I heard an interesting metric today. A friend of mine had just gotten back from Taiwan where he was looking for a contract manufacturer for his clever new device. He is an old mobile hand, having been in consumer electronics for a couple decades.

He told me that he was surprised to find a very high level of interest from the biggest design manufacturers or ODMs.  He had been dealing with these companies for years, but his current product is his own venture,  and he only expected to make a few thousand units. Typically, the biggest ODMs only get interested when units are measured in the millions. He had originally anticipated working with one of the smaller, Tier 3 ODMs, but his contacts at the larger firms encouraged him to meet with them.

He came back and told me everyone he spoke with only had one metric they cared about for their own business – their percent of “non-PC” revenue. Metrics are very important, and as an analyst I always paid attention to what terminology company management used when describing their own business.  This idea was a large part of Clayton Christensen work’s in the “The Innovator’s Dilemma”. And the best example in recent memory was the realization several years ago that Apple was not pursuing market share for the iPhone, but was instead focused on “profit share”. There are countless other examples. Companies with their own fabs or chip manufacturing lines tend to focus first and foremost on capacity utilization because it has such a big impact on their reported profitability.  Other companies fall into the trap of focusing too much on market share, or one hard technological problem, or worst of all, their day-to-day stock price.  I have found that most companies tend to use a basket of metrics since there is usually no single, all-encompassing figure that truly captures the health or future prospects of their business .

For companies in Taiwan’s electronics manufacturing supply chain, business is hyper-competitive with very slim margins. So in the past, I find that they cared about a lot of metrics  – new design wins, operational efficiency, material costs and gross margins – to name a few. Now admittedly, my friend’s travel tales are not data, they are just one anecdote. Nonetheless, I suspect that he is not far off the mark; the fact that he heard the same metric so many times certainly points to that. So for the sake of argument, let’s assume he is right and that there is a big concern in Taiwan for new revenue sources and getting away from PCs as quickly as possible.

Obviously the demise of the PC is not news. This has been in the press and blogosphere for years. However, those constituencies, as well as Street investors, tend to inflate the seriousness and immediacy of big changes. “PCs unit shipments to decline 2% in Q3” is not as exciting a headline as “PC Industry in terminal decline”. In my work, I have always found visiting Greater China to be an important antidote to the Valley’s echo chamber. It is one thing to write a blog post, it is quite another to run a mammoth manufacturing enterprise on that same basis. They supply chain tends to take a very ‘grounded-in-reality’ approach to operations. This is not always to their benefit, but I like hearing both sides of perception.

So I think there is more to this ‘data’ than just another headline in on an ongoing theme.  If, in fact, we are seeing the Taiwanese complex seriously disengage from building PCs, it will have important implications for the whole industry. As much as everyone in the Valley (and I am as guilty of this as anyone) has been prophesying the end of the PC era, PC sales are still very large. However, if the ODMs begin to disinvest in their PC production, we will likely see an acceleration of PC declines. The PC complex is built on steady increases in volume, and optimized for economies of scale. If the best minds and most-skilled engineers at the ODMs turn their efforts to other products, this will no longer hold true. At  the very least, we will see a change in PC price dynamics. Prices have been roughly flat for a few years, but there have still been improvements in features. This will likely stall as well.

I do not want to pile on the gloom. And I do not anticipate some shocking, cliff-like decline in PC sales any time soon. Nonetheless, I do think we are on the verge of a sea change in industry dynamics.

Perhaps the biggest factor preventing a sudden shift is the fact that it is not clear where the ODMs can turn. The solution is not thousands of small-batch, niche products. Or if it is, then there will be a lot of pain as manufacturing lines undergo serious adjustments. At CES, everyone was talking about wearables and the Internet of Things (IoT), but no one is yet clear what that means or what products will work.

Nor should we assume that the pain of industry transition will be felt solely by the Taiwanese ODMs. They could very well find a new industry to attack and transfer the pain to someone else. For instance, I know of one major consumer electronics company, dissatisfied with their PC and smartphone prospects, who are now turning to automotive electronics. This company has the scale and resources to make a serious effort here. They also bring with them the lessons learned in building products that consumers like, in a disruptive fashion. For companies in staid industries with significant electronic content, be forewarned, there could be a lot of excess manufacturing capacity in Asia coming towards you.