Thanks for your feedback. I reached the very similar conclusion that you outlined: the lack of VC funding needs to be compensated by other creative funding sources. The strategic partners appear as the most likely target. In my mind, it is the completion of the 'circle of innovation'. Let me explain briefly in the context of EDA. No VC means no new technology, so the customers (semis) are starved for innovation, the smart ones will start investing strategically in companies that provide the new technologies they need to create their products.
The flipside is that semi companies sometimes choose to invest into internal development rather than fund promising start-ups. It is happening right now in some of the major semis. It is the traditional 'make vs buy' turned into make internally vs make externally'.
I think the issue is not availability of fund, issue is understanding semi/ EDA business and creating a sustainable RoI model. Major reasons why a VC does not fund in semi/ EDA start-up
1. High fund requirement: May not be true for EDA or IP strat-up, but true for fabless strat-up
2. Longer return time: Any semi/ EDA company takes at least 2 - 3 years (for fabless it is more) to strat earning decent revenue. VCs whose typical investment horizon is 3 - 5 years this is a long duration
3. Absence of tangible asset: The asset created by semi/ EDA company is intangible in nature, if the company is not able to earn revenue or not able to see enough customer traction the exit value is very small
4. Lack of understanding: Good number of VCs does not understand the high tech market very well and they are reluctant to invest
Let us see how these problems can be addressed
1. Creation of strategic investors: A strategic investor is a person/ company who is not only investing in the start-up, but also is a potential customer/ acquirer of the product. In that scenario, his return from the investment goes beyond the cash he received by selling his equity in the start-up. A start-up in semi/ EDA will always look for such investor. Also the big companies who are hungry for new technology to survive create a dedicated fund for start-ups whose technology is relevant to his line of business
2. Equity in exchange of services/ tools: A start-up needs to drive down the cost and should be ready to give up some equity in exchange of services/ tools he needs. For example an IP company can give some equity to an EDA company to get access to the EDA tool. A fabless company can negotiate with a design services company to get design services in exchange of equity. Off course EDA and design services company should also be prepared for that
Agreed: there are two ways EDA can expand beyond its present mature 'perimeter'. First, get into new markets: software development is an often mentioned candidate; new applications such as wearable devices - as you suggest - might be another opportunity. The other key is revamping the business model: EDA 's market size is 1-2% of the semiconductor industry! Clearly EDA is not extracting value from its technology. Cloud offerings might help; more solutions oriented delivery (srevices?) have worked in some cases.
@Michel: I appreciate your candor and the follow up. EDA market in my opinion has attained a high level of maturity and there are only few new product options (other than IP). I think the EDA industry, in order to keep innovative enterprises thriving, has to offer many more solutions than the current generation legacy ones. This is easier said than done but the evolving trends are clearly visible -to cite a few examples, wearable devices, products made with molded interconnect devices (which can be printed with 3D printers), etc. Some of these newer markets that can be easily addressed by EDA companies are not done so because the product pricing and positioning conflicts with existing models.
MP - Fully agreed about the bleak VC funding environment for semi/EDA. I commented about this in a previous blog, 'Improving the EDA funding environment' (7/23/2013). I believe that for EDA the early stages have to covered by angels and 'sweat equity' from the founding team. The process I propose applies to the angels interaction.
The crowdfunding is very interesting but appears to be catching on when the funder is a potential consumer or has some other affinity with the product. This crowd is limited with semi/EDA and reduces itself mostly to angels who have been involved in the industry - either as producers or consumers.
The article is entirely VC-centric and from what I gather, it does not look too encouraging for new hardware startups. On the contrary, look at what is happening in the wearables market including Quantified Self -an majority of the startups in this area have been crowdfunded. Perhaps the author can comment on the subtle and/or direct differences in how a startup team can pitch to crowdfunding sources?
As a marketer by training, who is not an engineer but has an eye for engineering, a concentrated electronics industry has left the guild of engineers with few tenured specialists in marketing, business management and economics.
Concentration has also left primarily three types of business opportunities: hardware design within large incumbents, among incumbent compliments, or for innovators chasing emerging platform standards.
It may be difficult but family obligation can make it practical to accept servitude, at a major, working out one's innovation's responding to query by in house attorney 'the product work is only capable of that specified within the contract assignment'. Never stop learning.
Enabling new competitive platforms where compliments can attach around open infrastructure to create subordinate economic effects has become much more difficult under incumbent restraints. However, where new enterprise compute and communication platforms are emerging, to take from Intel and Cisco, software development offers a vastly unlimited business opportunity to grow new component and hardware system businesses.
There is a burning need and the primary hardware springboard is production economics. Without scale innovations and their innovators remain boutique developments. No one will supply the innovative substitute unless it can be manufactured in volume. Yes it must be whole with hardware and software standards that make sense because they are broadly applicable. Still among this incumbent crowd, 88,000 units produced in 2012 growing to 1.2 million by 2016 does not make an industry and will not sustain a competitive business.
Minimally multiple by 100 for a standard platform's market.
Breaking through this engineered myopia adds competition and intellectual property law, industrial management, cybernetics, system and value theory, psychology, team building, network dynamics, market relations and responsibility in technocracy. Modern engineering curriculum and enterprise often neglect some of these important components for contributing to the sustainable business. Whether incumbent or start up.
With only so many seats at the table, job security a preference, communications sometimes limited by reference engineers tend to hire other engineers. This can leave marketing and management professionals out where diversity for research, market, business, product plans and customer approach can add exponentially to the effort.
For the group dynamic does 1 + 1 = 4? Do As hire A+'s? Do B's hire Cs? How limited or broad are our views? How conservative or innovative are our approaches? Do we look beyond the box that we ourselves and the system paint; for warrant? Do we modify our techniques? How well do we define questions that will deliver real answers for consensus in our structured decision making?
Intelligence is a prerequisite that comes from many quarters and 'in the know' one of its most sustainable competitive advantages. To know where to look, where not to look, to start, to continue, to amplify, to escalate, to expand, to change, too put on hold, to look somewhere else and even to start again.
Diversity of information in the process of identifying whole products worth engineering includes addressing not only the tangible benefits, but intangible merits perceived across many minds of the market.
There is more than one market and there are many minds. Intellectually how to connect with them requires engineering that is more than engineers. Who are mass adopters? What is the profile of economic buyers? Where will leverage be gained with fast followers? How can mass and leverage grow the business from a process of constituent assembly well ordered?
Taking a day from the bench to visit the foundation of use markets and their channels offers a basis for delivering unique and differentiated products. The knowledge and support network to commercialize them. Where software system integration smells the money there's likely a market. Frequently this does not mean platform validation at the large incumbent's customer sites, but developing channels left out of their game to establish a new game.
Once the new game is running on coordinated advance, constantly reconfiguring the competitive status quo can secure challenger's ability to lead, for position, sustainable advantage and a viable business.
As you point out, the key to maintaining leadership is to create a culture of innovation. Add to this retention and motivation of the good people.
The VC's were - past tense since there are few left in semi - making a bet on one chip: get it right and you make a lot of money, acquisition or IPO ensues and the VC's gets out. So the VC's were making a bet there would be a large market for the chip 2-3 years (the IC design cycle) ahead. Tough call for a concept that is really innovative and requires at least a $20M investment.
Semi business model is inherently scalable, the main caveat to this would be if a heavy dose of services is required for 'design-in' of the new chip.
My Mom the Radio Star Max MaxfieldPost a comment I've said it before and I'll say it again -- it's a funny old world when you come to think about it. Last Friday lunchtime, for example, I received an email from Tim Levell, the editor for ...
A Book For All Reasons Bernard Cole3 comments Robert Oshana's recent book "Software Engineering for Embedded Systems (Newnes/Elsevier)," written and edited with Mark Kraeling, is a 'book for all reasons.' At almost 1,200 pages, it ...