How to pitch new silicon to a system company: knowing the timing and the process are key ingredients for success.

Raul Perez
7 min readJul 23, 2020


Since I’ve been on both sides of the table when it comes to buying/prospecting or selling/pitching silicon components, I want to put down some thoughts about how things work out timing and process wise such that the buyer is most receptive to the pitch.

Some key steps are:

  • Understand the product release date. There are many companies that release their products at a known time of the year. For some it is Christmas time, for others it’s more dependent on a season of the year. Yet others are in regulated markets, and they are less sensitive to any particular consumer seasonal behavior. Whatever it may be, you need to figure out your target system company’s product release schedule.
  • Understand the development cycle, and how much time there is between builds. Every system company is trying to converge to a final shippable candidate design, and they will have a series of builds with different objectives. Typically these builds are named Proto, EVT, DVT and PVT. The time to input new silicon is preferably during the Proto builds, and up to the EVT build. So when a system company is most likely to accept pitches from silicon component suppliers is when they are planning their Proto builds. The more complex the silicon, the more difficult it will be to win a socket, and the more advanced in the build schedule the less likely there is a chance that the system company will design in new silicon. Probably the worst case scenario is when a company is selling processors (or some other complex component), since these tend to impact everything in the system: software development, firmware development, EE, PCB routing and subsystem control. So a supplier selling processors needs to be really early pitching to be able to get a design win for a system. In contrast, a supplier that sells diodes or fets, has more time to design in their components, in the case of diodes even late EVT builds could be a possibility to design in a diode. The timing between builds matters in cases where you need to spin silicon again. You need to ensure that you will be able to intercept a late EVT build with final silicon. Sometimes, if the silicon changes are minor, there could be a silicon revision targeted for the DVT build, but this is not advisable in general.
  • Understand what is the process the system company uses to engage with suppliers like you. Usually for simple components such as two pin parts, the company will have a GSM (global supply manager) in charge of those from the sourcing perspective, and they will work with EEs to see if they are willing to add these components to the BOM and test them during the builds. However, for custom silicon the system companies may have a different engagement process that involves a silicon manager following a process such as the one implements. In these types of engagements there are complex interactions between supplier and system teams, and the level of dedicated support is much higher on both sides. It is very important to understand exactly how each system company engages with suppliers in order to plan effectively and have a good chance of success.
  • Understand what is the impact of designing your silicon into the system company. A two pin part such as a diode is fairly straightforward to test and only requires EE sign off. A part such as a processor is very complex, and therefore will require multiple teams to sign off such as the software team, firmware team, EE team, thermal team, signal integrity team, possibly the ME team and other sub-system teams that will be feeding data to the processor or be controlled in some way by it. So as you can see, you need to understand how difficult it is to design your silicon into the system so that you can plan ahead and provide the system company with the types of tools and support needed to make their life easier and be able to do it at the right time during their development cycle.
  • Preserve your resources for real opportunities, and focus them early on them. Typically the GSMs are in the Operations team at a system company, and part of their job is to continuously reduce the BOM cost even while the product is in development and after it launches. As discussed above, it isn’t practical to design-in some types of silicon products that are more complex after the first EVT, but the GSMs need to keep negotiating price down whichever way they can. So you may end up being the useful loser! What I mean is that there is the possibility that a GSM keeps you engaged after the engineering team has already selected a different supplier, but they keep you engaged to maintain some credible leverage against your competitors. This is actually pretty common practice, and you need to be mindful of how much resource you invest in this opportunity after you know for a fact that the last proto build has been completed. It could be that you get lucky and your competitor has a catastrophic fallout with the system company at EVT or even after, and then they switch to you, but it is very unlikely that will happen because the objective of the system company is to ship on schedule and use their resources efficiently. It’s possible that a silicon supplier’s sales team focuses too much on a bigger system company’s opportunity after there is no realistic way they can win that design, and loose other smaller opportunities in which they actually had the right timing and had a real chance of winning but got distracted just because volume was larger for the bigger system company’s opportunity.

There are more considerations that come to mind, but I hope the list above at least provides a quick way to help silicon suppliers do a basic check for each opportunity on their list, and helps you prioritize accordingly. It is always in everyone’s best interest in our electronics ecosystem that system and silicon suppliers have the most efficient deployments of their products so we can continue earning a living and making great products for the consumer.

— — — — — — — — — — — — — — — — –

At we specialize in managing custom silicon engagements with our battle tested process. Custom system silicon when done with the assistance of silicon experts puts the system company in control of its own destiny. Hiring silicon experts full time at your company may not be reasonable due to insufficient continuous work for them, and that is why provides you a solution so that you can engage with chip suppliers on custom silicon programs and mitigate all the risks associated with it.

Developing custom silicon can have huge benefits from an economic, engineering and market perspective for system companies, but it takes a structured and detailed approach to ensure proper take off and a successful landing. Don’t hesitate to contact us at for any further questions, or help you may require.


  1. Hardware engineers speak in code: EVT,DVT,PVT Decoded by Anna-Katrina Shedletsky
  2. How to keep your leverage while single sourcing custom system silicon by Raul Perez

Originally published at

Other related articles: