Dan Stern– Engineering Inhouse/ Manufacturing Outsourced


Editor’s note-  Dan Stern is a remarkable man.  He designs and sells a number of specialty car parts, specifically around automotive lighting.  Dan is a regular wherever people are discussing the finer points of automobile engineering.  He is the editor of  Driving Vision News and the proprietor of Daniel Stern Lighting Consultancy. 

Dan has a point of view and an opinion or two.  His respect for quality engineering makes him an inspiration for those that know him. He has a special regard for vintage Chryslers and their engineering.  He works to source and integrate parts from others and designs and manufactures some parts himself.   When I have been a customer of his, I get more than a product that he helped engineer or sells, but an education as to how it works, and why it matters. 

This interview is important, because it shows a common pattern– Keep the engineering in-house, and find good partners to build your product.  Jonathan Werner uses this approach, and Chris Anderson has written extensively about this as well.  In some cases, this has proven to be a good way for a Manufacturing Entrepreneur to build his company.

Tell us a little bit about what you make.

I make easily-accessible explanations for difficult-to-grasp ideas. My product is expertise to help people understand the intricacies of automotive lighting. It’s a field that seems simple on first blush; people tend to think they’re just lights, either they work or they don’t, and that’s about it. But in fact, there is an enormous amount of physics and engineering and physiology—and even psychology and philosophy—in just about every aspect of the design, performance, and regulation of every lamp, light, and reflector on a roadgoing vehicle.

When did you start making this? 

I’ve been at it for most of two decades.

How long did it take you to develop it? 

Most of two decades! This is not a static subject that can be learned once and taught indefinitely on that basis.

What changed along the way? 

Almost everything. The technology and technique in headlamps and other automotive lighting is presently undergoing a near-total revolution as LEDs displace older light sources. For the first time, both the light source and the optics used to gather, focus, and distribute the light are changing at the same time. But while this is certainly the biggest revolution the field has seen, it’s not the first. We’ve had new light sources before: high-intensity discharge “Xenon” replaced halogen, which replaced tungsten, which replaced burning wicks. And we’ve had new optics, too: projectors, condensers, light guides, complex-surface reflectors. New materials have come in; glass and metal have given way to a variety of advanced polymers and other engineered materials. Computers and cameras have given us smart car lights that can follow the curve of a road and otherwise adapt the lighting performance to match prevailing conditions. On a parallel track, the regulations for car lights have had to change to accommodate the new technology and its performance potential. And that has spurred philosophical discussions about the finer and grosser points of regulation itself.

Did you run into times that you needed to redesign part or all of your product?

That’s a constant given. My range of products is diverse, because my clients are diverse; there’s no one-size-fits-all. The law enforcement agency trying to understand which cars to pull over for dangerous lights, the attorney working on a case involving vehicular lighting, the guy who wants to be able to see better while driving at night, the working group or government agency trying to update a technical standard or a regulation or a law, the lighting supplier considering additions to its product line…each of these different kinds of client needs the right set of information, presented the right way. So I have to tailor my product to its intended use whether I’m serving as an expert witness in a court case, or as a freelance consultant, or as an answer-man on a podcast or call-in show or column, or as a writer hired to produce a report on some aspect of the automotive lighting field, or in my positions on various technical standards boards, or in my role as General Editor of DrivingVisionNews, the global automotive lighting and driver assistance industry journal.

How did you figure out your manufacturing process to make this?  Have you changed it since you first laid it out?  Did you consider outsourcing or off shoring?

The irreplaceable basis of the process is comprehensive knowledge of the subject matter.It’s possible to bluff it effectively enough to make the cash register ring at a retail level—there’s a lot of that out there—but when the product is billed and reputed as expertise, sought by and marketed to those relying on the information to be all the way complete and correct, there is no substitute for thorough knowledge. And it can’t be one-dimensional; it’s not possible to give a usefully broad or deep explanation of the differences in American and European and Asian automotive lighting practices, for example, if one has only studied the field from within the American context. Beyond that foundation, the product development process is adaptation all the way. Adaptation to changing technology, changing facts, changing queries, and changing interpretations.

Outsourcing? Sure, there are services out of India that will generate a report or technical paper or whatever you want. It’s a great way to create a murky whirlpool of words on any subject, but that’s the direct antithesis of my product, which is clarity. It’s the same with product-development work; it’s possible to farm out the design of a headlamp (turn signal, fog lamp, whatever) to a job shop in China, say.

Outsourcing like this can be made to look like a great savings on paper…but only if we pretend that price and cost are the same thing. They are not! An outsourced technical report or a headlamp design-and-tool job with an attractive low price often winds up with a staggeringly high cost. Not only is the muddling effect of a language barrier greatly magnified when discussing a highly specialized technical subject, but a huge proportion of the outfits offering tempting services can’t deliver on their promises. I was doing a product-development job about five years ago on a line of motorcycle headlamps, and the boss said they had to be sourced in India or China for the low price. Every job shop I sent specs to would always come back and say they could do the job to an excellent standard, no problem, “sign here”. I drafted an unbuildable specification—a spec for a lamp that physically could not be manufactured—and sent it around. More than 9 out of 10 job shops still came back with “Yes, no problem, sign here”. Those very few shops that expressed some doubt about the spec were the only ones worth trying to talk to, and even most of those failed miserably. They just didn’t have the expertise and ability they claimed to have. On that project alone, the requirement to go to a “low-cost country” drove the cost of the project up far beyond what it would have taken to have the lamps designed and tooled in a top-flight American facility. And that’s just one example.

How did you find capital to get started? 

Incrementalism, carefully controlled to remain within my actual means all along. I started out selling vehicle lighting equipment out of my college dorm room—headlamps and such—that I had evaluated and handpicked for being the best in their category. That’s how I started growing my reputation: headlamp customers took a chance that I was telling them the truth, bought what I said to buy, and found their night-driving problem completely solved. A good reputation began to develop, and then I did it again with specifiers, as when a truck builder asked which lamps should go on a fire truck being developed for the Australian market. Then I began gradually shifting towards a knowledge-based product, as when the government of a country in Southeast Asia wanted to know how they should update their headlamp regulations. Incrementalism, never spreading myself too thinly.

If you could go back in time, and talk to yourself at the beginning of this adventure, what you advise yourself to do?   You must have learned a lot since then, so what lessons would you say stick with you?

Probably the main thing is that automotive lighting is a pretty rarefied field, and most of the participants are at least aware of one another. I think I would go back and tell myself to do a better job, in the early years, of allocating less of my time and attention to the big talkers, and more of it to those few bright stars in the field who really know what the hell they’re talking about, really know what they’re doing. And then I’d tell myself, once I sorted out whom to listen to, to talk less and listen more. Pretty standard older-self-to-younger-self advice, I suppose, but the Dunning-Kruger Effect is a big impediment to effective learning and teaching alike: we humans are poorly equipped to know just how much we don’t know, and the less we know, the more we think we know. I try to keep that in mind and never stop learning.

What is the worst advice that somebody gave you as you were building your operation?

“The customer is always right”. That’s terrible advice! Actually the customer is frequently wrong. by dint of ignorance. That might sound judgmental, but only because “ignorant” is often used as an insult. We have to keep in mind that there is no shame or fault in ignorance as such, and it is easily curable with knowledge and information. The fact that the customer seeks expertise or some other service or product amounts to “you have something I need; you know something I need to know”. There’s always an element of education involved, whether the product is pure knowledge, information analysis, some tangible product, or whatever else. The thoughtful customer seeks to replace guesses and assumptions with knowledge and facts, and the thoughtful provider seeks the same result.

Trouble comes in the form of the customer who isn’t really shopping for knowledge or information or a product, but rather seeks reinforcement of existing guesses and opinions and preferences. This buyer is determined not to let mere facts topple guesses, preferences, folk “wisdom”, and baseless conclusions. The provider must make a sincere good-faith effort to provide what the customer actually needs, but when faced with a customer who has decided to prioritize his ego over his project, the thoughtful, conscientious vendor does not—cannot afford to—cater to the customer’s jealously-guarded ignorance. There’s nothing to gain once the would-be customer insists that his preferences and opinions must take precedence over the facts, and there is much to lose. The thoughtful vendor, in fact, has no qualm about promptly telling such a customer or client to take his business elsewhere.

My grandfather, a public-utilities regulation consultant renowned internationally, gave much better advice. He told his clients and customers, “You are entitled to an opinion; you are not entitled to an uninformed opinion”. That sounds harsh, but it’s critically important: each of us, whether we’re the maker or the vendor or the consultant or the customer or the client, is entitled to his or her own opinion, but not to his or her own facts. When that basic reality is disregarded in favor of a bromide like “the customer is always right”, everybody’s actual, real interests go unserved—supplier and customer alike.

How do you innovate? When you get the next great idea for a product, or improving an existing product, where do the ideas come from, and how do you make it real?

The answer here ties in with what I’ve been saying throughout this whole interview: never stop learning. It’s not enough to keep up with the developments in the field itself; it’s crucial to continually seek new ways of looking at facts, new ways of thinking about existing questions, new ways of looking at existing reality. Inspiration can hit anywhere, at any time. In the shower, or while stuck in traffic, or on the way to the fridge for a midnight snack, sure, but also when talking and listening with people. It’s really important to listen not only to my direct buyers, but also to the people whose lives are affected by what happens in my field of automotive lighting.That’s pretty much everyone, since most of us drive and just about all of us interact in some fashion with cars every day. Everyday viewpoints and experience are vital to hear and understand, and they don’t filter up into the “expertsphere” on their own—you have to go strike up conversations, then close the mouth and open the ears.

But that “never stop learning” coin has another side, too: never stop teaching! It’s not enough to just pump people for info, listen and nod and go “Mm. Mm-hmm. Very interesting. Thanks for your thoughts.” As new viewpoints and perspectives get integrated into the knowledge, and thence into the product, it’s important to give back, to work towards greater accuracy of information on the subject accessible to the general public. Work at it hard enough, consistently enough, and eventually you get a nice positive feedback effect. The basic quality of people’s questions starts to get noticeably better. Facts and science start to edge out guesses and suppositions in the general chatter on the subject. It’s not only heartening to see and hear, but it lets more people make better decisions. Everyone wins.

How do you connect with your customers and fans?  How did you build a community of people interested in your product and ideas?

The means, modes, and methods of connection have to match up with whoever I’m trying to reach. It doesn’t work to plop a product down and say “Here y’go, come and get it”. For the commercial side of things it’s word of mouth, online and offline. I’ve never bought an advertisement of any kind, and I deleted my Facebook account three years ago—don’t miss it, either; that turned out to be a great decision. My ability to buy groceries and keep the roof over my head and the floor under my feet lives or dies by my reputation…now that’s motivation!

As for building and keeping an interested community: that comes back to engaging directly with the community. That’s central and vital, but a lot of business don’t seem to think so. They make it impossible for users (or non-users!) of their products to get in touch and share their thoughts. You can’t call and talk to them. You might be able to send e-mail, but usually the best you can hope to get back is a bland, unresponsive “thank you for contacting us” from someone unequipped, uninterested, and unauthorized to engage meaningfully.

I really think that’s the wrong way to do it. When I was a teenager, long before I started any kind of business activity, I read Guy Kawasaki’s book “The Macintosh Way”. I was much too young to grasp much of what he presented, but two crucial concepts stuck with me: doing the right thing’s useless if it’s not done the right way, and personal engagement is centrally, utterly important. That second point was really hammered home for me when I finished the book and came upon Mr. Kawasaki’s phone number, right there in plain text. I called him up and he answered the phone and talked to me. He didn’t condescend or patronize, he really talked with me. It was as substantial a conversation as a highly successful executive and a 14-year-old could have had, and it really stuck with me: whoever sends me an email gets a personal reply from me, and I answer my own phone.

At the same time, there’s a tendency for individual participants in an industry to clam up and refuse to talk to other participants. Of course thoughtful discretion and diplomacy are needed; nobody wants to give away the proprietary secrets, and making unkind remarks about one’s fellows in the industry is seriously unwise. But there are huge benefits to intra-industry communication and community at a much higher level than we tend to see, especially in North America. Hey, there’s a trade journal and a trade association for just about any industry you care to name: the people who make bottles and cans have one. The people who make art supplies have one. But until a few years ago, there was none for the automotive lighting and driver assistance industry. Now we have one, DrivingVisionNews. It got started in Europe, where there’s less hesitation to form a community wherein people actually talk and listen with each other, even though they work for competing companies. Now we’ve got a lot of American and Asian companies on board, we’re holding workshops and panel talks and round-table discussions to capacity crowds, and the benefits of the improved communication are starting to be felt. Whole sectors of the industry, instead of remaining constrained by outmoded regulations blocking really good innovations from reaching the market, are putting their heads together. They’re paddling the canoe in the same direction, talking to regulators with a coördinated voice, and things are moving in a productive direction as never before. This is much better!

How do you do design work vs sourcing?  When do you say- “ I know what I want, and I need to design it and send it out for production” vs. “I can make this product work exactly the way that I want””?   

There’s no firm heuristic for this decision; it really is on a case-by-case basis. A lot of it depends, casewise, on whom I’m working with, in what context, and what constraints exist in terms of time, cost, confidentiality, etc.

What can you tell me about finding good manufacturers?  How do you know that you found one that you like? 

I like to do careful evaluation of a company’s existing products. Obviously there are objective tests for compliance with whatever standards and regulations are applicable, and those are a good go/no-go sorting criterion. But most standards and regulations in my field don’t require a good product, just a compliant one. So it becomes important to scrutinize the company’s products subjectively, too. A hold-in-the-hand test can be very revealing, if done by someone who knows what to look for, what’s good, and what’s not.

Likewise, it’s important to interact with the company at as many levels as possible. How quickly, accurately, and usefully do they reply to inquiries from potential business partners? From potential customers? From existing customers? If they’re slow and vague to respond to an enquiry before the project starts, they probably won’t speed up once it’s under way.

Last on my list is scrutinizing their certifications and processes and suchlike. Of course I want to know about the processes and protocols a company has in place, but I am not a big fan of things like the ISO 9000 and ISO 14000 series of quality assurance and quality control certification schemes. I just don’t think they’re anywhere near as helpful as they’re promoted to be. Of course, being human, I’m subject to confirmation bias just like everyone else, but in my experience there’s no reliable correlation between ISO certification and the ability and consistency of a company in producing a quality product. I’ve seen consistently excellent products from uncertified companies, and I’ve seen garbage from certified companies.

Do you have any horror stories about a bad vendor or source? 

Unfortunately so. In that product-development work I mentioned when you asked about outsourcing and offshoring, I did not have free reign to choose vendors and sources. One of the project parameters was “product must be sourced from a low-cost country”, which in that particular case was code for China or India. I had severe qualms about this, because all my research and scrutiny told me it was not possible to get a satisfactory product of this type out of those countries.  Nevertheless, that edict stood, so I made my objections known in writing, picked a vendor in India I hoped could be coached adequately to produce something at least usable, drew up a very highly detailed specification spelling out every last little detail in words of one syllable along with all the requisite drawings, crossed my fingers, and forward we went. Only we didn’t go forward! It was more like we were jammed in Neutral. What should have been a four- to six-month development timeline dragged out to two and a half years. There were long periods of time with no contact from the vendor—emails not returned, faxes not acknowledged, phone calls deflected. Every once in awhile they would send an unacceptable sample; we’d evaluate it, list and describe the faults for them, and after a few months they’d send another sample…with one of the listed problems partly addressed and three new ones. Lather-rinse-repeat! It was ridiculous, especially since there was nothing innovative about the product we needed. This was in 2005 or so, and it was utterly standard, basic technology—state of the art circa 1972 all over the world.

Eventually it was decided I should travel to India to see what was the matter. The vendor had me to their factory, we sat down at a conference table with cookies and tea, and they finally explained why they hadn’t sent a satisfactory sample: they’d never received such a closely detailed specification from a customer, they said, which meant we were the most knowledgeable customer they’d ever had, they said, and so they felt timid about asking questions, they said. 30 hours in Delhi sandwiched between 17-hour plane flights for that lame excuse! In the end we never did get a satisfactory product out of them. We had to unwind the deal and nobody wound up happy.

What would you tell somebody that wants to do something like what you are doing?  Maybe it is automobile brakes, instead of lights—The pattern you built is what I am trying to suggest is repeatable.

Don’t overlook the ABCs! I see so many people screw this up; they get so enthralled with their gimmicks, slogans, novelties, and innovations that the basics get done in a slapdash, minimally-satisfactory way (if at all). The primary main objective of a headlamp is to light the driver’s way safely through darkness—not to look cool. The primary main objective of brake parts is to stop the car quickly and safely—not to come in a sexy box advertised on a fabulous website. There are tons of vehicle lighting products marketed with all kinds of claims and hype, expensive packaging, catchy slogans and promos, but just try asking how well they work in terms of objective safety performance. If you get a meaningful answer at all, it’s something like “Oh, they fully comply with all applicable regulations”. Again, the regs don’t require a good product, just a compliant one, so that’s an evasive non-answer. The same goes whether the product is a tangible item (headlamp) or an intangible (knowledge). Don’t be that kind of consultant who borrows the client’s watch and charges him to write up a report saying what time it is—have the passion, knowledge, and resources to give the client genuinely new and pertinent information, or do something else for a living.

And treat your customers as intelligent grownups. The market is crowded with fluff and garbage, and there are a lot of buyers for it. But there are also a lot of adults out there, and adults have more discretionary money than kids. They’re tired of being lied to, tired of being sold shoddy products whitewashed with lavish warranties, tired of feeling bewildered by a mountain of hype, tired of being the victim of “if it’s priced this high, it must be good!” games. These are people who want to see as well as possible at night, or want the best possible brakes (windshield wipers, whatever). They don’t care if it comes in a plain brown box, they’re not impressed with infomercials, they just want an honest deal on an honest product. The marketing decisions needed to make a successful go of such a strategy—here again, whether it’s a tangible product or a knowledge-based venture—are sometimes counterintuitive, but this market segment will always exist, even if it’s invisible to some of the louder marketeering strategists.

Where will your market, product, or business be in 5 years?

Well, the technology is advancing at an unprecedented fast pace. And the rate at which that pace is accelerating is also faster than ever seen. Every time we turn around and blink, LEDs are getting brighter and cheaper—and that’s just one example. There are new mirrors that give a broader field of vision without distorting the view. There are tiny cameras so sharp and fast that whole new driver assistance systems are possible that we couldn’t have imagined just a couple of years ago. In five years’ time, we’ll be thinking much less in terms of separate vehicle lighting systems and driver assistance systems, and more in terms of integral driver vision systems. Eventually cars will drive themselves and we won’t have much need of headlamps, stop lamps, turn signals, and suchlike. But that day’s not coming for quite awhile, so if you’ll excuse me, I’d better go get back to work on an analysis of how different types of headlight bulbs affect headlamp performance!

Posted in Uncategorized | Leave a comment

Jim Meyer — “Innovation happens when the constraints are present. “

Editor’s note– James Meyer’s company, Quarq ( now a part of SRAM LLC) is a customer of Ursa Information System, and a user of OpenERP.  We met Quarq about 1 year ago, and were incredibly impressed with their product, and their staff. They are a great manufacturing company, making high end parts for the Bicycle racing community.  Additionally, they are located in Spearfish, South Dakota, in the beautiful Black Hills. 


Jim is Quarq’s co-founder and now Technology Director. Jim has an undergraduate degree in mathematics from Rose-Hulman Institute of Technology in Indiana. After Rose-Hulman, Jim attended the Massachusetts Institute of Technology where he gained a Master of Science in Mechanical Engineering and delivered a graduate thesis on racecar data acquisition.

Jim is a triathelte, and a three-time Ironman finisher, including the Ironman® World Championship in Kailua-Kona, Hawaii in 2004. Since 2007, Jim has focused on road racing and endurance MTB. He won the Solo 30 Men’s category at Breck Epic 2012 and finished second in the Open Men’s category at La Ruta de los Conquistadores. 

To say that Jim is exceptional is an understatement.

 Why did you decide to start a business in the Black Hills of South Dakota?

I grew up here.   My father was a dentist.  He and my mother drove into Spearfish on a sunny afternoon in June 1972 and then they never left and got carried over to me.

But mainly, I like the small town peace and not waiting in traffic.

We’ve got a lot of beautiful country around here.  The other resources necessary for our business are either present or easy to get.

 Can you tell me a little bit about what you make?

We build a power meter for a bicycle which is an instrumented bicycle crank set that measures the force that you’re putting on the pedals as well as the speed that the pedals are turning and that’s usual for training for neither professional cyclists but also for cycling enthusiasts… a measure of the training that quantifies your performance.

 Have they been around awhile?

The first ones came out about 20 years ago, but then in the last 10 and particularly 5 years, they have become much more popular.  25 years ago, they were basically kind of a sports science tool.

What gave you the idea to start with this?

I actually went to buy a power meter in 2006.  At the time, I was training for a Triathlon and I looked at the options that were available and realized that there was some market opportunities that were still available. That’s when I decided that I should try to fill that and build my own.

 Was there something that you saw in the existing systems that you didn’t like?

There were several systems on the market but the 2 primary ones were SRM from Germany (another crank set system, similar to ours) and Powertap from Wisconsin and the Powertap system was in the rear wheel and that was unavailable to me because I had a set of carbon tri-spoke rear wheels and the Powertap system wasn’t compatible.  So that meant that I needed an SRM but the SRMs were very expensive.   So that led me to think that there was a hole in the market for somebody to build a crank set system at Powertap level pricing. That’s kind of the basic, original market hole that we went out to fill.

How long did it take you to develop it?

It took 3 months to build the first—to build and test the first prototype. In 3 or 4 months, it went really quite quickly from a prototype to something real.

It’s pretty amazing what you can do with a laptop and an internet connection.  You can really learn and design a lot and get things made and done.  From the time we tested that first prototype until shipping was 2 years

I think that was big lesson, because it’s a much smaller project to make one thing work once in one environment.  It’s  a whole other project to make all of them work all the same in all sorts of environment and that kind of shows the differences that 3 months to make a prototype versus 2 years to make a product.

The 2 year development was a lot about the actual—the design of the product itself and distributing out any production process that would be able to make it work.

 How did you figure out manufacturing?

A combination of everything.

We originally have some partners that helped us with things which are great for some things because they bring in resources but then you know if you have outside partners there’s always knowledge that is outside of building them and so after we’ve gone, we brought more and more in-house and tried to capture and understand and kind of drive that knowledge internally within our organization.

As we’ve grown, we’ve hiring outside people, with new skill sets.

 That’s an interesting pattern you’re bringing up.  There seems to be two patterns for small manufacturers- the “I’m outsourcing everything”  versus the  “I want to try it internally.”  How did you make the decision?

We just started with all the outside people and then the supply chain and the lead times involved get pretty long.   We ran into problems where somebody is having a problem at one vendor that they don’t necessarily know how to solve and the solution is actually out of a different vendor and when those things are in different places with people that don’t talk to each other and they only coordinate through us.

It’s hard to find those solutions and so just from by necessity and practicality, we’ve kind of naturally stumbled into bringing more and more things in house, mainly because then you can chop out the waste and lead time between the steps, but then you also you get the learning where you watch what you’re doing in one step and see how that affects the next step; and of course that’s once you’ve developed the process and make it more efficient.

Where there times where you needed to redesign part of all or part of your product?

Yeah, there’s been a couple of times when we had to stop and regroup, I think that one of the reasons why we’re successful was the first version that we made was pretty good and so the first units that we shipped in 2008.  We made one design change, but beyond that we went all the way around until 2012 before we introduced a full new ground from the ground up product.

That was pretty good result to get 3-4 years life cycle out of that first generation design before we switched to that second generation.

That is really exceptional.  That’s hard to do and it says that you got your first one pretty right when you got it out the door.


Important topic for entrepreneurs– How did you find the capital to get started?

I saw a presentation once on “How to be an Olympian?” and it had 5 things you needed to do and I don’t remember what they were—work hard blah, blah, blah, but the fifth one was “Choose your parents right.”– And this is more important than other other factors put together.

My dad was a dentist and he started the business when I was young and he did well enough on that to sell it; so then when I wanted to start this, both my parents knew what it was like to start a business from scratch and they decided to help.

So you used the “family and friends” method which is the most common way I think people start businesses.

My dad didn’t have that advantage and so he went out and sold stocks to people in town, and these friends became involved in his first business.

And that’s what we would have done had the resources from them not been available.

Had my parents had not been available, we would have gone out and found some share holders.

 If you could go back in time and talk to yourself at the beginning of this adventure, what would you tell yourself ?

There are a couple of things.  One, we messed around with some vendors that weren’t entirely helpful to us.  The more we brought in-house, the better off we were.

I summarize this as there were times when we had low leverage with our vendors and I’d really avoid that again. If you are using outside vendors,  you want to be able to ask direct questions and get direct answers. If the vendor for whatever reason can’t or won’t be forthcoming about what’s going on in the process and how things work and if they’re not a partner in getting things solved then you kinda need to move on elsewhere, either bringing in-house or finding a new vendor; because if you got a vendor helping they really need to be a partner particularly really on– in a business.

 How do you use technology?

We do a lot of CAD work for design and then we do a lot of automated data collection for testing and QA/QC.  I think that we’re pretty data driven and that’s something that we want to do even more of as we develop the manufacturing further.

I’d like to get to be more structured in the way we do process control and really use a lot of those formal process controls and SPC tools to drive what we’re doing but from day 1 we’ve done a lot of automated data collection through all our testing and very early on, we built a data acquisition system—a database that could show test results.

 So, over time do you see more of these tools coming in or you’re looking to automate more, what’s the right balance for you?

Well, we like to have data to make data driven decision but certainly it’s not really value added to be entering data into the ERPs to spend time entering data into the ERPs system so that time that you’re just purely on data entry is kinda waste.

So we do a lot to try to make ERP provide data input automatically within the process.

There are places where we take don’t automate, such as when we get the process lean enough that things are moving through the process very quickly.  We have a physical model of being able to see what’s going on in the production flow either with Kanban or a fast enough production flow.

But of course, you can’t run the entire business that way.  To me, the ideal use for the technology is to free people up to add value, but all the data and the reporting is available for analysis because the data was collected intrinsically within the process.

How do you innovate?  When do you get that next great idea for a product or improvement and how do you make it real?

To me, the innovation piece really comes when you got a problem, and embracing the constraints when you’re stuck.

When you “can’t do this because of that  and you can’t do that because of this” and once you start to kind of have those boundaries or you start to get boxed in,  the problem drives the creative moment.

Once you get all those constraints, and think about what you’re really trying to achieve, generally, there’s some sort of base assumption that you realize is wrong.  To me, that’s when the innovation happens, when the constraints are present.

 So, are you a fan of Elihayu Goldratt?

James:   Yeah, yeah I haven’t read “The Goal” in a long time but certainly—yeah.

When I hear someone who knows about the Theory of Constraints a little bit I—that perks up my ears.

Yeah, and of course you could think of it from a purely manufacturing stand point.  But even Finance, when you get boxed in financially, you have to think “Okay, what are we really trying to achieve” and that’s when it really make you focus and try to maximize the results.

 How do you connect with your customers and fans and how did you go about building this community of people interested in what you do?

We participated in forums directly from the beginning and we tried to talk to the customers as directly as we could.  We had a big advantage on that and that our product was to generate the following and there’s a forum  on Google of a  bunch of bike nerds talking about power meters there everyday.

So we interact directly there and with basic posts and not trying to be really promotional but just having a conversation here of building a power meter and this and that and as we grew we found out that we had to interact less, mainly because we get our own fan base developing and they lay down their questions for us.

 One of the things that I noticed from your website, is you are celebrating the victories and the whole career of your customers on your blog.  You go out of your way to acknowledge the people out there who use your product.

Yup. That’s the important piece. We’ll hop in and say a word here and there and show that we’re watching, show that we’re listening, and that we care.

What’s the best advice that you have ever gotten?

I was getting  coaching and mentoring from a trusted advisor.   He encouraged me to raise our prices early on.  We had more demand than what we have in supply and we were trying to keep the prices low to compete with Powertap price point.

We raised the prices and that turned out to be the best thing that we ever did.

Our advisor knew you need to understand that the price somebody will pay as related to the value that they perceive from the product and it’s fully unrelated to how much it cost to build the product.

The learning in there is to really make sure that you understand the value proposition that you’re creating and what’s that worth to people.

Let’s say you’re cost is $50, and you’re selling it for $100,  if you can sell it for $120, you just made a lot more money on every sale and that’s the key thing to understand.

 Where will your market and product and business be in 5 years?

We’d really like to grow into other products within the bicycle industry relating to data acquisition– other sensors or recording and display units or integrating into other components of the bicycle, that’s where we’re looking.  Certainly Power is the big key product but then there’s more places to go.

Posted in Uncategorized | Leave a comment

Jonathan Werner– :” …Like it or not, globalization is here.”

jw at desk

Jonathan Werner, Director of Procurement for Custom Traxx

Custom Traxx is a leader in the model train hobbyist community, creating an exceptional product for a high end market.  You can find out more at :



Disclosure—I have known JW for 30 years, since we were kids together in Bismarck, ND.  He and I got into mischief together, and I hope we are in the clear now, due to the statute of limitations.  Jon  is an engineer, entrepreneur, and former Marine.  He is also a sports raconteur, and a great friend.  

Tell us a little bit about what you make.

We are really a niche within a niche. Take the niche market of model railroaders, then take that down to the sub-set of enthusiasts who only model the streetcars which used to be a part of everyday life in this country.

Pennsylvania 2009 118

Custom Traxx supports the HO scale electric traction modeler by supplying authentic detailed decal finishing sets, heralds, destination signs, and other markings for streetcars, and interurban railway vehicles as they ran in most American and Canadian localities. We also offer a growing line of parts including selected traction re-powering items; HO scale traction mechanisms, trolley poles, and HO scale ORR girder rail, turnouts and crossings, as well as a number of specialty parts from our strategic partner Bowser as well as several other manufacturers in the streetcar modeling industry.

Currently, our flagship product is the line of HO scale models we have produced with Bowser of the historic traction vehicles operating on San Francisco’s F Market Line. We just gave our first public glimpses of the newest model in this line, the Perley-Thomas 900 class car which originally ran In New Orleans, and are running today in both New Orleans and San Francisco. We are scheduled to have this model available by the end of the year in not only in its original New Orleans paint scheme, but also in the marking of a few other cities where these cars were in service, such as Atlantic City, Chicago, and Philadelphia.

A more detailed overview overview of Custom Traxx can be seen in a public album on our Facebook page.

When did you start making this?

Pennsylvania 2009 040

A PCC body during the F Market Line project testing phase , May 2009

This has been approximately a five-year process. Custom Traxx has actually been in existence since 1992, but the F Market Line was our first “full-blown” venture into mass production.

Pennsylvania 2009 011

One of the first F Market testing prototypes, May 2009

How long did it take you to develop it?

The F Market Line went into the “drawing board” phase in 2008. We introduced our working prototypes at the East Penn Traction Club’s bi-annual meet in Philadelphia in May of 2009, and the first models in that line were on store shelves by the end of that year. Since them, we have introduced new models in groups of four or five roughly twice a year.

What changed along the way? 

Our entire concept of how to manufacture these things. Before the F Market Line, traction modeling was a world of scratch building, modifying existing kits, and generally scrounging for parts and taking a “make-do” approach because it is truly a niche hobby.

Then along came a technique known as resin casting. Resin had some serious upsides, namely it was economical and patterns and molds could be made inexpensively, which allowed for small and/or custom production runs. That fact in and of itself was a massive factor because most modelers want a specific model from a specific time, and from a specific city.

Resin also proved troublesome from a few perspectives. First of all, certain resins have  durability issues, such as warping and cracking. Mass production was a problem because the molds generally didn’t have a long enough lifespan. But most importantly, working with resin could be dangerous, particularly inhaling resin dust from sanding the bodies before painting.

In other words, resin casting put the idea in our heads about taking the hobby out of the scratch-building world, but resin casting also proved it would not be the vehicle to get us there.

Did you run into times that you needed to redesign part or all of your product?

Pennsylvania 2009 016

Custom Traxx CEO George Huckaby with Bowser CEO Lee English during the F Market project’s testing phase, May 2009.

Constantly. There was the aforementioned issue with how we would even manufacture these models. Then, there were constant issues with fitting all the needed component into a cramped physical space. For a host of reasons, weight of the models proved to be a crucial issue. Then came the challenges of putting digital command and control (DCC) systems into these models. They were becoming all the rage in other model railroading circles, and it was clear to us that failing to incorporate this technology doomed us to obsolescence. However, that also meant fitting a circuit board with wiring into an already cramped space.

How did you figure out your manufacturing process to make this?  Have you changed it since you first laid it out?  Did you consider outsourcing or off shoring?

30-Group Shot at Sanda Kan

George Huckaby and Lee English with some other project members at the manufacturing facility in Shenzen, China

Let me take that question by question. We came to our manufacturing decision essentially through reverse engineering. When the resin casting option proved problematic, we asked the question “How else can we build what we want, and in the quantities we want?”

It is important to note we asked this question at a point where we hadn’t yet decided to make this a business; we were still just a bunch of hobbyists looking to build something we wanted. Custom Traxx was still a “garage operator,” providing decals and other peripherals. We’d never done a “full-blown” effort into models. That means we never really laid out a vision of a manufacturing process as much as we laid out requirements. As we continued through that process, going off-shore proved to be the only real option, for several reasons.

Many people think the option to go overseas is simply financial, but there’s more to it than that. Granted, in our case the cost-per-unit to manufacture simply could not be ignored. Even after you factor in time, transportation, and the other logistical issues involved in having your production floor 10,000 miles and twelve time zones away from your management capacities, the decision to go off-shore was an easy one.

But another reason that made that decision an easy one was all the regulatory stuff the multiple levels of government will put you through in this country. Without getting into specifics, when you factor in the time and money involved in navigating that sea of red tape if you wanted to manufacture something in this country, you can have something built in China, let it spend two months in transit, and still have it on the shelves in America before you can get through the governmental hoops.

In my opinion, there’s something fundamentally wrong with that, and if we as a nation want to remain economically relevant 50 or 100 years from now, we need to stop the finger-pointing over “they are sending our jobs overseas” and start asking why that is happening. That includes asking the hard questions nobody wants to deal with, such as if we waved a magic wand and created 5 million new manufacturing jobs in the country, would our infrastructure even be able to support that?

The answer is probably not, because we haven’t made any significant upgrades to our infrastructure in 50 years. That’s in direct contrast to countries like China, Vietnam, and several of the former Soviet Bloc nations who are all increasing their manufacturing capacity.

The fact is that like it or not, globalization is here.  It has been here since the days of the Silk Road; it’s the whole reason man learned to build ships to cross the oceans, and the technology available to us today is only furthering that.  The people who will tell you differently tend to be good, old-fashioned isolationists, and any economic historian who is being intellectually honest will tell you there was a precise term for the last time America deliberately choked off foreign trade. It was called the ”Great Depression.”

Twenty years ago, Mexico was the “stealing our jobs” boogie-man. Now it’s China. In the future, it will be Vietnam or one of those former Soviet bloc nations I mentioned. The other day, I bought a shirt that was made in Lesotho.

Like I said, globalization is here. It’s time to address that as fact, then plan and act accordingly.

How did you find capital to get started?

We didn’t.  After the financial panic of 2008, we knew we had no chance of getting a loan. This all came out-of-pocket and leveraged with partnerships. In many ways, all we really did was take technology that either existed or was emerging, and brokered it into packages that fit our needs. As long as we kept making  those so-called “win-win” deals, the momentum kept growing.

But to be honest, this isn’t a “big-cap” venture. Again, it really started as a couple of hobbyists who weren’t so much about the proverbial “better mousetrap;” rather we just were looking for a way to build the mousetraps we wanted.

If you could go back in time, and talk to yourself at the beginning of this adventure, what would you advise yourself to do?   You must have learned a lot since then, so what lessons would you say stick with you?

My answer for that is more of an observation about advice, and from whom you get it.

When you take advice, you really must understand where the advisor is coming from. For example, the words from the guy who tells you not to worry about risk when you discover he’s floating on some pain-free venture capital…well, those words ring a bit hollow. The same goes for the guy who tells you social media isn’t important, then you find out he’s running his whole business off an old-school land-line answering machine in his basement. My point is sometimes we tend to listen to advice that fits our vision; the old “people hear what they want to thing.”  When words fit into our ears just right, sometimes we don’t do the due diligence. And that’s to our peril.

What technology do you use for design?

16-Reviewing the status of the project

George Huckaby reviewing specifications with a project engineer in Shenzen, China

We are heavy users of CAD. Our partners at Bowser use a lot of CNC. I can see a need for ERP in the future, especially as the partnership continues to develop. CAD is mission-critical because one thing we learned about manufacturing off-shore is that our procurement packages needed to be flawless. When your production floor is 10,000 miles away, there’s no such thing as “fixing in production.” Once you flip the switch, the process doesn’t stop until you have a finished product. If it wrong, you just have to eat it and start over. For us, or for anybody like us, that would have been a fatal blow.

We have experimented with several CAD applications, but we really aren’t married to any one in particular. That may sound a bit strange, but we have a unique problem regarding complex curves. Without getting to detailed, we are in a world which demands a high level of accuracy in a model as compared to the prototype. Compound that with the fact we are modeling things which were designed anywhere between 60 and 100 years ago, with design techniques ranging from clay sculptures to pencil drawings…to make a long story short, there are challenges in replicating curves with a computer that weren’t necessarily designed with a mathematically-based approach. So, we are always looking out for whoever might build the proverbial “better mousetrap.”

Having said that, personally, I prefer programs that support *.dwg files. They are the closest thing the CAD world has to universal, and when you want to keep the ability to move around, that’s important.

It’s also a crucial component because as I’ve said before, when dealing with manufacturers who are on a different continent and have a different culture, precision in your procurement packages is essential.

How long does it take you to get a new product designed and out to the market?

The timeline in our product development has some fixed factors and some variable ones as well. The actual production itself, meaning the time from the molds are cast to the time the model bodies are complete is essentially a function of time per unit multiplied by number of units being made.

If we are talking about measuring from conceptual phase to putting boxes on store shelves, I can describe the time line like this:

Concept: “Let’s build car X. There’s one at Y museum we can use as a prototype.” Assuming we believe there is a viable market for car X, that’s the extent of this step.

Design: Variables based on phases…The complexity of the body type to be modeled and the number of paint schemes in which the model will be produced.

Production: As I said, time per unit multiplied by number of units.

Foreign Transport: Expect 30 days by sea from Asia

Customs: One of the biggest variables in the process. Dealing with Customs is like dealing with a building inspector…you never really know what you are going to get. Expect five days

Domestic Transport: Everything goes to our partner’s facility in Pennsylvania, and distribution to retailers happens from there, as well as direct on-line sales.

The bottom line. From start to finish, roughly six months.

What is the worst advice that somebody gave you as you were building your operation?

George Huckaby (Custom Traxx founder and CEO, and JW’s father) may have a different answer to this question, but to me, the worst advice I heard came from the nay-sayers to manufacturing off-shore. These were the people who told me nobody would buy models made overseas; these were the same people who lined up to buy them.

How do you innovate?   When you get the next great idea for a product, or improving an existing product, where do the ideas come from, and how do you make it real?

Pennsylvania 2009 435

George Huckaby pointing our details on a SEPTA PCC-II in Philadelphia

The re-emerging interest in public transit in this country is doing that for us. We started this with the PCC cars running on the San Francisco’s F Market Line. Since then, several other cities have gone on to bring back vintage rail vehicles. Just a few examples; San Diego has some PCC cars running on a line its Gaslamp district. Toronto is running some restored PCCs. Kenosha, Wisconsin has a heritage railway outfitted with PCC cars. And these aren’t museum pieces, these are fully-restored vehicles operating in revenue service.

We have a firm belief in only modeling things which are running on real rails somewhere, because people model what they see.  As more cities bring back these vehicles, it simply becomes a process of determining which ones have enough market appeal to warrant putting them into production.

In terms of technology, the big thing we’ve done is to be on the edge of bringing DCC into the traction modeling world. This is the technology that allowed model railroaders to have models with realistic sights, sounds, and lighting, not to mention the ability to control multiple trains on the same layout from a single controller, or to stream video from a train to your smart-phone, or to even use to your smart-phone as a control device.

Again this was the results of our building relationships with companies already in the field like Soundtraxx and Train Control Systems. The Tsunami©  decoder from Soundtraxx, and Train Control System’s advances like WowSound©  and KeepAlive© allowed us to develop model functions specific to traction vehicles.

How do you connect with your customers and fans?  How did you build a community of people interested in your product and ideas?

glh driving 717 at oerm

George Huckaby operating a vintage Pacific Electric vehicle at the Orange Empire Railway Museum, Perris California

Actually, this is part B to the answer to your last question. Before all these cities began running these vintage vehicles, to see them you had to go to a museum. To this day, the various railways museums around the country are the best places to do research for new traction models.

Really, this is another area where our emphasis on relationship-building pays off.

When many of these cities started putting these vehicles back on the streets, they didn’t have anybody who knew anything about maintenance and operation of them. They literally had no choice but to go to museums.

George Huckaby has been a member of the Orange Empire Railway museum in Perris, California for forty years. Not long after he started taking me out there, I was in the booth selling tickets . Now, forty years later, George is the CEO of the museum, and we are both using our experiences at OERM in many ways today.

That is really why we believe so strongly that the relationships between modelers, museums, and present-day transit systems can be very beneficial to all parties involved.


Jonathan Werner at the controls of a PCC car at Orange Empire Railway Museum, Perris, California

For example, the car shown above, Los Angeles Metropolitan Transit Authority (formerly Los Angeles Railway) Car #3165 was originally built for Twin Cities Rapid Transit in 1948 before being sold to LAMTA in 1953.  Since we had intimate knowledge of its history, we knew that if we were to model this car, we would have to account for some of the unique details specific to cars built to operate in the harsh winters of Minnesota; features that may not necessarily be present in the car’s current Southern California museum life.

The knowledge gained from those relationships places us in a unique position to work with many modelers, other manufacturers, and even municipal transit operations to provide technical assistance, specification evaluation and verification, functional testing and design revision, and even maintenance and operational advice for the transit operators. Let me give you some more examples.

Massachusetts Bay Transit Authority (Boston) contacted OERM with questions about the braking system on their fleet of pre-World War II fleet air-electric PCC cars. OERM has been running just such a car for years;  Los Angeles Railway #3001, a pre-war “Air-Electric” PCC built in 1937. The knowledge we had of that particular car also allowed us to offer paint and specification to another manufacturer for their production of a model of that car.

Through the relationship Custom Traxx established with San Francisco MUNI during the development of the F Market PCC models, we gained the ability to gather specification data on other MUNI vehicles, such as MUNI’s Breda Light Rail Vehicle. We then shared this data with another model manufacturer and offered technical assistance on their Breda LRV model project.

Another benefit of our relationship with MUNI was that we were granted permission to examine some pieces held by the Western Railway Museum, largely due to that organization’s relationship with MUNI. This had a major impact on our plans to model the San Francisco “Torpedo” double-end PCCs.

So, how does this all answer your question? The traction modeling world is genuinely a small one; the more places where we can make our mark can only mean getting visibility to our efforts where we may have not had it before. Many transit workers become museum volunteers, and many museum volunteers are modelers. On top of that, I mentioned earlier we only model vehicles which are actually running somewhere. By being able to say we are part of the resurgence of the use of vintage rail vehicles in actual revenue service on several transit systems, when people want to model those systems, they will already have the Custom Traxx name in their heads.

It seems like one of your competitive advantages is your R&D.  You are intimately aware of your target product’s life history, and technology that is demanded by your customers.

The beautiful thing about model railroad enthusiasts is that they are not shy in sharing their opinions. As I’ve said, these are people who demand a very high level of accuracy in a model.  So, when you unveil a prototype, you will get plenty of input. Again, this is where our relationships with various museums and transit agencies are in valuable. It gives us the ability to get almost any type of information we need for either no or very little cost. That approach also plays big when it comes to developing new technology. The partnerships we have formed with other manufacturers in the industry mean we can actively share information and expertise to develop new products and/or technology.

How did you figure out pricing for your products?  This is a key issue for a lot of new manufacturers?

That’s a really good question, because pricing was really a key part of making the decision to build these ready-to-run models. Let me explain that. Ready-to-run models (meaning those that you could pull out of the box, drop on your track and away they go) were a rarity in the traction modeling world before Custom Traxx. Everything was kits…kits you assembled, painted, and put decals on yourself. With some manufacturer’s kits, you just got the body and the floor; the power and the decals were up to you to find. That meant you could pay $150 for a kit, and another $75 to power it and another $25 to put decals on it, if you could find them.

We figured out if we took a mass-production approach and integrated all the stuff you would need to buy to complete a kit into one ready-to-run model, we could offer a basic unit originally priced around $150. Naturally, technology advances such as DCC and sound were priced above that, but we knew for a basic painted and running model we had to stay below $200.

Where will your market, product, or business be in 5 years?

I’m not a big “crystal ball” guy, but I will say this. The opportunities to model vehicles which are running on the street are limited. That means we are headed for a change no matter what. Personally, I think it is probable we will be making models of the current light rail vehicles in the future. It is definite we will have a new market in five years. If you go to a model railroad show now, especially one geared toward traction modeling, you are going to see a disproportionately high number of gray-haired people. Our ability to remain viable is directly tied to our ability to get younger enthusiasts engaged in the traction modeling hobby.

This is why OERM hosts a number of “Thomas the Tank Engine” functions per year. All we have to do is get the kids onto the museum property and let them see out fleet of traction vehicles in operation.  Since I am based in Lafayette, Indiana, this is why I have regular interaction with a large and established model railroad club at Purdue University.

Regardless, five years from now we will be marketing a different product to a different market.

How have you dealt with “Near Death Experiences?”  Any advice to a new entrepreneur, who is scared out of his/her wits?

Well, just about anything can be a “Near Death Experience” in a start-up if you don’t make the right decisions at the right time. In our case, we really were fortunate in several senses. First, we knew a lot of the pitfalls from doing business in this market from having talked to several other entrepreneurs. Second, we knew where the fatal mistakes were in the process because we were able to draw on the experiences of others. Most importantly, we knew that over-reaching was a death sentence. This isn’t a market where you can generate a wide range of revenue streams. We knew we had to be deliberate in our actions, very much more the “tortoise” than the “hare.” I get that approach may not work in other areas, but I can’t see any arena in which interacting with other business owners and having as much useful knowledge as possible can’t improve your decision-making.

Posted in Uncategorized | Leave a comment

“If you’re waiting for the answer to become black and white, too late.”– Chris Anderson, CEO of 3D Robotics

Chris Anderson is  the CEO of 3D Robotics and founder of DIY Drones. Prior to this, Chris was the Editor in Chief of Wired Magazine.  Chris is also the author of a number of books, including “Makers”.

In full disclosure, 3D Robotics is also a customer of Ursa, and uses OpenERP for their production management.

Greg Mader:  So, how do you tell people what you make?  How do you describe what you do to a family member?

Chris Anderson:  We make drones.

Greg Mader:  How do you feel about the “drone” word?  That’s controversial within the community isn’t it?

Chris Anderson:  I created a site called DIY Drones, so obviously I’m okay with it.  Like many similar words, it starts with military connotations and turns out to be tamed for civilian uses.  Remember computers used to be scary, the internet used to be ARPA-Net.  There is a long tradition of taking military technology putting it the hands of civilians, and reclaiming the words.

Greg Mader:  You mention how you got started building drones in your book “Makers”.  Can you tell us a little bit more about how you got started?

Chris Anderson:  I started with my children, building Lego Mindstorms  robotics kits  and I was blown away by what you can do with the parts  –  my children were getting bored and I was trying to get something that would impress them, that they hadn’t seen before, like a robot.

I thought “what would I like?”  So, we built a Lego Autopilot, it kind of almost worked.  It did not impress the children, but it impressed the hell out of me.  And the rest is history.

Greg Mader:  How long did it take you to develop  your first autopilot and — and were bumps in the road, or places where you thought, “You know, I need to try different approach.”?

Chris Anderson:  I think that every day, you have one of those moments. I had a lot to learn.

It turns out that it was not a very good autopilot, But, it was going from kind of a bad autopilot out of Legos, to platforms we could use, and along the way, learned about embedded computing and physical computing platforms, ending up with what we know now.

Along the way, we learned about embedded computing, and physical computing platforms, ARM processors, sensors, and GPS.

I think every day, it was like, “Oh, yeah it turns out it’s a lot more complicated, that one particular thing is a lot more complicated, I think that’s where I need to switch gears.”

But, you know, the good news is that, it was really easy to switch gears.

The tools were getting better.

When we started, accelerometers were cheap, but gyros were expensive.  A couple years later, accelerometers and gyros were both cheap together on the same chip, and the GPS got better and cheaper, processors got faster, and, hey, wait 6 months and this industry will churn out another version, and then you have the next miracle, and, that problem would get easier.

The stuff I was doing was impossible ten years earlier, incredibly difficult five years earlier, by the time I got there it would be easy, and getting easier by the day.

So although we made lots and lots of mistakes, and had to change course all the time, we were able to do it. There were no brick walls.

Greg Mader:  How did you figure out your manufacturing process?

Chris Anderson:  I’m the wrong person to ask, I’m afraid.  Jordi ( Jordi Munoz) really did all of it.  I’m up in Berkeley.  The reason I quit my job, and came on as CEO is because Jordi kept sending me updates, “Here, we moved into a new space,” or “Look at the new machine.”

Every update was more jaw dropping than the last.  Like, “Here’s our book keeper,”. Okay, we have a bookkeeper, and then he telling me, “Here’s our inventory system.”  Okay, we have an inventory system,  and then “here’s our pick and place machine,” and I ask, “What’s a pick and place machine?”  He’s tells me, “I bought it on eBay.”

I think it was the day that he showed me a picture of the forklift that we had, so I thought, “Okay, this is a real company, it’s time to get serious here.  I’ve got to quit my job.”

Greg Mader:   [laughter],

Chris Anderson:  I guess, Jordi was 21 at that time, with a team of other 21 year olds. Maybe it’s just something that 21 year olds just can do.  They just have to go on Google, and they ask questions, and they dig around, and they just do stuff.  But they got themselves ISO 9001 quality engineering and production, by studying on the Internet.

I stand in awe, because, the average age at the company is less than half my age.

I knew nothing about Mexico, until I met these guys, randomly met Jordi on the internet, and I thought Tijuana was full of drug cartels.

What they’ve opened my eyes to is that this is the future of North American manufacturing with electronics.

That this is where these kind of world class manufacturing skills now exist.

We’ve lost so many people in US manufacturing, but Mexico has only grown them.

What we want to hire somebody that has been working for a world class multinational for 10 years like Samsung or Foxconn?  The answer is we’re not going to do that in the United States, we’re going to do it in Tijuana.

So, we have, we’ve learned from our employees, we’ve hired people on the production line who have spent the last 5 years at Samsung.  And we ask, “how did you do at Samsung,” and they’re like, “this is how: this goes here, this goes here, this goes here.”

We hire employees, and they keep up.  [laughter]

Greg Mader: How did you find the capital to get started?  That’s a question, a lot of people have?  “I have this idea and I want to make it real but I don’t know how to talk to people about money or what kind of risk I’m willing to take myself.”

Chris Anderson:  It didn’t take any money.

Over 3 years, we ended up putting like $50,000.  It’s basically credit card.

We started small, we started organic, and we had cash flow on day one.  I think we’ve been basically profitable every month of our existence because it’s so easy with hardware.

We have a model to charge 2.6x raw materials cost.  A little margin for us and a margin for the retailer, and we grew organically.

The simple answer is that we didn’t raise venture capital until we were three years old at which point – I’m going to round off here – but at that point, we were a business, we were already a large-ish, (well I guess small by global standards) but large for our industry.  We were a large, successful company, and we were able to raise venture capital with good valuation.

Had we tried to raise money first, before we had even been in business, we would not have been able to raise as much money or at nearly that valuation.

So the good thing about this kind of business is that it doesn’t take much money, and the money is at Kickstarter rates of money.  You can build a manufacturing business for $50,000.

Greg Mader:  That’s profound statement. Would it possible for a Maker to go cheaper, if he could send the design abroad and have at least to have an initial run?

Chris Anderson:  Oh, absolutely.  If you can outsource, you can do it even cheaper but you know the reason that you can build a factory so cheaply now.

Part of this trend are tools like OpenERP.  So much of the software is open-source for free.  The tools are getting faster and cheaper.

It’s just like everything else in technology, the cost of entry is falling so rapidly, as the internet and innovation model kicks in.

In my book, I talked about that poor story from Flash of Genius from the guy that tried to build an intermittent windshield wiper factory.

And I talked about how you would do it today and, you know, the answer is, you probably would have started a factory initially, you have the ability today, and when it was time to start a factory, it would just be so much cheaper.

The supply chain around them got that much better, the services on the web got that much better, the software that we use got that much better.   If it’s hard now, wait 6 months and it’ll be easy.

Greg Mader:  If you could go back in time five years ago and talk to yourself at the beginning of this what would you advise yourself to do?  Would there be anything you change?

Chris Anderson:  I think I probably would have gone even faster. At a certain point, I said, “You know what, I need to focus on one thing and quit my job and run this.” And, the moment I made that decision, it went really quickly, and I realized I probably should have made that decision earlier.

I had to think about 5 kids and a wife, and I had to convince her that it was a good bet as well.  If business was ready to be taken to the next level a year earlier, I wasn’t ready, so I blame myself.

Greg Mader: That’s fair.  I actually think that’s a problem with every entrepreneur I’ve ever met. It’s easy to talk other people, it’s often harder to believe in yourself and say “this is the moment we got the right solution. Let’s go for it.”

Chris Anderson:  Yeah and, you know, I also had to convince my wife.

Greg Mader:  In your book, you talk a lot about technology- your company is using ERP, you are using pick and place machines.  Do you see the balance using machinery versus people who do assembly?

Chris Anderson:  Yeah we talked about that every day.  For example, yesterday, we were trying to decide whether to invest in these optical quality checking machines versus people.  So, these machines cost like, you know, $60,000 and they optically inspect circuit boards, to see if there are any mistakes.

These machines, when they’re properly set up, they’re very high accuracy, and they have very high throughputs.  Their costs go down over time, they can be amortized.

Whereas the people, they’re lower up-front cost, lower accuracy, and are a continuing expense.

And so, we had to model the business out.  How many people would it take – here’s people doing the job, versus here’s machines doing the job.

Then, we asked  “What do Year 1 costs look like, what do Year 3 costs look like?”

In the end, we decided not to go with the optical quality machines,  we decided to go with people, because we were concerned that the machines would be hard to set up, and that, we would actually have to hire at least one person just to run the machine.

That was just yesterday, and we’ll do the same thing today.  And we also do an in-source/out-source calculation.  You know, injection molding, wire cutting, packing – we are constantly trying to decide if this is our core competence, the place where we can add value, or should we try the job with someone else around us.

For the injection molding, we considered buying our own machine, but then we realized that the capacity of the machine was so high that we would be able to satisfy our entire yearly demand in one week, so the machine would sit idle for 51 weeks of the year.  And that was an easy decision, but you know, we go through this every day.

Greg Mader:  To switch gear slightly Chris, what’s the worst advice someone gave you as you were building your business?

Chris Anderson:  Everyone, I mean, everyone told me that it couldn’t be done and that they were all these regulatory barriers and legal risks.  Everyone told me all of these all these reasons starting companies was hard and risky.

I kind of chose to ignore all of them.

Partly because I thought that they didn’t understand our model, and we had actually built a model where we could diminish a lot of that risk.

If you call up a lawyer, and you explain what you want to do, the lawyer will come up with these explanations to death.  He will tell you that there’s high risk here, or that they can’t vouch for this, or that the law is unclear on this and you should pay extra for this additional insurance, or whatever.

If you call up people who are focused on risk, they will tell you that what you’re doing is risky – and sometimes you just need to ignore them.  Or you need to take that advice and say, “I understand that there’s a probability involved here, I’m willing to take that risk, let’s go for it.”

Greg Mader:  I’ve certainly heard all the doom and gloom-sayers, too, over the years. The funny thing is I keep seeing businesses that choose to ignore it and innovate instead and they come up with something amazing or beautiful and — and they did it because they didn’t listen to anybody.

Chris Anderson:  Right, take Paypal.  If Paypal had gone to a lawyer –( I’m making this up, I have no if they did or not – but I bet you,) if they had gone to a lawyer and said “we want to start this, like, currency, this like, money trading but we don’t want to regulated like a bank, is that okay?”

I’m sure the lawyer would have said, “No, that’s not okay, you need to go through the whole banking process to do what you’re doing. “

But they didn’t, and they’re not regulated like a bank, and as a result they’re labeled to be innovative, banks are not.

And, at the end of the day, it was a gray area, and they chose to innovate in a gray area, and all great innovation happens in a gray area.

If you’re waiting for the answer to become black and white, too late.

Greg Mader:  What’s the relationship between your customers and fans and your innovation?

Chris Anderson: We are community driven company so, our customers are the innovators.

We have an open platform which doesn’t just work, but also makes it easy to modify.

Then we have a cultural and licensing process that encourages innovation to come back into the community for the benefit of all.

So that’s what, you know, I mean if there are – we’re the Android in our marketplace, and there are iPhones in our marketplace, and, you know, if you want something that’s closed and polished there’s certainly that in the marketplace.  If you want something that’s open and nearly as polished, maybe not quite as much because of the sake of the openness, then that’s what we offer.

We’re a platform, and the great thing about platforms is that they make innovation easy by sort of extracting all the hard stuff and recreating a layer where anybody can add something, add their own ideas, their needs, their passions or whatever it is.  There’s a really easy way for people to contribute.

Greg Mader:  One last question: Where do you see your product and business being in three to five years?

Chris Anderson:  I think the Android example is relevant.  Android is something that started, kind of Linux-y, which is to say, very open but hard to use, lack of polish.  Eventually gained the polish, but it retained the openness, and beat iOS, beat Apple, for the top spot.  That’s our inspiration, if you will.

We’re in that transition phase, we basically have a very functional product and it’s open but without polish, but as it will soon be, so we’re focused on adding that polish.  Polish meaning, easy to use, reliability, non-expert.

So that’s the big pivot we’re going through right now, to kind of get that sort of out-of-the-box experience.

In terms of our company size, we’re building an aerospace company.  Pretty different kind of aerospace company, we’re a bottoms-up aerospace company, we don’t sell to the government or the military.  We don’t carry people, so we don’t go through that regulatory process.

But we think this is the shape of the 21st century aerospace company.  We think a 21st century aerospace company looks like a Silicon Valley company, in the same way Tesla, a 21st century car company, looks like a Silicon Valley company.  You know, it’s a laptop with wheels, we’re a laptop with wings.

And we think that being able to take the innovation model that worked so well with computers and the internet will apply to aerospace.  Aerospace is an exciting opportunity, a disruptive opportunity and I think we’re even pushing through that.

Posted in Uncategorized | 3 Comments