a lot of people are using the USB port to connect to audio I/O subsystems of various types. It got me thinking about how people view the analog world.
A couple of years ago I was involved in a consulting project that involved segmenting the analog IC market. For those of you who have not made the transition from engineering to marketing, "segmentation" is the practice of separating the potential customers for products or services into identifiable "segments" with common language, needs, and so forth. Identifying market segments is the first and most critical part of developing a product marketing strategy, since everything else follows from it. It is mostly common sense, but some companies are a lot better at it than others.
A big part of segmentation is defining what customers care about, and tailoring the products and marketing of those products to serve those needs. Sure, everyone cares about cost, but focusing on low cost as the only possible customer benefit is a tough way to live. More profit is generated if you can identify a key customer need for which customers are willing to pay a premium… preferably a need that can be filled at a lower incremental cost than the premium that can be extracted.
The client for this segmentation exercise had already started the work based on their understanding of the analog IC market. It was a major, multi-national, vertically integrated company, with end-equipment divisions that made almost everything from consumer electronics to medical equipment to industrial machinery. They figured that their corporate organization was a decent proxy for the entire electronics market, which was pretty clever. Most of their subsidiaries bought a lot of the silicon that the semiconductor division produced, which was of course the whole reason for the vertical integration in the first place. Now the client wanted to improve the profitability of the semiconductor division.
But their "built-in market" turned out to be the source of their biggest blind spot in the whole segmentation exercise.
Their consumer electronics divisions consumed the majority of the semiconductor division's analog IC output. And naturally, those high-volume markets revolved around the processing of analog and video signals. The obvious segmentation that the client arrived at was that the analog IC market could be divided into segments defined as "audio," "video," and "other." There were other sort-of analog niches labeled "RF" and "power," but those were handled separately.
It did not take long to figure out that one of the problems with the profitability of the semiconductor division was that the focus on high-volume audio and video chips for their barely-profitable consumer divisions ultimately led to product lines that were also barely-profitable. They never really got into analyzing what all the other (more-profitable) divisions were using for analog ICs, and what those segments needed. I told them that they needed to figure out what the profitable divisions were buying from other analog IC suppliers, find out why they were more appealing than in-house ICs, and find big enough common needs (segments) to define some new products for them.
And if there weren't any product ideas that met their business models for volumes, testing, support, etc., then they needed to focus on either changing their business model or figure out how to reduce their costs if their main emphasis was going to be on consumer ICs, and a race to the bottom on price.
The analog IC business is hard.
Have you been in this situation?