Jed Hurwitz

Jed Hurwitz reveals what he wished he'd known when starting up his business.

I entered the semiconductor field slightly by accident, graduating with a general electronics degree; I faced a choice of career paths as diverse as power stations to microelectronics. I went with "chips" partly because a friend was offered a job by the same company, Plessey Semiconductors, and partly because I thought it would be fun. I wish I could say that my decision was more informed, but I really did not know that much about it - sometimes you just have to go with your instincts.

"Invest in people, identify their potential and make them feel that they are part of the success."

I was fortunate in my first job, to find job, to myself attracted into CMOS analog design and to have a boss, Ian Philips, who taught me two good lessons that I try to carry through, in every endeavour that I enter into. Firstly, technology can be applied to anything to "improve" its function; it is just a question of informed imagination - you just need to identify what are the actual problems that need to be solved in order to make a better product. Secondly, invest in people, identify their potential and make them feel that they are part of the success - financial reward is not the only factor; external recognition, training and responsibility are often equally, if not more important. However, make sure you do it by consultation and occasionally by surprise - it is much more motivating than if they have to ask. I cannot re-iterate too much, how important it is to be actively involved in the future of your team members; if it is done right they will repay you 10 times over.

When you build a team, it is vital that you get a good mixture of skills and backgrounds. There needs to be people with industrial experience as well as those from academia. For a product to reach market, it is not about making one near perfect demonstrator, it is about making millions reliably and with high yield. People who have done it before are invaluable. I take a sharp intake of breath if I hear that a start-up consists of 9 PhD's and 1 non PhD. You should also make sure your team consists of a mixture of innovators and finishers, commercialising technology is not just about the ideas, it is also about precision execution, if you're extremely fortunate you'll have some people who are able to do both. Finally there needs to be a commercial and financial balance to your team, unfortunately many technologists feel they do not need these people... yet. But they are wrong. You need specialists in these areas to provide balance and ensure your plans are reasonable.

"A classic mistake is to assume your technology is the centre of the world."

This leads me to my next theme - it is really important to analyse the real value of your technology and the market opportunity properly, don't just fit the technology to the "best" market, but try to analyse what are the application fundamentals, what are the real trends, who the real stakeholders are and why. A classic mistake is to assume your technology is the centre of the world. You need to understand what role it plays, what the alternative is (often the incumbent solution) and most importantly, what things you can do to enhance your value. Sometimes you need to take a step back and be critical of your own solution. Don't just look at what aspects it improves, but look at what are its weaknesses, or what might change to enhance or weaken your position.

There is a natural tendency to be blinkered by your own technology and your own experiences of the alternative approaches, you need to balance your enthusiasm for your approach, remember your competitors and customers are capable people too, and the best technology does not always win if the other factors - commercial, manufacturability and strategic - are not satisfied. Customers will not easily back themselves into using a "proprietary" solution if there is not a sufficient difference in price/performance, as they generally have more to gain from a multiple source strategy. This can be frustrating, but this is the reality, and you may need to consider a second source or licence strategy to eventually address this concern and ensure there is a big enough market.

Just as it is important to release your technology for the right reasons, to the right partners, at the right time, you also need to be careful about developing NIH (Not Invented Here) syndrome. If it is necessary to adopt another approach you should be brave enough to do it.

Another tough decision when building a technology company is to try not to kill today for tomorrow, OR tomorrow for today. In general, you will start as an R&D centric organisation, and then when you get the first market traction you will move into a mode where you try to maximise the return from the first generation of your product, but please ensure that you are already planning to design it out with your second generation. If the market is big enough, and you don't design yourself out, somebody else will.

"Although you need to make sure your plans are aggressive; they must also be realistic and achievable."

You will be judged by your investors and your customers on your execution - the toughest thing to do is to deliver a project on time, in budget and to the customer's requirements. If your plans are too optimistic or have too much fat, they will not be achievable or survive diligence. There are three things to watch out for:

• Being too optimistic about the time to build your team, and the overheads of getting them functional and performing. Include milestones that are measurable and that de-risk the project, but also motivate the team, and provide evidence of progress to the outside world.
• Not having sufficient time to validate or stabilise the product before launching it to many customers, will cause you problems and may damage your reputation - that will take much more effort to recover. So it is best to restrict the early customers to the minimum strategic needed, and those who are already familiar with the technology or are stakeholders somehow in the technology.
• Relying on the hockey stick sales curves. It is in everyone's plans, but I have never seen it actually happen. Customers don't always move at the pace you want, and often there are seasonal effects to the markets or customer design cycles that you need to respect. Make sure you know what your contingency plans are. If one month kills you, it is not a good plan.

Finally, I'll touch on the question of finance, to get financing there are four elements your proposal must have:

• There needs to be a significant market, and that market should be growing rather than declining.
• There needs to be something unique about your offering that can be turned into something sustainable, and can be mapped into the fundamental needs of the target markets.
• There needs to be good people in the organisation, both technical and management, ideally with track records of success.
• There needs to be a plan for a product that the market wants that is 'just mad enough' - if it is too easy you will not be able to answer the question of why can't the competition do it, if it is too difficult you will not be able to justify that you can do it.

Even then, there are many other hurdles to cross. You will be funded to keep you lean and mean until you prove that you can manage the money and the project successfully. Make sure that you communicate effectively with your investors, and work at maximising your value while managing the risk and it will work out well, and they will have to ensure that they maximise the potential in the next round of funding.

I hope that by now I have not disillusioned you. I should have stressed that although it is hard work and at times very stressful, being part of a technology start-up is hugely rewarding. You get to attempt to change the world, to make the seemingly impossible become possible, you get to build a team and a technology from scratch, and hopefully you will get the unexplainable pleasure from seeing your "baby" reach the shelves (and get significant financial reward). I can vouch that after 20 years of being involved in silicon technologies, and the last 12 years of being involved in start-ups, it is still fun and I think I made the right decision, all that time ago.