Should you be chasing standards with your chips or playing the proprietary game? This question goes to the heart of your company business plan, and you need a well defined answer. It probably seems like the answer should always be to stay on the "standards" approved side of the chip design, but that has not always been the case. Traditionally, one company pioneered a new technology and got an early lead in the market. The late-to-the-party competitors immediately got together and formed a committee to draft a standard that negated the first-mover advantage of the pioneer. It was easy to separate competitors in those days … the winner was the one making all the money and the losers were on the committee.
Of course there are good and valid reasons for having standards, and with the complexity of things like multimedia formats, standards are now a practical necessity.

Who can afford to create a chip with millions of gates hoping that the company can then drive the market by itself? Silicon Image created the HDMI interface for HighDef television, and then found it didn't have the marketing muscle to drive that interface into the market alone. Ultimately it was forced to open the HDMI spec to others in order to overcome the natural reluctance of the market to a completely proprietary interface. Their gamble, in this case, was that even after opening up the spec and making it an open standard, they could leverage the pioneering work they had done to maintain a technical lead. Only time will tell if they were right.

Some of these standards attempt very broad coverage that encompasses multiple levels of performance and feature sets. H.264, for instance, has a variety of "profiles" each targeting a different screen size and resolution. It is not enough for a chip to be H.264 compliant - it must also state which profiles it implements. Over time, some of those profiles will command major market shares and others will slip into obscurity. Examples like this prove that there are many opportunities for new chips that are both standards compliant but still assemble those standards and profiles in ways at will be attractive to the market.

So what is your strategy in this standards confusion? First, we should say there is nothing wrong with being the market leader and driving a de-facto standard into the market. If you can do it - if you have the market power - this can be a winning and highly profitable strategy. But even if you go the proprietary route and win, there will be a standards committee in your future as competitors get on board to endorse and expand your work. Regardless of whether you're the first mover that started the whole thing, or a later entry, if this is a key product/technology for you, you HAVE to be on that committee. If you're the technical originator, you have to be on the committee to keep from being victimized by your competitors. If you're a later entrant, you have to be on the committee to make sure it does not move in a direction that negates all your chip development you've already done. If you're new to the market, it's the place where you'll find out what your competitors are doing technically in this product direction, and by extension, if what you're doing is going to be a strong product offering or not. In short, you HAVE to be on that committee.

In many cases, these standards committees are formed and driven by some sort of trade association or alliance. Companies interested in the technology band together to share information and push the technology by creating the standard. There is usually a fee - not insignificant - to join these alliances, but it's money well spent, and there is frequently some sort of lower-cost affiliate membership available as well. The key is that it is not just the chip companies in the alliance -- there will be some major potential customers there too. After all, it's in their interest to know and influence the direction of that technology as well as keep track of the players. By being part of the alliance, you have an opportunity to meet the major customers on a monthly or quarterly basis and demonstrate your knowledge and progress. When the early major deals go down, they will generally be between the customers and the chip providers in that alliance, so you definitely want to be part of it. As standards are developed and agreements reached, there will be press activity generated by the alliance and, as a member, you'll be included. This is an easy way to get recognition as a leader in the technology.

Organizationally, I usually drive this activity out of marketing, either the technical marketing or the strategic marketing groups. I find that typically my client companies will be in three or four of these alliance / standards committees, but even then it is not a full-time job. Still, it is important to recognize that the primary reason for your company to be a member is not technical but business. You're there to get competitive information, influence directions, and work early customer adopters. The lobbying that goes on in these groups can rival presidential electioneering, and you're going to need a pro at it. Many of the meetings are best characterized as babysitting -- just tracking activity rather than heavy technical discussions. So I have marketing act as the main contact point, and then make sure the company provides technical resources on an as-needed basis.

Submitted by contributing analyst, Dave Guzeman.  He is the president of MindPik, www.mindpik.com.