The Customer Development Manifesto: Reasons for the Revolution (part 1)

This post makes more sense if you read the previous post – The Leading Cause of Startup Death: The Product Development Diagram.

After 20 years of working in startups, I decided to take a step back and look at the product development model I had been following and see why it usually failed to provide useful guidance in activities outside the building – sales, marketing and business development.

Every startup has some methodology for product development, launch and life-cycle management. At their best, these processes provide detailed plans, checkpoints and milestones for every step in getting a product out the door: sizing markets, estimating sales, developing marketing requirements documents, prioritizing product features.  Yet at the end of the day even with all these processes 9 out of 10 of new products are failures.

So what’s wrong the product development model? The first hint lies in its name; this is a product development model, not a marketing model, not a sales hiring model, not a customer acquisition model, not even a financing model (and we’ll also find that in most cases it’s even a poor model to use to develop a product.) Yet startup companies have traditionally used this model to manage and pace not only engineering but also non-engineering activities.

In this post I’m going to describe the flaws of the product development model.  In the next few posts that follow, I’ll describe more specifically how this model distorts startup sales, marketing and business development. And how thinking of a solution to this commonly used model’s failures led to a new model – the Customer Development Model – that offers a new way to approach startup activities outside the building. Finally, I’ll write about how Eric Ries and the Lean Startup concept provided the equivalent model for product development activities inside the building and neatly integrates customer and agile development.

Product Development Diagram

Product Development Diagram

1. Where Are the Customers?
To begin with, the product development model completely ignores a fundamental truth about startups and new products. The greatest risk in startups —and hence the greatest cause of failure—is not the technology risk of developing a product but in the  risk of developing customers and markets. Startups don’t fail because they lack a product; they fail because they lack customers and a profitable business model. This alone should be a pretty good clue about what’s wrong with using the product development diagram as the sole guide to what a startup needs to be doing. Look at the Product Development model and you might wonder, “Where are the customers?”

The reality for most startups today is that the product development model focuses all their attention on activities that go on inside a company’s own building. While customer input may be a checkpoint or “gate” in the process, it doesn’t drive it.

2. The Focus on a First Customer Ship Date
Using the Product Development model also forces sales and marketing to focus on the end point of the process – the first customer ship date. Most sales and marketing executives hired into a startup look at the “first customer ship date,” look at the calendar on the wall, and then work backwards figuring out how to do their job in time so that the fireworks start the day the product is launched.

The flaw in this thinking is that “first customer ship” is simply the date when engineering thinks they “finished” the 1.0 release of the product. The first customer ship date does not mean that the company understands its customers, how to market or sell to them or how to build a profitable business. (Read the preceding sentence again. It’s a big idea.)

Even worse, a startup’s investors are managing their financial milestones by the first customer ship date as well.

The product development model is so focused on building and shipping the product that it ignores the entire process of testing your basic hypothesis about your business model (customers, channel, pricing, etc.) before you ship. Not testing these hypotheses upfront is a fundamental and, in many cases, fatal error most startups make.

Why? Because it isn’t until after first customer ship that a startup discovers that their initial hypotheses were simply wrong (i.e. customers aren’t buying it, the cost of distribution is too high, etc.) As a result the young company is now saddled with an expensive, scaled-up sales organization frustrated trying to execute a losing sales strategy and a marketing organization desperately trying to create demand without a true understanding of customers’ needs.

As Marketing and Sales flail around in search of a sustainable market, the company is burning through its most precious asset—cash.

3. The Focus on Execution Versus Learning and Discovery
The product development model assumes that customers needs are known, the product features are known, and your business model is known. Given this certainty, it’s logical that a startup will hire a sales and marketing team to simply execute your business plan. You interview sales and marketing execs for prior relevant experience and their rolodexes, and hope they execute the playbook that worked for them in prior companies.

All of this is usually a bad idea.  No one asks, “Why are we executing like we know what we are doing? Where exactly did the assumptions in our startup business plan come from?”  Was the sales revenue model based on actually testing the hypotheses outside the building? Or were they a set of spreadsheets put together over late night beers to convince an investor that this is going to be a great deal?

No newly hired sales and marketing exec is going to tell a founder, “Hey my prior experience and assumptions may not actually be relevant to this new startup.” Great sales and marketing people are great at execution – that’s what you hired for. But past experience may not be relevant for your new company. A new company needs to test a series of hypothesis before it can successfully find a repeatable and scalable sales model. For startups in a new or resgemented market, these are not merely execution activities, they are learning and discovery activities that are critical to the company’s success or failure.

4. The Focus on Execution Versus Agility
The product development diagram has a linear flow from left to right. Each step happens in a logical progression that can be PERT charted with milestones and resources assigned to completing each step.

Anyone who has ever taken a new product out to a set of potential customers can tell you that the real world works nothing like that. A good day in front of customers is two steps forward and one step back. In fact, the best way to represent what happens outside the building is more like a series of recursive circles—recursive to represent the iterative nature of what actually happens in a learning and discovery environment. Information and data are gathered about customers and markets incrementally, one step at a time. Yet sometimes those steps take you in the wrong direction or down a blind alley. You find yourself calling on the wrong customers, not understanding why people will buy, not understanding what product features are important. Other times potential customers will suggest a new use for the product, new positioning or even a much better idea.

The ability to learn from those missteps, to recognize new opportunities, and to rapidly change direction is what distinguishes a successful startup from those whose names are forgotten among the vanished.

5. The Outsourcing of Founders Responsibility
The Product Development model separates founders from deeply understanding their customers and market. The responsibility for validating the founders original hypotheses is delegated to employees – the sales and marketing team.

This means the founders are isolated from directly hearing customer input – good, bad and ugly. Worse, founders really won’t understand whether customers will buy and what features are saleable until after first customer ship.

When an adroit and agile founder gets outside the building and hears for the nth time that the product is unsellable they will recognize, regroup and change direction. A process to give the founders continuous customer interaction – from day one – is essential.

6. The Focus on a Finished Product Rather than a Minimum Feature Set
The passion of an entrepreneur coupled with the product development diagram drives you to believe that all you need to do is build the product (in all its full-featured glory) and customers will come. A Waterfall development process reinforces that inanity. The reality is quite different.  Unless you are in an Existing Market, (making a better version of what customers are already buying) you’ll find that your hypothesis about what features customers want had no relationship to what they really wanted.

Most startup code ends up on the floor.

7. Investor Focus on a Broken Model
Ask VC’s why they use the Product Development model to manage a startup and you get answers like, “It’s the way my firm has always done it. Why change something that has worked so well over the last three decades?” Or, “Look at our returns, its always worked for us.” Or at times an even more honest answer, “My senior partners say this is the only way to do it.”

Some firms correctly point out that, “It’s fine if 8 out of 10 of our companies fail if the remaining two return 20x our money. That’s a better return than having 10 out of 10 companies succeed and each return 2x our money.  Therefore we don’t want startups doing anything but swinging for the fences.”

The fallacy is that the product development model is the most efficient model for new ventures swinging for the fences– this year, last year, last decade, or since the first startup met their first investor.

Venture portfolio companies don’t succeed because they used the Product Development model they succeeded in spite of using itThe fact is most successful startups abandon the product development model as soon as they encounter customers.

Today, startups using the product development model iterate and learn and discover by burning investor cash. When cash is tight, they go out of business – or they adopt a more efficient model.

—–

Part 2 of the Customer Development Manifesto to follow.

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to FurlAdd to Newsvine

The Leading Cause of Startup Death – Part 1: The Product Development Diagram

When I started working in Silicon Valley, every company bringing a new product to market used some form of the Product Development Model.  Thirty years later we now realize that its one the causes of early startup failure. This series of posts is a brief explanation of how we’ve evolved from Product Development to Customer Development to the Lean Startup.

The Product Development Diagram
Emerging early in the twentieth century, this product-centric model described a process that evolved in manufacturing industries. It was adopted by the consumer packaged goods industry in the 1950s and spread to the technology business in the last quarter of the twentieth century. It has become an integral part of startup culture.

At first glance, the diagram, which illustrates the process of getting a new product into the hands of waiting customers, appears helpful and benign.  Ironically, the model is a good fit when launching a new product into an existing, well-defined market where the basis of competition is understood, and its customers are known.

The irony is that few startups fit these criteria. (None of mine did.)  We had no clue what our market was when we first started. Yet we used the product development model not only to manage product development, but as a road map for finding customers and to time our marketing launch and sales revenue plan. The model became a catchall tool for all schedules, plans, and budgets. Our investors used the product development diagram in our board meeting to see if we were “on plan” and “on schedule.” Everyone was using a road map that was designed for a very different location, yet they are surprised when they end up lost.

Product Development Diagram

Product Development Diagram

To see what’s wrong with using the product development model as a guide to building a startup, let’s first examine how the model is currently used to launch a new product. We’ll look at the model stage-by-stage.

Concept and Seed Stage
In the Concept and Seed Stage, founders capture their passion and vision for the new company and turn them into a set of key ideas, which quickly becomes a business plan, sometimes on the back of the proverbial napkin. The first thing captured and wrestled to paper is the company’s vision.

Then the product needs to be defined: What is the product or service concept? What are the features and benefits? Is it possible to build? Is further technical research needed to ensure that the product can be built?

Next, who will the customers be and where will they be found? Statistical and market research data plus potential customer interviews determine whether the ideas have merit.

After that there’s a discussion of how the product will reach the customer and the potential distribution channel. The distribution discussion leads to some conclusions about competition: who are they and how they differ. The startup develops its first positioning statement and uses this to explain the company and its benefits to venture capitalists.

The distribution discussion also leads to some assumptions about pricing. Combined with product costs, an engineering budget, and schedules, this results in a spreadsheet that faintly resembles the first financial plan in the company’s business plan. If the startup is to be backed by venture capitalists, the financial model has to be alluring as well as believable. If it’s a new division inside a larger company, forecasts talk about return on investment.  in this concept and seed stage, creative writing, passion, and shoe leather combine  in hopes of convincing an investor to fund the company or the new division.

Product Development
In stage two, product development, everyone stops talking and starts working. The respective departments go to their virtual corners as the company begins to specialize by functions.

Engineering focuses on building the product; it designs the product, specifies the first release and hires a staff to build the product. It takes the simple box labeled “product development” and makes detailed critical path method charts, with key milestones. With that information in hand, Engineering estimates delivery dates and development costs.

Meanwhile, Marketing refines the size of the market defined in the business plan (a market is a set of companies with common attributes), and begins to target the first customers. In a well-organized startup (one with a fondness for process),  the marketing folk might even run a focus group or two on the market they think they are in and prepare a Marketing Requirements Document (MRD) for Engineering. Marketing starts to build a sales demo, writes sales materials (presentations, data sheets), and hires a PR agency. In this stage, or by alpha test, the company traditionally hires a VP of Sales who begins to assemble a sales force.

Alpha/Beta Test
In stage three, alpha/beta test, Engineering works with a small group of outside users to make sure that the product works as specified and tests it for bugs. Marketing develops a complete marketing communications plan, provides Sales with a full complement of support material, and starts the public relations bandwagon rolling. The PR agency polishes the positioning and starts contacting the long lead-time press while Marketing starts the branding activities.

Sales signs up the first beta customers (who volunteer to pay for the privilege of testing a new product), begins to build the selected distribution channel, and staffs and scales the sales organization outside the headquarters. The venture investors start measuring progress by number of orders in place by first customer ship.

Hopefully, somewhere around this point the investors are happy with the company’s product and its progress with customers, and the investors are thinking of bringing in more money. The CEO refines his or her fund-raising pitch and hits the street and the phone searching for additional capital.

Product Launch and First Customer Ship
Product launch and first customer ship is the final step in this model, and the goal the company has been driving for. With the product working (sort of), the company goes into “big bang” spending mode. Sales is heavily building and staffing a national sales organization; the sales channel has quotas and sales goals. Marketing is at its peak. The company has a large press event, and Marketing launches a series of programs to create end-user demand (trade shows, seminars, advertising, email, and so on). The board begins measuring the company’s performance on sales execution against its business plan (which typically was written a year or more earlier, when the entrepreneur was looking for initial investments).

Building the sales channel and supporting the marketing can burn a lot of cash. Assuming no early liquidity (via an IPO or merger) for the company, more fund raising is required. The CEO looks at the product launch activities and the scale-up of the sales and marketing team, and yet again goes out, palm up, to the investor community. (In the dot-com bubble economy, the investors used an IPO at product launch to take the money and run, before there was a track record of success or failure.)

The Leading Cause of Startup Death
If you’ve ever been involved in a startup, the operational model no doubt sounds familiar. It is a product-centric and process-centric model used by countless startups to take their first product to market.  It used to be if you developed a plan on model that looked like this your investors would have thought you were geniuses.

In hindsight both you and your investors were idiots. Following this diagram religiously will more often than not put you out of business. The diagram was developed to be used by existing companies doing product line extensions - not startups creating new markets or resegmenting existing ones. Most experienced entrepreneurs will tell you that the model collapses at first contact with customers.

VC’s who still believe in the product development model in the 21st century offer no value in building a company other than their rolodex and/or checkbook.

Coming next Part 2: What’s Wrong with Product Development as a Model?

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to FurlAdd to Newsvine

The End of Innocence

I love TechCrunch. If you’re a startup raising money or just want to see your name online, there’s not a better blog on the web.  Reading this TechCrunch post made me remember the first time I saw someone confront a worldview they didn’t expect.

TechCrunch PRDiscovering that your worldview is wrong or mistaken can be a life-changing event. It’s part of growing up but can happen at any age. What you do when it happens shapes who you’ll become.

Dinner in a Strange Land
When I was in my mid 20′s working at ESL, I was sent overseas to a customer site where the customers were our three-letter intelligence agencies. All of us knew who they were, understood how important this site was for our country, and proud of the work we were doing. (Their national technical means of verification made the world a safer place and hastened the end of the Soviet Union and the Cold War.)

As a single guy, I got to live in a motel-like room on the site while the married guys lived in town in houses and tried to blend in with the locals. When asked what they did, they said they worked at “the xxx research facility.”  (Of course the locals translated that to “oh do you work for the yyy or zzz intelligence agency?”)

One warm summer evening I got invited over to the house of a married couple from my company for a BBQ and after-dinner entertainment – drinking mass quantities of the local beer. The quintessential California couple, they stood out in our crowd as the engineer (in his late 20′s, respected by his peers and the customer) had hair down to his shoulders, sharply contrasting with the military crewcuts of the customers and most of the other contractors.

His wife, about my age, could have been a poster child for the stereotypical California hippie surfer, with politics that matched her style – antiwar, anti government, antiestablishment.

One of the rules in the business was that you didn’t tell your spouse, girlfriend, significant other who you worked for or what you worked on – ever. It was always a welcome change of pace to leave the brown of the unchanging desert and travel into town and have dinner with them and have a non-technical conversation about books, theater, politics, travel, etc. But it was a bit incongruous to hear her get wound up and rail against our government and the very people we were all working for. Her husband would look at me out the corner of his eyes and then we’d segue the conversation to some other topic.

That evening I was there with three other couples cooking over the barbie in their backyard. After night fell we reconvened in their living room as we continued to go through the local beer. The conversation happened to hit on politics and culture and my friend’s’ wife innocently offered up she had lived in a commune in California. Well that created a bit of alcohol-fueled cross-cultural disconnect and heated discussion.

Until one of the other wives changed a few lives forever with a slip of the tongue.

Tell Me it Isn’t True
One of the other wives asked, “Well what would your friends in the commune think of you now that your husband is working for intelligence agencies x and y?”

As soon as the words came out of her mouth, I felt time slow down. The other couples laughed for about half a second expecting my friend’s wife to do so as well. But instead the look on her face went from puzzlement in processing the question, to concentration, as she was thinking and correlating past questions she had about who exactly her husband had been working for. It seemed like forever before she asked with a look of confusion, “What do you mean agencies x and y?”

The laughter in the room stopped way too soon, and the room got deathly quiet. Her face slowly went from a look of puzzlement to betrayal to horror as she realized that that the drunken silence, the dirty looks from other husbands to the wife who made the agency comment, and the wives now staring at their shoes was an answer.

She had married someone who never told her who he was really working for. She was living in a lie with people she hated. In less than a minute her entire worldview had shattered and coming apart in front of us, she started screaming.

This probably took no more than 10 seconds, but watching her face, it felt like hours.

I don’t remember how we all got out of the house or how I got back to the site, but to this day I still remember standing on her lawn staring at strange constellations in the night sky as she was screaming to her husband, “Tell me it isn’t true!”

The next day the site supervisor told me that my friend and his wife had been put on the next plane out of country and sent home (sedated) along with the other couple that made the comment. By the time I came back to the United States, he was gone from the company.

It’s been thirty years, but every once an awhile I still wonder what happened to the rest of their lives.

———-

The End of Innocence
In much smaller ways I’ve watched my children and now my students discover that their worldview is wrong, mistaken or naive. I’ve watched as they realize there’s no Santa Claus and Tooth Fairy; the world has injustice, hypocrisy and inequality; capitalism and politics don’t work like the textbooks and money moves the system; you can’t opt out of dying, and without regulation people will try to “game” whatever system you put in place.

Learning to accept the things you can’t change, finding the courage to change the things you can and acquiring the common sense to know the difference, is part of growing up.

While I love TechCrunch, the post and the quote about the PR agency (“one PR firm has discovered a dynamite strategy, throw ethics out the window”) left me wondering; how do PR agencies interact with TechCrunch and other blog and review sites? Is this behavior an outlier or is it the norm in the PR industry?

Or is it just someones end of innocence?

Listen to the blog post here


Download the podcast here or here

Coffee With Startups

I’ve just met four great startups in the last three days.

An Existing Market
All four were trying to resegment an “Existing Market.” An existing market is one where competitors have a profitable business selling to customers who can name the market and can tell you about the features that matter to them. Resegmentation means these startups are trying to lure some of the current or potential customers away from incumbents by either offering a lower cost product, or by offering features that appealed to a specific niche or subset of the existing users.

Some of the conversations went like this:

Startup 1
Entrepreneur -“I’m competing against Company x and have been following the Customer Development process and I’ve talked to lots of customers.”
Me – “Have you used Company x’s product? Do you know have they distribute their product? Do you know how they create demand? Do you know how many units they are selling? Do you know the archetype of their customers?
Entrepreneur -“Well no but my product is much better than their product and I have this great idea….”

Rule 1: In an existing market Customer Development means not only understanding potential customers, but your competitors in detail – their product features, their sales channels, their demand creation strategy, their business model, etc.

Startup 2
Entrepreneur -“I’m competing against Company x and we are going to offer a lower-cost, web-based version. We’re about to ship next week.”
Me –“That’s a great hypothesis, do customers tell you that they’d buy your version if it was cheaper or on the web?
Entrepreneur -“Well no but my product is much cheaper and everyone’s on the web and I have this great idea….”

Rule 2: In an existing market Customer Development means understanding whether your hypothesis of why customers will buy match reality. This is easy to test. Do this before you write code you may end up throwing away.

Startup 3
Entrepreneur -“I’m competing against Large Company x and we solve problems for a set of customers – I’ve talked to many of them and they would buy it.”
Me – “So what’s the problem?”
Entrepreneur – “We just started letting early customers access the product and adoption/sales isn’t taking off the way we thought it would. We only have 20 customers, and Large Company x has millions.”
Me – “How are you positioning your product?”
Entrepreneur – “We tell potential customers about all our features.”

Rule 3: In an existing market directly compare your product against the incumbent and specifically describe the problems you solve and why Company x’s products do not.”

Startup 4
Entrepreneur -“I have something really, really new. No one has anything like it.”
Me – “Isn’t it kind of like Twitter but better?”
Entrepreneur – “You don’t get it.”

Rule 4: You may want to think twice positioning as a New Market. If customers immediately get an analogy for your product, don’t dissuade them. Save the “New Billion Dollar Market” positioning for the investors, not customers.

Lessons Learned

  • Deeply understand the incumbents that make up the Existing Market
  • The “hypotheses tested to lines of code written” ratio ought to be high
  • Position against the incumbents weaknesses – their customers will tell you what they are
  • Existing Markets adoption rates are measured in % market share gained, New Markets have adoption rates which may occur in your company’s lifetime

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to FurlAdd to Newsvine

The Secret History of Silicon Valley Part X: Stanford Crosses the Rubicon

This post is the latest in the “Secret History Series.”  They’ll make much more sense if you read some of the earlier ones for context. See the Secret History bibliography for sources and supplemental reading.

———————–

Swords Into Plowshares
After the end of World War II, returning veterans were happy to beat swords into plowshares (and microwave tubes) on the Stanford campus. From 1946 until 1950, Stanford’s Electronic Research Lab conducted basic research in microwave tubes.  Although this reseearch would lead to the development of the Backward Wave Oscillator and Traveling Wave Tube for military applications, Stanford was building tubes and circuits not entire systems.  The labs basic research was done by graduate students or Ph.Ds doing postdoctoral internships, supervised by faculty members or hired staff (many from Fred Terman’s WWII Electronic Warfare lab.)

In 1949, with the detection of the first Soviet nuclear weapons test, the Iron Curtain falling across Europe and the fall of China to the Communists, Cold War paranoia drove the U.S. military to rearm and mobilize.

Source: Center for Arms Control and Non-Proliferation (in constant 2009 $’s)

Source: Center for Arms Control and Non-Proliferation (in constant 2009 $’s)

We’ll Do Great in the Next War
Early in 1950, just months before the outbreak of the Korean War the Office of Naval Research asked Fred Terman to build an Applied electronics program for electronic warfare. All branches of the military (the Air Force and Army would fund the program as well) wanted Stanford to build prototypes of electronic intelligence and electronic warfare systems that could be put into production by partners in industry. The Navy informs Terman that, “money was not a problem but time was.”

Pitching the idea to the President of Stanford, Terman enthusiastically said, “In the event of all-out war, Stanford would become one of the giant electronic research centers…”  (A bit optimistic about the outcome perhaps, given that both the U.S. and the Soviet Union had nuclear weapons at this point.)

Crossing the Rubicon – The Applied Electronics Lab
Setting up a separate Applied Electronics Lab for military funded programs doubled the size of the electronics program at Stanford. The new Applied Electronics Laboratory was built with Navy money and a gift from Hewlett-Packard. With the memories of WWII only five years old, and the Cold War now a shooting war in Korea, there was very little discussion (or dissension) about turning a university into a center for the production of military intelligence and electronic warfare systems.

The work in the applied program focused in fields in which faculty members or senior research associates specialized.  Many of the other staff in the applied program were full-time employees hired to work solely on these military programs.

ELINT, Jammers and OTH
The Applied Electronics Lab used the ideas and discoveries (on microwave tubes and receiver circuits) from Terman’s basic research program in the Electronic Research Lab. The Applied Lab would build prototypes of complete systems such as Electronic Intelligence systems, Electronic Warfare Jammers, and Over the Horizon Radar. The Applied Electronics Lab also continued work on the Klystron, pushing the tube to produce megawatts in transmitted power. (Stanford designed Klystrons producing 2½ Megawatts were manufactured by Varian and Litton would power the radar in the BMEWS (Ballistic Missile Early Warning System) built at the height of the cold war.) The close tie between the two labs was a unique aspect of the Stanford Lab. Stanford had a Customer Development loop going on inside their own lab. The discoveries in tube and circuit research suggested new electronic intelligence and countermeasure techniques and systems; in turn the needs of the Applied Lab pushed tube and circuit development.  With the Applied Electronics Lab Stanford was becoming something akin to a federal or corporate lab run under university contract.  The university found government contracts profitable as the government reimbursed their overhead charges (their indirect costs.) This means they could fund other non-military academic programs from this overhead.

The Stanford Applied Electronics Lab built prototypes which were handed off to the military labs for their evaluation. Subsequently military labs would contract with companies to build the devices in volume. In some cases, branches of the military contracted directly with Stanford which worked with local contractors in Silicon Valley to build these components or systems for the military. The prototype ELINT receivers built by the Applied Electronics Lab used the Stanford Traveling Wave Tubes. They quickly went into production at Sylvania Electronic Defense Labs down the street in Mountain View and Hallicrafters in Chicago. Later versions would be built by numerous industry contractors and installed on the fleet of ELINT planes orbiting the Soviet Union. These traveling wave tubes would also become the heart of the panoramic receiver used on the B-52 by the electronic warfare officer to get the bomber through the Soviet Air Defense system.

Jammers built by the Stanford Applied Electronics Lab used the Stanford Backward Wave Oscillators to produce high power microwaves. Unlike the simple noise jammers used in World War II, Soviet radars were becoming more sophisticated and newer designs were fairly immune to noise. Instead the jamming signal needed to be much smarter and have a deep understanding of how the targeted radar worked. Taking the information gleaned from our ELINT aircraft, Stanford built prototypes of jammers modulated with two new deception jamming techniques – angle jamming and range-gate pull-off. Some form of these deception jammers would eventually find their way into most electronic warfare defense systems used in the Cold War; first in the U-2, A-12 and SR-71. (Ironically the B-52 bomber, which would become the airborne leg of our nuclear triad, would use dumb noise jammers for two more decades – the Air Force opting to put the smart jammers on the B-58 and B-70, high altitude supersonic bombers – one soon obsolete and other never made it into production.)

The last major area of research that the Applied Electronics Lab group investigated was how radio signals propagated within the earth’s ionosphere. Over the next fifteen years this Radio Science Laboratory would receive the most funding of all departments in the lab (from the CIA) to build a ground based ELINT system. They would build and deploy two Over The Horizon Radar (OTHR) systems to detect Soviet and Chinese ballistic missile tests using ground based radars.

Guards at the DoorStanford Joins the Cold War
In 1953 the Office of Naval Research told Terman that all military-funded projects (basic or applied, classified or not) needed to be in their own separate physical building. As a result Stanford moved the Applied work from the Electronics Research Lab into its own building.

In 1955, the pretense of keeping unclassified and classified work separate imposed too much of an administrative overhead and Stanford merged the Applied Electronics Lab and the Electronics Research Laboratory into the Systems Engineering Lab. The Applied Electronics portion of the lab was now the size of a small company.  It had 100 people, 18 of them full time faculty, 33 research associates and assistants and 33 other tube technicians, draftsman, machinists, etc. Over half this lab would hold clearances for military secrets. (Top Secret: Terman, Harris, McGhie, Secret: 44 others, Confidential: 8 others. Terman, Harris and Rambo also had Atomic Energy Commission “Q” clearances.)  Some students who were getting their engineering graduate degrees wrote masters and PhD thesis that were classified. Unless you had the proper clearances you couldn’t read them.  Terman and Stanford had just made a major bet on the cold war, and Stanford ranked sixth among university defense contractors.

A security guard was stationed at the door of the Applied Electronics Lab to ensure that only those with proper security clearance could enter. The law of unintended consequences meant that this most casual addition in front of a university building would result in the occupation and destruction of the lab (and its twin at MIT) and the end of the program 14 years later.  (More on this in a later post.)

Show and Tell – The Stanford ELINT and Electronic Warfare Contractors Meeting
During a typical year, the Applied Electronics Lab would host classified visits from military labs and defense contractors. By early 1950’s Stanford started holding a two day meeting for contractors and the military.

 

1955 Stanford Contractors Meeting

1955 Stanford Contractors Meeting

 

The 1955 attendee list gives you a feeling of the “who’s who” of the military/industrial establishment: RCA, GE, Motorola, AIL, Bendix, Convair, Mepar, Crosley, Westinghouse, McDonnell Aircraft, Douglas Aircraft, Boeing, Lockheed, Hughes Aircraft, North American, Bell Aircraft, Glen Martin, Ryan Aeronautics, Farnsworth, Sperry, Litton, Polarad, Hallicrafters, Varian, Emerson, Dumont, Maxson, Collins Radio.  Other universities doing classified ELINT and Electronic Warfare work attended including University of Michigan, Georgia Institute of Technology and Cornell. Over a hundred government contractors reviewed Stanford’s work on tubes and systems.

Stanford Contractors Meeting 1955 Attendees

Stanford Contractors Meeting 1955 Attendees

This was a classified conference at a university, the contractors not only got to hear the conference lectures, but also visited exhibits on the devices and systems the lab had built. The lab would repeat the conference the following week for government agencies doing military work.

Barely noticed at the 1955 conference, a year before the first transistor company opened in Silicon Valley, one of the sessions described how to use a new device called a“transistor” to build wide-band amplifiers. (Terman had sent faculty and graduate students to the University of Illinois in 1953 to learn transistor physics.)

The World Turned Upside Down
The Applied Electronics Lab solidified Stanford’s lead as one of, if not the place in the U.S. military for advanced thinking in ELINT and Electronic Warfare.  It would turn on its head the relationship of universities and corporations.

Traditionally universities chased corporations for funding and patronage, but the military’s dependence on Stanford’s and Fred Terman’s judgment turned that relationship on its head.  Now the military was listening to Terman’s advice about which military contractors should get the order for to mass produce the Stanford systems.  The contractors were now dependent on Stanford.

Terman the Rainmaker
During the 1950’s Fred Terman was an advisor to every major branch of the U.S. military. He was on the Army Signal Corps R&D Advisory Council, the Air Force Electronic Countermeasures Scientific Advisory board, a Trustee of the Institute of Defense Analysis, the Naval Research Advisory Committee, the Defense Science Board, and a consultant to the President’s Science Advisory Committee. His commercial activities had him on the board of directors of HP, Watkins-Johnson, Ampex, and Director and Vice Chairman of SRI.  It’s amazing this guy ever slept.  Terman was the ultimate networking machine for Stanford and its military contracts.

Stanford Industrial Park – Microwave Valley Booms
By the early 1950’s many of the corporations that attended the yearly Stanford Electronic Warfare conferences would establish research labs centered around Stanford for just this reason – to learn from Stanford’s basic and applied research and get a piece of the ELINT and Electronic Warfare contracting pie.

Stanford Industrial Park was the first technology office park set up to house local and out of state microwave and electronics startups. First occupied in 1953 it would include Varian, Watkins Johnson, Admiral, HP, General Electric, Kodak, Lockheed.  Other east coast companies which established branches in Microwave valley in the 1950’s included IBM, Sylvania, Philco, Zenith and ITT.

The Future is Clear – Microwave Valley Forever
By 1956 Fred Terman had every right to be pleased with what he had helped build in the last ten years in and around Stanford.  The Stanford Electronics Lab was now the center of ELINT and Electronic Warfare.

Startups were sprouting all over Microwave Valley delivering microwave tubes and complete military systems, slowiy replacing the orchards and fruit trees. Granger Associates was a 1956 startup founded by Bill Ayer, a graduate student in the Applied Electronics Radioscience Lab, and John Granger, a former RRL researcher, building ELINT and Electronic Warfare systems (the Granger jammer was carried on the U-2.) Four years later Ayer and another Granger engineer would leave Granger and found one of the preeminent electronic warfare and ELINT companies: Applied Technologies.

The future of the valley was clear – microwaves.

1956 – Change Everything
Yet in 1956 two events would change everything.  At the time neither appeared earthshaking or momentous. First, a Bell Labs researcher who had grown up in Palo Alto, had his own interesting World War II career, and recently served as a military advisor on cold war weapons systems, decided to follow Fred Terman’s advice to locate his semiconductor company near Stanford.

The second was when a Southern Californian aircraft company decided to break into the missiles and space field by partnering with Stanford electronics expertise. It moved its electronics research group from Burbank to the new Stanford Industrial Park and built its manufacturing facility in Sunnyvale.

Shockley Semiconductor Laboratory and Lockheed Missiles Systems Division would change everything. Read about it in Part XI of the Secret History of Silicon Valley here.

I’m From the Government and I’m Here to Help You

Apologies for a “Secret History” post that appeared today and now is gone from the blog.  (Those of you with a reader still have it.)  It was a rough draft you’ll see again when I can finish complete sentences.  (Something you can appreciate if you’ve read my class text on Customer Development.)

My only excuse is that this week I’m doing public service in my role as a California Coastal Commissioner.  Long days, lots of items on the calendar, dedicated staff, very bright fellow commissioners. Greatest hits here.

Touching the Hot Stove – Experiential versus Theoretical Learning

I’m a slow learner.  It took me 8 startups and 21 years to get it right, (and one can argue success was due to the Internet bubble rather then any brilliance.)

In 1978 when I joined my first company, information about how to start companies simply didn’t exist. No internet, no blogs, no books on startups, no entrepreneurship departments in universities, etc.  It took lots of trial and error, learning by experience and resilience through multiple failures.

The first few months of my startups were centered around building the founding team, prototyping the product and raising money. Since I wasn’t an engineer, my contribution was around the team-building and fund raising.

I was an idiot.

Customer Development/Lean Startups
In hindsight startups and the venture capital community left out the most important first step any startup ought to be doing – hypothesis testing in front of customers- from day one.

I’m convinced that starting a company without talking to customers is like throwing your time and money in the street (unless you’re already a domain expert).

This mantra of talking to customers and iterating the product is the basis of the Lean Startup Methodology that Eric Ries has been evangelizing and I’ve been teaching at U.C. Berkeley and at Stanford. It’s what my textbook on Customer Development describes.

Experiential versus Theoretical Learning
After teaching this for a few years, I’ve discovered that subjects like Lean Startups and Customer Development are best learned experientially rather than solely theoretically.

Remember your parents saying, “Don’t touch the hot stove!”  What did you do?  I bet you weren’t confused about what hot meant after that. That’s why I make my students spend a lot of time “touching the hot stove” by talking to customers “outside the building” to test their hypotheses.

However, as hard as I emphasize this point to aspiring entrepreneurs every year I usually get a call or email from a past student asking me to introduce them to my favorite VC’s.  The first questions I ask is “So what did you learn from testing your hypothesis?” and “What did customers think of your prototype?”  These questions I know will be on top of the list that VC’s will ask.

At least 1/3 of the time the response I get is, “Oh that class stuff was real interesting, but we’re too busy building the prototype. I’m going to go do that Customer Development stuff after we raise money.”

Interestingly this response almost always comes from first time entrepreneurs.  Entrepreneurs who have a startup or two under their belt tend to rattle off preliminary customer findings and data that blow me away (not because I think their data is going to be right, but because it means they have built a process for learning and discovery from day one.)

Sigh.  Fundraising isn’t the product.  It’s not a substitute for customer input and understanding.

Sometimes you need a few more lessons touching the hot stove.

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to FurlAdd to Newsvine

The Secret History of Silicon Valley Part IX: Entrepreneurship in Microwave Valley

This post is the latest in the “Secret History Series.”  They’ll make much more sense if you read some of the earlier ones for context. See the Secret History bibliography for sources and supplemental reading.

———————–

In the 1950′s Stanford University’s Electronics Research Laboratory (ERL) continued to develop innovative microwave tubes for the U.S. military. This next product, the Traveling Wave Tube, would have a major impact on electronic intelligence. Stanford’s Dean of Engineering, Fred Terman, encouraged scientists and engineers to set up companies to build these microwave tubes for the military. Funded by military contracts, these 1950′s microwave tube startups would help build Silicon Valley’s entrepreneurial culture and environment.

Why Electronics Intelligence?
Starting in 1946 Electronic Intelligence aircraft (ELINT) had been probing and overflying the Soviet Union to understand their air defense system. During the 1950′s, the U.S. Air Force Strategic Air Command, U.S. Navy and the CIA were the primary collectors of tactical and operational ELINT on the Soviet PVO Strany Air Defense system. (The NSA owned COMINT collection.) They flew an alphabet soup of Air Force and Navy planes (Navy PB4Y-2′s, P2V’s, P4M’s and EA-3′s, Air Force B-17s, EC-47′s, RB-29s, RB-50′s, and the ultimate ELINT collector of the 1950′s – the RB-47H.) Common to all these planes (generically called Ferrets) is that they were loaded with ELINT receivers, manned by crews called Crows.

The Strategic Air Command needed this intelligence to understand the Soviet air defense system (early warning radars, Soviet fighter plane radar, Ground Control Intercept radar, Anti-Aircraft gun radar, and radars guiding Soviet Surface to Air Missiles.) We needed this data to build radar jammers that could make the Soviet air defense radars ineffective so our bombers with their nuclear payloads could reach their targets. The information we collected would be passed on to defense contractors who would build the jammers to confuse the Soviet air defense radars.

ELINT Tasking
The ELINT program sought answers to operational questions like: What was the Radar Order of Battle a penetrating bomber would face? Were there holes in their radar coverage our bombers could sneak through? What was the best altitude to avoid the Soviet defenses? ELINT operators on each flight were tasked to gather basic data about the characteristics of the radar: is this a new type of radar or an existing one? What is its frequency, power, pulse repetition interval, rotation rate, scan rate, polarization, carrier modulation characteristics, etc. Then they would use direction finding equipment on their aircraft to locate its position.

ELINT Receivers
Early ELINT receivers were not much different then the radios you had at home – someone had to manually turn a dial to tune them to the correct frequency. By the 1950’s these receivers could automatically “sweep” a frequency band, but this action was mechanical and slow. That was fine if the Soviet radar was operating continuously, but if it was just a brief radar transmission or burst communication (which Soviet submarines used), we would probably miss hearing it. (The Soviets kept their radars turned off to stop us from recording their signals. So at times multiple ELINT planes would fly on a mission – one to run at the Soviet border appearing to attack, the other to pick up the signals from the air defense network as it responded to the intrusion. Keep in mind that 32 of these planes were shot down in the Cold War.)

The ultimate dream of ELINT equipment designers was a “high-probability of intercept” receiver, one that would pick up a signal that came up on any frequency and capture even a single pulse, however brief.

This was a two-pronged challenge: the U.S. needed receivers that could tune much faster than any of the manual methods that existed, and it needed receivers that could tune a much broader range of frequencies along the electromagnetic spectrum. Again Stanford technology would solve these challenges.

Rapid Scan/High Probability of Intercept – Stanford’s contribution
In the last post we described Stanford’s high power, electronically tuned microwave tubes (the Backward Wave Oscillator) which made high power, frequency agile airborne jammers possible.

Now Stanford’s Electronics Research Laboratory delivered another tube which forever changed electronic intelligence receivers - the Traveling Wave Tube (TWT.) Invented in Britain and further developed at Bell Labs, this tube would deliver the “holy grail” for ELINT receivers - instantaneous scan speed and extremely broad frequency range. A Traveling Wave Tube (TWT) could electronically tune through microwave frequencies at 1000 times faster than any other device, and it could operate in a frequency range measured in gigahertz.  As a microwave preamp, it had high gain, low noise and extremely wide bandwidth. It was perfect for a new generation of ELINT receivers to be built into the Ferret planes searching for signals around the Soviet Union. Later on TWTs would be built that could not only be used in receivers, but also actually transmit broadband microwaves at high power.

Invention Versus Commercialization
While Stanford was doing its share of pure research, what’s interesting about the Electronics Research Laboratory (ERL) was its emphasis on delivery of useful products for its customers – the military – from inside a research university.  The military had specific intelligence requirements and that meant that a TWT needed to be rugged enough to withstand being put on airplanes. This military/university collaboration for deliverable products is where the Electronics Research Laboratory (ERL) would excel – and ultimately end up leading to its destruction.

twt schematic

Traveling Wave Tube – Source: Thales Electron Devices

The Rise of “Microwave Valley” – More Stanford Tube Startups
The Traveling Wave Tube generated another series of startups from Stanford’s Electronics Research Laboratory.  R. A. Huggins, a research associate at the Stanford’s Engineering Research Lab, left in 1948 to start Huggins Laboratories in Palo Alto and put the first commercially manufactured traveling wave tube on the market. With a boost from military R&D contracts, Huggins Labs continued to expand, diversifying into backward-wave oscillators, low-noise TWTs, and electrostatic focused tubes. (In the 1970′s Huggins Labs sold to an east coast company, Microwave Associates (which became M/A-COM.)

Stanley Kaisel, a research associate at the Stanford ERL tube laboratory, left to join Litton’s startup. He left Litton in 1959 and started Microwave Electronics Corporation (MEC) to make low power, low noise TWTs. He sold the company to Teledyne in 1965.

Venture Capital, Microwaves and the OSS
Dean Watkinsthe leader of TWT research at Stanford’s Electronic Laboratory, left Stanford in 1957 and co-founded Watkins-Johnson (with R.H. Johnson the head of Hughes Aircraft microwave tube department) to market advanced TWTs to the military. Unlike the other Stanford tube spinouts which were funded with military contracts, Watkins-Johnson would be one of the first venture capital funded companies in the valley. Its first round of funding came from Tommy Davis (an ex-WWII OSS agent) then at the Kern County Land Company who knew Fred Terman through his military contacts. Terman and Davis negotiated the Watkins-Johnson investment and would sit on the Watkins-Johnson board together.

Frustrated with Kern’s lack of interest in investing in more technology companies, Tommy Davis would go on to found one of Silicon Valley’s first VC firms with Arthur Rock, creating Davis and Rock, founded in 1961. They would be one the first venture firms to organize their firm as a partnership rather than an SBIC or public company. They would also set the standard for the 20% carry for general partners. Tommy Davis would go on to found the Mayfield Fund in 1969.

These Stanford tube spinoffs joined the growing list of other microwave tube manufacturers in the valley including Eitel-McCullough, Varian, Litton Industries and Stewart Industries. Others would soon join them. By the early 1960s, a third of the nation’s TWT business and a substantial share of the klystron and magnetron industry was located in the Santa Clara Valley– and almost all of these companies emerged from one engineering lab at Stanford.

But microwave tubes were just the beginning of Stanford’s relationship with the military. Fred Terman was just getting warmed up. Much more was to come. Read about in Part X of the Secret History of Silicon Valley here.

Thirty-Six Years Later

One of my first posts (here) was learning about bats, moths and electronic countermeasures in natural systems in Thailand in the middle of the War in Vietnam.

Catching up on my back issues of Science magazine all I could do was smile when I read the title of an article in the July 17th issue:  Tiger Moth Jams Bat Sonar

Add to FacebookAdd to DiggAdd to Del.icio.usAdd to StumbleuponAdd to RedditAdd to BlinklistAdd to TwitterAdd to TechnoratiAdd to FurlAdd to Newsvine

The Secret History of Silicon Valley Part VIII: The Rise of Entrepreneurship

This post makes sense in context with the previous post.

——–

The Korean War catapulted Stanford University’s Electronics Research Laboratory (ERL) into a major player in electronic intelligence and electronic warfare systems. Encouraged by their Dean, Fred Terman, scientists and engineers left Stanford Electronics Research Laboratory to set up companies to build microwave tubes and systems for the military. Funded by military contracts these 1950′s startups would help build Silicon Valley’s entrepreneurial culture and environment.

The Beginnings – “Vacuum Tube Valley” Ecosystem circa 1950
From its founding in 1946 Stanford’s Electronics Research Laboratory (ERL) did basic research into vacuum tubes that could operate at microwave frequencies. The research was funded and paid for by the Office of Naval Research (ONR) and later by the Air Force and Army. Much of the basic research work was done by advanced students or by recent Ph.Ds doing postdoctoral internships, supervised by Stanford engineering faculty members or senior research associates (staff.)

In a 1950 proposal to the Navy Fred Terman noted that the work that Stanford proposed “correlates almost ideally with related industrial activities in this area.”  There were already “tube manufacturers in the area (Eitel-McCullough, Litton Industries, Varian Industries, Henitz and Kaufman and Lewis and Kaufman) that represented an integrated set of tube facilities for basic research, advanced development, engineering of new tubes, model shop and pilot and quantity production. And that circuit work is carried on by several organizations in the neighborhood, with Hewlett Packard Company being especially notable in this regard.” Terman was describing the valley’s already existing ecosystem for building vacuum tubes in 1950.

But unlike the majority of existing tube manufacturers in the valley who were making products for radios, Stanford Electronics Research Lab tube group had a special customer with very special needs – the U.S. Air Force and its Strategic Air Command.

So what exactly was the Electronics Research Lab designing? What were these microwave tubes? Why were they so important to the military? And what were these electronic intelligence and warfare systems used for?

Stanford Joins the Cold War - Microwave Power Tubes
Stanford’s work in microwave power tubes would solve two of the Strategic Air Command’s most important cold war problems.

During a nuclear war in the 1950′s the Strategic Air Command was going to fly its bombers with nuclear weapons into the Soviet Union. To protect their country, the Soviets were building an air defense network to warn, track and destroy these attacking bombers. Our bombers used jammers to confuse the Soviet air defense radars. But the jammers that we built in WWII were no longer sufficient to protect the planes we wanted to send into the Soviet Union.

These 1940′s jammers (built by the wartime lab headed up by Terman and his team now at Stanford) had been built around tubes originally designed for radio applications, put out 5 watts of power. This miniscule amount of jamming power was acceptable because each WWII bomber flew in formation with hundreds of other planes, together attacking just a single target each day. The combined jamming power of all the bombers on a mission was enough to saturate and confuse German radar. But in a potential cold war attack on the Soviet Union, our bombers were not going to fly in a massed formation to attack one target. Instead we would attack multiple targets in the Soviet Union at the same time.  And while a few bombers would penetrate the periphery of the Soviet Union together, each plane — now able to carry more explosive power than all the bombs dropped in WWII — would approach its target individually. As a result of this change in strategy (and explosive capacity), each bomber had to supply enough jamming power to defend itself.

 

B-47 - primary Strategic Air Command Bomber in the 1950's

B-47 – primary Strategic Air Command Bomber in the 1950′s

As a result, to protect its bombers flying over the Soviet Union the U.S. Air Force needed power tubes that had hundreds of times more power than WWII devices.

The U.S. Air Force also needed improvements in frequency agility to protect its cold war bombers. Frequency agility can be best described by what happened over Germany in WWII. As the allies jammed Germany radar, the Germans tried to avoid the effect of jamming by changing the frequency on which their radars transmitted. This was possible since the jammers in U.S. planes’ could only transmit on a narrow band of frequencies (providing spot jamming) and could not be retuned in the air. To cover all the possible frequencies German radars might be operating on, allied technicians pretuned the jammers before each bomber raid so that each plane transmitted on a different frequency. The combined effect of hundreds of planes in the bomber stream was to cover a broader frequency range than one jammer could by itself.  (This technique of covering a broad range of frequencies was known as barrage jamming.)

(A good Radar tutorial is here, on the Radar Range equation here and Electronic Warfare tutorial is here. The links will download PowerPoint presentations.)

 

But nuclear warfare over the Soviet Union in the 1950′s meant that a single bomber needed jammers that could cover multiple frequencies, and could be tuned instantaneously. Not only did the US need more more powerful microwave power tubes, the power tubes had to be frequency agile, (able to be tuned in the air to different frequencies) to jam the Soviet radars. (For example, the Soviet P-20 Token was an early warning radar our bombers would encounter.  It transmitted on 5 different frequencies over a band 300mhz wide. To jam it, all five frequencies had to be jammed at the same time. Our WWII jammers couldn’t do the job.)

Terman’s Systems Engineering Research Lab at Stanford would develop microwavepower tubes that offered a solution to both challenges and would be a a game changer for electronic warfare at the time.

High Power, Instant Tuning – Stanford’s contribution
Stanford’s Electronics Research Laboratory first contribution to high power microwave tubes for airborne electronic warfare in the 1950’s was the Backward Wave Oscillator (BWO). Stanford engineers realized that this tube, which had been invented in France, could electronically tune through microwave frequencies while producing almost a 1,000 watts of power – (equivalent to the output of 200 jammers over Germany in WWII.) Perfecting this tube for use as an airborne jammer became one of the labs primary objectives.

This was a critical development to support the new tactics of single bombers penetrating the Soviet Union. Equipping a bomber with several jammers built around Backward Wave Oscillator could give it enough power to use barrage jamming against multiple radars and get it through to its target.  Stanford gave its Backward Wave Oscillator design drawings to tube manufacturers throughout the U.S. By the 1960′s, the U.S. Air Force would ultimately equip its B-52 bombers with 6,000 jammers using these these oscillators.

The Rise of “Microwave Valley” Stanford Tube Spinouts
A technician in Stanford’s ERL tube shop, Ray Stewart, thought he could build these Backward Wave Oscillators commercially, and left to start Stewart Engineering in Scotts Valley near Santa Cruz.  The company had more orders from the military than it could handle. (Stewart would sell his company to Watkins Johnson, one of the most financially successful of the Stanford microwave tube spinoffs. More about Watkins-Johnson in the next post.)  Stewart joined a growing list of other microwave startups beginning to populate the valley.

One of the early microwave spinouts from Stanford was built around a microwave power tube called the Klystron, invented by Terman’s students Russell and Sigurd Varian and William Hansen. In 1948 the Varian brothers along with Stanford professors Edward Ginzton and Marvin Chodorow founded Varian Corporation in Palo Alto to produce klystrons for military applications. (Fred Terman and David Packard of HP joined Varian’s board.) While the Klystrons of the 1950’s had too narrow and bandwidth and were too large for airborne use, they could be scaled up to generate megawatts of power and were used to power the U.S. ground-based Ballistic Missile Early Warning System (BMEWS) radars (and the Stanford Linear Accelerator.)

klystronAnother of Terman’s students, Charles Litton, would start several Silicon Valley companies, and in the 1950’s Litton Industries would become the leader in pulse and continuous wave magnetrons used in jammers and missiles. Magnetrons were the first high power microwave device invented in WWII. Used in radars systems and missiles, magnetrons could produce hundreds of watts of power.

More to Come
These first microwave tubes were just the beginning of a flood of innovative
products for the military.  The next Stanford tubes and systems would revolutionize the Electronic Intelligence aircraft that were circling (and flying over) the Soviet Union.

More in the next post, Part IX of the Secret History of Silicon Valley.

The Secret History of Silicon Valley Part VII: We Fought a War You Never Heard Of

These next series of posts chronicles the untold story of how one professor returning from one war decides to enlist Stanford University in waging the next one and by accident, laid the foundation for Silicon Valley, venture capital and entrepreneurship as we know it today.

These posts cover two distinct periods – the first, the rise of “Microwave Valley” chronicles the decade of 1946-1956 as Stanford University became the hub of military/industry contracting in the Bay Area.

The second series of posts, the rise of  “Spy Satellite Valley,” starts in 1956 with two game changing events– one very public – the valley’s first semiconductor company and one very, very private – the valley as the home of the first optical and ELINT spy satellites and submarine-launched ballistic missiles.  The story ends in 1969 with campus riots at Stanford.

————–

These posts will make a lot more sense if you look at the earlier Secret History posts.  If you read only one previous post, read this one (or this one.)

————-
The Birth of Entrepreneurship in The Hot Cold War
Silicon Valley entrepreneurship was born in the middle of a secret war with the Soviet Union. It’s a war you probably never heard of since most of it was classified, and both parties never wanted it public lest it got out of hand.  Yet it was a war in which tens of thousands of Americans fought and hundreds died. Frederick Terman, Stanford’s Dean of Engineering, enlisted Stanford University as a major arms suppliers in this war. In doing so he accidentally launched entrepreneurship in Silicon Valley – with the help of the U.S. military, the CIA and the National Security Agency.

Stanford as a Center of Microwave and Electronics
In 1946 after running the military’s secret 800 person Electronic Warfare Lab at Harvard, Fred Terman returned to Stanford as the dean of the engineering school. Terman’s goal was to build Stanford’s electrical engineering department into a center of excellence focused on microwaves and electronics. Having already assembled one of most advanced electronic labs in World War II, Terman was one of the few academics who could do it.

terman

Fred Terman

Terman’s first step was to recruit 11 former members of his staff from the Harvard Radio Research Lab — “Congratulations, you’re now Stanford faculty.”  Not only were they all great researchers, but they also had just spent three years building electronic warfare systems that were used in World War II. They would become the core of Stanford’s new Electronics Research Lab (ERL.)  While officially in the electrical engineering department, the lab reported directly to Terman.

Next, Terman used his military contacts to secure funding for the Lab from the Office of Naval Research, the Air Force and the Army Signal Corps. (Although the country had returned to peace, some in the military wanted to preserve our ability to fight the next war.) By 1947 the U.S. military was funding half of Stanford’s engineering school budget. Terman proudly pointed out that only Stanford, MIT and Harvard had a military sponsored electronics program.

Stanford Leads in Electronic Intelligence and Electronic Warfare
In the 1950’s Stanford Engineering Research Lab (ERL) made major contributions to electronic intelligence and electronic warfare.  Its basic research focused on three areas: microwave receiving and transmitting tubes, radar detection and deception techniques and understanding the earth’s ionosphere.

Stanford became one of the leading research centers in advancing the state of microwave tubes including the klystron which could provide high-power microwave in pulses, magnetrons which could provide continuous wave microwave power, and backward wave oscillators and traveling wave tubes – both electronically tunable microwave tubes.

Stanford’s research on the earth’s ionosphere would lead to meteor-burst communication systems and Over the Horizon Radar used by the NSA and CIA to detect Soviet and Chinese missile tests and ultimately to the research that made Stealth technologies possible.

Its studies in radar detection and deception techniques would lead Stanford to the applied part of it mission.  Stanford would build prototypes of electronic intelligence receivers (high probability of intercept/rapid scan receivers) for use by the military. These applied systems were prototypes of the jamming devices found on our bombers and receivers found in NSA ground stations and the fleet of ELINT aircraft flying around and in the Soviet Union and later on in the U-2, SR-71 and ELINT ferret subsatellites.

Later posts will talk about these technologies and the startups that spun out of Stanford to build them. But first, to understand what happened at Stanford and in Silicon Valley under Fred Terman, some context about the Cold War is helpful.  (Skip the next section if you’re a history major.)

The Cold War
After World War II ended, our wartime ally the Soviet Union kept its army in Eastern Europe and forcibly installed Communist governments in its occupied territories.  Meanwhile the U.S. demobilized its army, sent its troops home, scrapped most of its Air Force and mothballed almost all its Navy. As tensions rose, there was a growing fear that the Soviets could invade and occupy all of Western Europe.

In 1949, the Soviets exploded their first nuclear weapon and ended the U.S monopoly on atomic weaponry. That same year China fell to the communists under Mao Zedong, and the Nationalist government retreated to Taiwan.  A year later the Korean War turned the cold war hot, as communist North Koreans attacked and overran most of South Korea (except for a small defensive perimeter in the south.) American and United Nations troops entered the war fighting North Koreans and then Communist Chinese ground troops, and Soviet fighter pilots for three years.  34,000 U.S. soldiers died in battle.

To the U.S. the Soviet Union seemed bent on world conquest with Korea just a warm-up for an atomic war with massive casualties. (This was not an unreasonable supposition after a conventional world war which had left 50 million dead.)  Faced with the reality of the Korean War, the U.S. began to rebuild its military. But now the Soviet Union was its target enemy, and nuclear weapons had become the principal instrument of offense. Instead of rebuilding its WWII forces, the U.S. military embraced new technologies (jets, electronics, missiles, nuclear subs) and built entirely new weapon systems (bombers with nuclear weapons, ICBMs, SLBM’s) for a new era of international conflict.

Europe, completely outnumbered and outgunned by the Soviet Union, built the North Atlantic Treaty Organization (NATO) as a bulwark against ground Soviet attack.  And the U.S. planned strikes with nuclear armed bombers if war in Europe broke out.

Stanford’s Electronic Research Lab (ERL) which had focused on basic research on microwave tubes from 1946 was about to scale up for the Cold War.

Smarter Intelligence
One of the major differences between the war with Germany and the cold war confrontation with the Soviet Union had to do with access.  The Soviet Union was a closed country. Unlike Germany in World War II, the U.S. could not fly across the Soviet Union to learn how their defenses were set up. We did not have radar maps of their cities. The Soviet’s secrecy fed our cold war paranoia. The U.S. was determined to find out what was going on inside.  And the way we were going to do it was with electronic/signals intelligence.

But the technology that supported intelligence gathering against our WW II enemies was not sufficient to penetrate the Soviet Union. The U.S. military had to develop new ways to collect intelligence. The engineering department and labs that Fred Terman established at Stanford University would play a key role in advancing electronic intercept and jamming technology to support the more sophisticated intelligence systems that the Cold War required.

The Air Force Needs to Know
By the Korean War, U.S. policy held that the Air Force, carrying nuclear weapons into the Soviet Union, would be the means to fight World War III.

Through World War II, the U.S. Air Force had been a part of the U.S. Army. It split off into a separate service in 1947. By the 1950’s, the Strategic Air Command (SAC) had become the U.S. Air Force’s long range bombing arm and the designated instrument of Armageddon.

On the other side of the Iron Curtain, the defense of the Soviet Motherland lay with the Soviet Air Defense Forces, called PVO Strany, a separate branch of the Soviet military formed in 1948 designed to detect U.S. bomber raids, target and aim radar-guided weapons and destroy the U.S. bombers.

Example of Radar Coverage - Japan in WWIIExample of Early Warning Radar Coverage – Japan in WWII

SAC needed intelligence to understand the components of the PVO Strany air defense system in order to shut them down and make them ineffective so our bombers with their nuclear payloads could reach their targets. (The information we collected would be passed on to contractors who would build jamming devices the bombers would carry.) It sought answers to tactical questions like: What was the Radar Order of Battle a penetrating bomber would face? (Were there holes in their radar coverage our bombers could sneak through? What was the best altitude to avoid the Soviet defenses?) What were the different types of Soviet fighter planes?  How many?  How effective? What about the anti-aircraft (AAA) gun defenses?  In addition the Soviets were adding a new type of defensive radar-guided weapon called the Surface to Air Missile (SAM).

Example of Jammer versus Radar Coverage- Germany in WW11

Example of Jammer versus Radar Coverage – Allied Jammers over Germany in WW11

The Strategic Air Command also needed to know what the navigational waypoints and the target would look like on their air-to-ground bombing radars. (These radars painted a map-like picture of the ground and prior to GPS, this is how bombers navigated their way to the target.)

And on top of all this the Strategic Air Command needed to understand the current state of the Soviet Air Defense Force readiness and deployment updated on a daily basis.

The CIA Needs to Know
While the Air Force was working on collecting intelligence to execute their tactical missions, the CIA, founded in 1947, was responsible for providing U.S. political leadership with a much bigger picture. They developed the National Intelligence Estimate –  a  series of reports which summarized their judgment about the size of the Russian threat. Also seeking to learn more about the Soviet Union’s offensive weapon systems, the CIA wanted intelligence to help them understand: What type of strategic bombers did the Soviets have? How many did they have? How would they reach the U.S.?  How would we know if they were coming? (Have they moved to their forward operating bases in the Artic?) The same was true about the Soviet defensive systems – how many fighters would they build and of what type?  How many Surface to Air Missiles – what was their range and accuracy?

And by the mid 1950’s the Soviets were testing ballistic missiles, both intermediate range that could reach Europe and intercontinental range that could reach the U.S.  Whatwas their range? What was their accuracy? How big of a nuclear warhead could they carry (throw weight and yield)?  The military needed to answer these same questions about the nuclear armed missiles the Soviets were putting on their submarine force.

To give our leadership an estimate of the Soviet’s nuclear production capacity, the CIA also had to estimate how many nuclear weapons could the Soviet Union make. Where were their production facilities? What was the yield of the weapons and their weight and size?

Throughout the 1950′s the CIA’s Office of Scientific Intelligence was heavily involved in the development of Electronics Intercept and Electronic Warfare Intelligence – and Stanford and the emerging startups around it would provide the systems and concepts to help.

The NSA and ELINT
In the 1950’s the Strategic Air Command and the U.S. Navy were the airborne ELINT assets for the U.S. Beginning in the mid/late 1950’s the National Security Agency (NSA) starting taking more and more responsibility for collection – first in communications intelligence, then in signals and telemetry intelligence. The NSA ultimately built up hundreds of ground stations, satellites and aircraft manned by tens of thousands servicemen (under the cover of the Air Force Security Service, Army Security Agency and the Naval Security Group.)

The “Hot” Cold War
Remember the Soviet Union was a closed country. To collect the intelligence to answer its questions about the Soviet threat, the U.S. military resurrected the signals Intelligence lessons and skills we invented in World War II. Starting in 1946 ELINT aircraft had been probing and overflying the Soviet Union. SAC, the CIA, the Navy and our British allies flew modified planes called Ferrets around the periphery of the Soviet Union to understand their air defense system (the crews were called Crows). (What isn’t well known is that the U.S. and Britain flew planes on deep penetration missions into and across the Soviet Union numerous times – well before a U-2 spy plane was shot down over the Soviet Union in 1960.)

British Canberra PR3 - Overflew Kapustin Yar 1953

British Canberra PR3 – Overflew Kapustin Yar 1953

The Air Force adopted a cover story that these were weather data gathering missions. These flights were no secret to the Soviets, (given the sheer number of surveillance flights around the Soviet Union it’s surprising they didn’t need their own air traffic control system,) and they started to protest diplomatically in 1948. When our flights continued, the Soviets took direct action. In 1950, two months before the Korean War started,  the Soviets shot down an ELINT plane over the Baltic. All ten crew members were killed. This was the beginning of a Soviet policy to stop ignoring incursions. They would attempt to force the ELINT planes to land in the Soviet Union or they would destroy them. Every year through the the 1950′s and the early ’60′s the Soviets attacked and shot down at least one of our ELINT ferret aircraft. This was a deadly game.

pb4y-2 Privateer Navy ELINT

pb4y-2 Privateer Navy ELINT

We kept on probing their defenses convinced that it was in our national interest to continue. The low-level conflict continued until the height of the Cuban Missile Crisis when the local Soviet commander shot down a U-2 over Cuba. Both countries realized that a miscalculation could have been a catalyst for World War III and the Soviets stop attacks on U.S. spyplanes. (The Communist Chinese continued to shoot down U-2′s flown by Nationalist Chinese pilots until 1970, and the Soviet Union accidently attacked two Korean airline passenger planes in the Far East, one damaged in 1978 and one destroyed in 1983.)

During the Cold War 32 U.S. ELINT planes were shot down by Soviet pilots with 225 U.S. airmen killed. (The numbers vary depending on the sources you read.) Regardless of the number, this was a deadly shooting war.

Stanford and an emerging set of Silicon Valley startups would be deeply involved in designing the technologies, techniques and ELINT systems on these planes. Microwave Valley was about to take off.

Details in the next post, Part VIII of the Secret History of Silicon Valley.

Follow

Get every new post delivered to your Inbox.

Join 137,441 other followers