This model is broken for all sorts of reasons. The big EDA companies just don’t have the stock valuations and the cash to make acquisitions in the $100Ms. As a result, almost no VC will fund a new EDA company or even put much more money into one that they already have on their hands. Plus, the slow adoption of the most advanced technology and falling design starts make it impossible to justify that type of valuation.

On the other hand, it remains really easy to start an EDA company. Find a problem, get together a few engineers who really understand it, and write some code. If you successfully solve a key problem for the leading edge semiconductor companies, they will buy your product.

Some things work in your favor. Part of the reason that the VC model isn’t working well for EDA is that it isn’t working well in a lot of segments: it just doesn’t cost enough to require VC-level investment. Computers are cheap. Cloud computing means that large capital investment is not required, you can have all the peak compute power and storage for almost nothing. If you are lucky, you may find enough friends, fools and family to invest in your company. Probably you won’t be able to pay anyone until you get your first product far enough along that you have a chance to raise some investment. The key is to keep that investment small, keep the burn rate low and run the company so that a $20M acquisition is attractive. Or that you have the old Metasoft (HSPICE) model of having a small highly profitable EDA company that throws off a lot of cash every year in bonuses, even if you don’t get acquired (although they did in the end when Gerry Hsu sent them an offer that they had 24 hours to accept). VCs disparagingly call these “lifestyle” companies, but if you have 15 people in the company and $10M in revenue that could be quite a lifestyle.

So why am I reiterating all this? Because Jim Hogan and I are leading a discussion on just this topic next Tuesday evening. “So you want to start an EDA company…” at the San Jose Doubletree on February 23 at 6:30-7:30 in the Oak Ballroom (bars are open in the area!). This is during DVcon but you don’t need to be registered to come along.

Be ruthless and cull all the director-level management of the existing product (Quicken Online in this case). Put the managers of the acquired product in charge.

This is one thing that I learned at Cadence (you might have noticed that Cadence has done a fair number of acquisitions over the years, to say the least). The first thing to do is to lay off all the managers responsible for the internal competing product. They will inevitably try and sabotage the acquisition in more or less devious ways, worry too much about users of the existing product and so on. The junior worker-bee programmers or designers can be reassigned; they are much less emotionally invested in the failed internal product and have the knowledge to merge any parts of the old product that make sense.

In the Quicken case they seem to be doing something different, based on what they have said anyway. The correct thing to do, in my opinion, is to put the mint.com guys in charge of everything. Not just their own product but also the Quicken Online product. And the managers of Quicken Online need to go. They probably weren’t in favor of the acquisition and will subtly try and show that it was a mistake and try and ensure as much as possible of their own work survives going forward. But it is the mint.com product where as much as possible must survive going forward, and the best way to ensure that is to put those guys in charge.

Steve Jobs did just this when he returned to Apple along with the operating system from Next (internally Mac code is still littered with classes that start NS for NextStep). He put the Next software managers in charge and pushed out the managers who had been responsible for the failed strategy that Apple had been pursuing. The Next managers could implement their strategy much more easily if they didn’t have another set of managers arguing with them about every decision.

Everybody knows that the big time sink in mergers is where products overlap. But the best way to handle this is to make sure that the managers of the successful, acquired, product are in charge of those decisions and not the managers of the failed product. This doesn’t make the problem go away completely, after all the customers of the existing product cannot typically simply be upgraded painlessly to the new product, but at least it means that the winning product will be the acquired one, which is essentially the decision that senior management had already determined is what they wanted to have happen when they decided to do the acquisition.

Not all mergers are like this, of course. Sometimes the new product line is completely complementary with no overlap. But often, under the hood, there is more overlap than is obvious. When Cadence acquired Ambit, they were already ahead of the curve because their internal synthesis product, Synergy, was doing so badly that they had killed it off six months before they acquired us. But one reason for acquiring Ambit was for its timing engine, which seemed to be the best in existence at that time, but the existing timing team at Cadence still controlled timing strategy. It took months to arrive at the foregone conclusion that the Ambit timing engine should “win” and become the Cadence timing engine, a decision that would have taken 5 minutes if Ambit’s timing team had been put in charge on day 1.

It is very difficult to keep innovation going after an acquisition, especially if it is done at a high price so that many individuals have made significant money and are really hanging around largely to vest the rest of their stock. Keeping a competing team around, and one that already is better connected politically, almost guarantees that innovation will stop and that the acquisition will be much less successful than it could have been.

There are two reasons for asking questions in an interview, when you are the candidate. One is that the type of questions you ask reveal that you are already thinking about the important issues affecting the company. And the other is that you genuinely want to know. In most cases, the questions serve both ends. In fact most questions you ask should help you decide if the company is going to be successful and whether you have the right skillset to improve those chances.

When you interview for a position at a startup, it is important to realize that you are interviewing the company as much as they are interviewing you. The point of working for a startup is that the stock they give you will be valuable (otherwise go do something else) and they need to convince you of that. When you interview at a big successful company it is much more of a case of them interviewing you. After all, if you’ve done your homework, you should know what makes them successful. Most of that information is in the public domain.

The most important question I like to ask is why the senior people in the company believe it will be successful. Since they work there, presumably they do but sometimes that have a hard time articulating why. The answer needs to be more than just having good people or good technology. The market that they sell into needs to be large enough and homogenous enough for their (or any) product strategy to have the possibility of being successful.

Another thing I like to ask are: what is the one reason people buy your product? Of course, just like John Bruggeman was pointing out on Tuesday, if they don’t have a good answer then there is all the more upside from doing a great job at marketing (if you are interviewing for a marketing position). But typically, if most of the company is engineers, they’ll have too many answers to this question rather than too few. Avoid the fine art and bicycles problem. City Slickers marketing is finding out the “one thing” and becoming focused on delivering that. If customers are all buying for different reasons, it is not possible to build a repeatable sales process.

A third question is to ask, which is good in non-startups too, is “If I got the job and was starting tomorrow morning at 9am, what would be the most important things to get working on?” They may not be the most important strategic things long-term, but if there hasn’t been any marketing before there is usually a backlog of urgent stuff: the customer presentation is hopeless, the website hasn’t been updated in ages, the company logo sucks, engineering needs a decision about which standard to support, or whatever.

So obviously cash is so important in a startup that it should never be spent? Well, not so fast. I’ve seen some really silly decisions about how to save money in startups over the years.

Most of the cash being burned in a software startup goes on engineers’ salaries. Consequently it makes sense to do everything to make their work environment as productive as possible. Do not force them to use old computers because they are already around. Computers are pretty cheap these days, a few days of an engineer’s salary will buy you something really high end. Do not equip the servers with so little disk space that they have to delete old data that will eventually turn out to be useful. Terabyte disk drives are under a hundred dollars. And don’t forget to make it easy for your engineers to work from home, by having good VPN and paying for them to have good internet connections. You pay your engineers more in an hour than their internet connection costs for a month.

One thing we discovered at Envis was that companies that provide PCs for gamers deliver the most bang for the buck. They overclock the designs, add special fast memory, have custom motherboards and so on. For a lot less than Dell will sell you a machine, you can get one that is half as fast again, with lots of cores. And a bonus, they look really cool.

Do not hire a consultant and then, for egalitarian reasons, give them a day on which it is their job to clean the kitchen at $150/hour. In fact, don’t make your engineers clean the kitchen anyway. That’s pretty pricy labor. And don’t annoy your engineers by charging for coffee or soda.

In a semiconductor company, there is an additional significant cost, namely design tools. Sometimes this is a cost in a software company too since they need tools for quality assurance and integration purposes. This is a hard balance to get right since too few tools again means that what looks like saving money on tools is really burning extra money on engineers’ salaries. Too many tools obviously wastes money more directly. When I was at Cadence we had a venture investment program where we would provide almost unlimited tools to startups for a mixture of cash and an equity position. We’d discovered that most startups underinvested in tools because they were so expensive but that this jeopardized their success.

Benefits, especially medical, are another area where startups can be penny wise and pound foolish. The most cost effective way to handle medical benefits, given that usually everyone is young and fit, is a combination of catastrophic coverage and a health savings account (HSA). In fact this is probably the best way to handle medical period, but that’s a political hot potato right now. It is what Whole Foods does and what John Mackey, the CEO, has recently got into trouble with the left for recommending as better than what congress is attempting to put together.

Bottom line: remember Gordon Bell’s line that cash is more important than your mother. But remember that engineers’  salaries are your biggest investment, and it is foolish not to do everything to make that investment as effective as possible.

When I say it’s time to move on I don’t mean that you need to move company, although that is certainly one option. If you move to work on a new product you’ll be learning stuff again. If you relocate to Japan you’ll be learning stuff again. If you move from application engineering to product marketing you’ll be learning again.

In particular, if you get promoted your job will change and you’ll be learning stuff again. This is especially acute the first time you are promoted into management. Typically you are the best engineer or salesperson or whatever on the team and so you get promoted. Now you have to learn about management, a subject that previously you may not have taken much interest in. It is an especially difficult transition since your comfort zone is not to do management at all, just do everyone’s jobs for them (after all, you were the best on the team so you are better than they are). It is a hard lesson to learn that as a manager your output is not what you do personally, it is the output of your group. It is not a positive that you did a lot of the work yourself, that means you are not doing a good job of nurturing the people in your group, not training them to be as good as you are.

People will often move on to another company anyway if they are bored since there might not be an appropriate position to move into, or a promotion to be had. This is especially true of new graduates who get fed up with some aspects of the company bureaucracy or culture and move to a new company to escape. However, the new company is typically the same (although different in details). It’s just the nature of companies that they don’t always do just what you think they ought to. The result of this phenomenon is that I think the best value people you can possibly hire are people who have already worked for at least one company and have 3-5 years experience. At that point they are enormously more productive than a brand new graduate, not about to leave because of company bureaucracy, and although they are paid more they are not paid a correct premium. The new graduates are probably overpaid and the 3-5 year people underpaid.

I know mostly about engineering and a good engineer is not 30% better than a poor one, they are ten times more productive. So 3-5 year guy is not 50% better than a new graduate, which may reflect the pay differential, they may be 5 times better.

The Apache/Sequence acquisition is interesting for a couple of reasons. One is that both companies are private and I think we will see much more of this. A lot of the smaller EDA companies simply will not make it and the big guys don’t have a lot of appetite for acquisition these days. And let’s face it, this is not a success story for Sequence.

Sequence itself has a long history, extending back to its two ancestor companies, Sente and Frequency Technology. A third company, Sapphire, got folded in a little later. Sente goes all the way back to the early 1990s and Frequency to the mid-90s. The company was completely recapitalized at least once, but that didn’t seem to be enough to get them to lift-off. Clearly the investors didn’t have an appetite to do it again and pulled the plug. Sequence managed to carve out a good niche with its PowerTheater product which has become one of the standard tools for power measurement. But Sequence never managed to truly make the transition to focus on that one successful product, I think since its older products were in use at key customers and they couldn’t afford the loss of revenue from stopping development on them. Maybe inside Apache that will be easier to rationalize.

Synfora acquiring Esterel Studio is, of course, another private company making an acquisition, although this time of technology rather than a company. Esterel was actually developed in Sophia Antipolis at Ecole des Mines (yes, the school of mining although it really is much more general engineering these days) to where a couple of years later I’d relocate. The Esterel is the large red rocky peninsula between Cannes and San Tropez. But I digress. Since Synfora apparently has several customer already using both products together, since they are largely complementary, I should think integration should be straightforward.

The other acquisition was Global Foundries and Chartered Semiconductor. Technically it is the government of Abu Dhabi that owns them both, but clearly the intention is to attempt to create a more integrated foundry business. I never really understood why Chartered wasn’t more successful than it was. One theory I’ve heard is that they were very slow to make critical decisions since every decision had to go back to Singapore and so they failed to get close to their customers (who, pretty much by definition, were not in Singapore since it’s tiny). Both companies are members of IBM’s foundry club although their focus is different. Global Foundries has a lot riding on the success of AMD (whence it was spun-out) and I’m afraid I’m skeptical about that. Chartered is in the mainline bulk-CMOS business and struggles against TSMC since wafer costs are largely a matter of scale. TSMC has incrementally been building a gigafab with 100,000 wafer starts per month, which conventional wisdom considers gives it at leas a 10% cost advantage over merely large fabs. That’s a big difference.

I think we are going to see more consolidation in both EDA and semiconductor. In EDA, I think the strong private companies will swallow the weak. But there aren’t that many strong swallowers so we’ll also see companies just fade away. In semiconductor I think that there may be some consolidation at the level of companies merging, but more likely are whole divisions being sold to create stronger focused companies, as happened with NXP selling their wireless business to ST, and Freescale selling their wireless business to…well, nobody wanted it.

I think all senior management jobs consist of two separate dimensions that have two separate skill sets. I call these management and leadership. Some people are good at both, some are good at only one.

Management is the basic operational management of the company. Presumably you already know how to do this, at least in your own domain (engineering, marketing, sales, etc) or you probably wouldn’t have been promoted. When you get more senior you have a new challenge: you have to manage people not from your own domain. If you are an engineer, it’s like salespeople are from another planet and you don’t understand what makes them tick. If you are a salesperson you may think the same about engineering. If the company is medium sized things are not so bad since you’ll have a sales manager and an engineering manager to insulate you. But if the company is small then you’ll have to manage the aliens directly. My recommendation is to get some advice. If you’ve never set up a sales commission plan before, don’t assume that because you are a smart engineer who knows Excel that you can just wing it. If you don’t know a friendly VP sales who can give you free advice, find a consultant and pay them. It’s a lot cheaper than making major mistakes.

As CEO you may have only an accountant (or maybe nobody) to support you in finance. I think it makes sense to get a “CFO for a day” consultant to help you unless you are very comfortable with all the finance issues and already have a good feel for how to put together a business plan, how to turn a sales plan into a cash-flow forecast and so on. If your eyes glaze over when you read my blog postings on finance, you need someone to help you. Whatever you do in finance, don’t treat it as a problem that will go away if you ignore it. You’ll need to get a financial audit done at some point, sooner than you expect, and cutting corners will then come to light.

If you are not an engineer by background, you can’t manage engineering. That’s not to say that you aren’t capable of managing engineering but just like salespeople won’t respect you unless you’ve carried a bag, engineering people want to be managed by someone who understands technology and development and knows what it takes to get a product out. If you don’t have an engineering manager you’ll at least need to trust one of the senior engineers to be feeding you the unvarnished truth.

The second leg of being a CEO or a senior manager is leadership. The most important aspect of this is to get everyone in the company committed to moving the company in the same direction. Unless you make truly stupid decisions, it is more important that everyone is aligned than that the decision is ideal. As General Patton famously said, “A good plan executed violently now is better than the perfect plan next week.” In business, “violently” is the wrong adverb but the sentiment is the same.

Having said that, it is also important that the overall strategy of the company is good and represents the best that the management team can come up with. It is also important to be flexible. If something isn’t working then you’ll need to try something else and preferably while you still have enough money in the bank to find out whether that new approach is good. Remember, most successful startups end up doing something somewhat or completely different from what they set out to do initially.

A general rule about management and especially being a CEO: if something good happens in the company, everyone will tell you about it. If something bad happens, nobody will tell you. Despite the proverb that "bad news travels fast," inside a company bad news travels really slowly so you need to make a special effort to discover it. In the early stages it is good to have someone in engineering who is a personal friend who will not hide bad news. Later on, you need someone in sales like that who’ll tell you what is really happening when the company tries to sell the product. You can’t sit in your CEO office and believe everything that you are told. You have to get out and dig.

Actually the two things are very different and not many people seem to be good at making the transition once they have got used to how a big company works, with their assistants, and finance organization, and HR department and all the rest.

When I was at VLSI and the fab was not running effectively, the company would hire a VP from TI or Motorola (where the CEO had previously worked and so knew good people he’d worked with before). These guys were used to running a fab that was running smoothly, with a large organization around them. They were not used to sorting out a dysfunctional fab with very few people to support them. When they didn’t work out, they were doubly expensive because they needed big severance packages to get rid of them.

When you become CEO of a startup, you have do everything yourself. Especially if the startup is attempting to run very lean with minimal cash burn, and conserve most of that cash for engineering. You want to put together a business plan? Fire up Excel. There’s at most a part-time accountant in the finance department and you can’t delegate it to them. Even if you have a “CFO for a day” part-time senior finance consultant, they don’t understand the business intimately like you should because that’s bound up with strategy which is not just something financial. They can help review the plan but they can’t do it for you.

If you’ve not got a very good engineering manager then you can’t rely on the current schedules. And you don’t have enough money to do what you would in a big company and hire a good engineering manager, or even a really good product management specialist. You have to do that yourself too. In a typical startup, as CEO, you will probably be the only person who isn’t writing code or designing chips.

Another problem with big companies is that people don’t really know how successful their business really is, since it is often very bound up in company-wide financial measures that are not closely enough tied to reality. So it is easy to look good when you aren’t, or looks undeservingly bad. If you are in a big EDA company, nobody knows how to really allocate revenue from big volume purchases to product lines. If you are in a semiconductor company, the cost model is rarely as accurate as necessary, and fab variances (because the fab is overloaded, or underloaded, or not yielding as expected) distort it again.

If your company has a few hugely profitable product lines (think Intel or Synopsys) then the smaller product lines may look good or not depending on how the overhead of the company is handled, and whether the profitable lines eat a lot of overhead leading to everyone else looking good (margin bleed-through), or the opposite, leading to everyone else looking worse than reality. It is too expensive to do full activity-based costing (ABC) and so overhead is often misleading. If cost of sales is a fixed percentage  of revenue, that assumes all products and all order sizes are equally easy to sell, which is clearly not true. But this may make some product lines look great (hire that manager) and others look poor (and he looked so promising) even though it purely an artifact of the underlying management accounting.

Although it is possible to make the transition from a big company to a startup, but both EDA and fabless semiconductor are littered with people who failed to do that. They were very successful at running a division of a big company, but were unable to translate that skill into success at either founding or coming into a startup and getting it to run well.

On the other hand, many startups fail because they are too early to market. In EDA, technologies tend to be targeted at certain process nodes which we can see coming down the track. There’s little upside in developing technologies to retrofit old design methodologies that, by definition, already work. Instead, the EDA startup typically takes the Wayne Gretsky approach of going where the puck is going to be. Develop a technology that is going to be needed and wait for Moore’s law to progress so that the world does need it. The trouble with this is that it often underestimates the amount of mileage that can be got out of the old technologies.

Since process nodes come along every couple of years, and even that is slowing, getting the node wrong can be fatal. If you develop a technology that you believe everyone needs at 45nm but it turns out not to be needed until 30nm then you are going to need an extra two years of money. And even then, it may turn out not to be really compelling until that 22nm node, after you’ve gone out of business. All the OPC (optical proximity correction) companies were too early to market, supplying technology that would be needed but wasn’t at that point in time. Even companies that had good exits, like Clearshape, were basically running out of runway since they were a process generation ahead of when their technology became essential.

The windows paradigm was really developed at Xerox PARC (yes, Doug Englebart at SRI had a part to play too). Xerox is often criticised for not commercializing this but in fact they did try. They had a computer, the Xerox Star, with all that good stuff in. But it was way too expensive and failed because it was too early. The next attempt was Apple. Not Macintosh, Lisa (pictured above). It failed. Too early and so too expensive. One can argue the extent to which the first Macs were too early, appealing only to hobbyists at first until the laser printer (also invented at PARC) came along. There are other dynamics in play than just timing but Microsoft clearly made the most money out of commercializing those Xerox ideas, coming along after everyone else.

Another means of being too early is simply having an initial product that it turns out nobody needs yet because it’s not good enough yet. Semiconductor development processes are all about risk-aversion, and any change has to mean that the risk of changing is less than the risk of not changing. For a startup with an early product in a process generation where the technology might be only nice-to-have this is a high barrier to cross. The startup might just serve as a wakeup call to everyone else that a product is required in the space, and eventually another startup executes better (having seen the first company fail) or the big EDA companies copy the technology into their own product line.

Overall, I think more startups fail by being too early to market than fail by being too late. Remember, it’s the second mouse that gets the cheese.

In the long run the technology is usually the most important aspect of the acquisition but the business is important for two separate reasons. Firstly, the revenue associated with the standalone business, ramped up by some factor to account for the greater reach of the acquiring company’s sales channel, is the way that the purchase price is usually justified. It is too hard to value technology except as a business. That’s why the venture capital euphemism for selling a company for cents on the dollar is “technology sale.” But more importantly, the business is the validation of the technology. Nobody can tell whether a startup’s technology is any good except by looking to see if anyone is buying it.

However, once the acquisition is done there is an immediate conflict. There is a running business to be kept going. After all that was the justification for the purchase price. It took the whole company to do that before acquisition, so presumably it will take the whole company afterwards. In the short term, the differentiation of the technology rests on its continuing to sell well. But the real reason for the acquisition is often to acquire the base technology and incorporate it into the rest of the product line. The only people who know the technology well enough to do this are the acquired company’s engineering organization. Suddenly they are double booked, developing the product to the plans that underpinned the forward bookings forecast, and working with the acquiring company’s engineers to do the integration.

An example. When I was at Cadence we acquired Cadmos for their signal integrity product SeismIC. plus other stuff in development. Googling back at the press release, I see that a person whose name sounds strangely familiar said:

"Adding CadMOS signal integrity analysis engines to established Cadence analog and digital design solutions provides us with the best correct-by-design timing and signal integrity closure capabilities in the industry," said Paul McLellan, corporate vice-president of custom integrated circuit (IC) products at Cadence.

Except, of course, to realize that vision required the Cadmos engineering team to work full-time on integration. Meanwhile, there is a business going full blast selling the SeismIC standalone. I believe there was also an earnout (part of the acquisition price depends on how much was sold) based on the standalone business only. A difficult balancing act for the engineering managers and myself.

We had similar issues when Cadence acquired Ambit. We needed to integrate the Ambit timing engine (and later the underlying synthesis technology itself) into Cadence’s whole digital product line at the same time as we were trying to give Synopsys a run for their money in the standalone synthesis business. Both of those goals were really important strategically but there was only one set of engineers.

Balancing these two conflicting requirements is probably the hardest aspect to manage of a typical EDA acquisition. It is really important, not just for financial reasons, to maintain the leadership position of the technology in the marketplace. At the same time, integrate that leadership technology so that it is available under-the-hood in other parts of the product line which, in the end, is probably how it will mostly get into customer’s hands. Preserve the differentiation while doing the integration.