Archive for March, 2012

Quotes For Entrepreneurs–March 2012

Written by Sean Murphy. Posted in Quotes, skmurphy

You can follow @skmurphy to get these hot off the mojo wire or wait until the end of the month when these quotes for entrepreneurs are collected on the blog. Enter your E-mail if you would like Feedburner to deliver new blog posts to your inbox.

+ + +

“Functionality is not the same as usefulness.”
Frank Hayes in “Conventional IT Wisdom

Originally mentioned in “A Good Idea is No Match for a Bad Habit

+ + +

“The revolution will not be televised—but it may be bootstrapped.”
Matt Wensing(@mattwensing)  “What They Can’t Tell You: Starting Up Outside a Hub

More context, concluding paragraphs of post:

As a founder born and raised in South Florida that’s had some measure of success (insofar as survival is a major component), this is my plan.  To tell people of the dangers in no uncertain terms, to take a chainsaw to their ideas, and to inspire them to find ways to collect money from customers first, so they can collect it from opportunistic investors second, and on their terms.  I won’t claim to be the first—there are bright and eager minds sprouting up left and right, and many more to be discovered.  But I will refuse to settle for spin.

The revolution will not be televised—but it may be bootstrapped.

+ + +

“…visual thinking is a mandatory literacy for innovation leaders of the future.”
Lisa Solomon (@lisakaysolomon) in “The Visual Thinking Revolution is Here

+ + +

“Searching for work is unpaid work”
Ville Hytonen

+ + +

“Success is largely about keeping your promises.”
Seth Godin in “Successful?

+ + +

“”Brevity is power.”
Josh Billings

+ + +

“”A codified, repeatable, reusable practice contradicts the nature of innovation, which requires difficult, uncomfortable work to challenge the status quo of an industry or, at the very least, an organization.”
Helen Walters in “Can Innovation Really Be Reduced to a Process?

h/t Chris Coldewey’s “Why Innovation Resists Codification

+ + +

“The business model canvas applications may be as much of a Procrustean bed for early stage startups as the 13 slide VC pitch deck.”
Sean Murphy

From an blog post I am still working on.

+ + +

“When action grows unprofitable, gather information; when information grows unprofitable, sleep.”
Ursula K. Le Guin “The Left Hand of Darkness”

h/t Esther Derby (@estherderby)

+ + +

“Talent acquisitions erode potential customers’ confidence, poisoning the well for future bootstrapped startups.”
Erik Dungan

opening quote for “Honor Customer Commitments to Avoid Poisoning the Well.

+ + +

“The true value of business software is not just in what it can do, but also in how quickly and easily it can be changed.”
Ed Weissman as (@edw529)

+ + +

“You aspire to great things?
Begin with little ones.”
St. Augustine

Used as a closing quote for “Small Wins Enable Larger Wins.”

+ + +

“The name ‘sprint’ may actually be harmful to software development, because the whole business is more like a marathon.”
Michal Paluchowski ?(@mpaluchowski)

+ + +

“People know what they want because they know what other people want.”
Theodor Adorno”

+ + +

“Sometimes you need to be on the dance floor dancing. Sometimes you need to be up in the balcony watching the dance.”
Will Kamishlian in “Planning in a Bootstrapped Startup

+ + +

“We assume that because we have the label we have the knowledge”
Jean Toomer

+ + +

“Make your next move from where you actually are.”
Frank Kuppner

+ + +

“Some must live with the consequences of what they didn’t dare do.”
Wieslaw Brudzinski

+ + +

“My two favorite institutions – universities and startups – have something in common, they both eschew management, to their detriment.”
Bob Metcalfe ?(@BobMetcalfe)

Which is especially odd because they both focus on fostering learning, and management would seem to be essential to the efficient allocation of resources. I think there is a fear that management is too associated with what are referred to as “delivery skills” in “The Innovator’s DNA” and that these come at the expense of “discovery skills” and creative thinking. I think this is a false choice.

+ + +

“The aggregation of individual data does not a commons make.”
Alexis Madrigal in “Paul Graham, The Commons, and How Google Stopped Being Google

Alexis Madrigal reaches a subtle and thought provoking conclusion that would seem to be at odds with a lot of “Big Data” evangelism but I think he is right.

+ + +

“Aim to encounter unknown difficulties that you may gain unexpected results.”
Jean Toomer

+ + +

“A horseless carriage was a common idea…ever since the steam engine was invented…”
Henry Ford in “My Life and Work

Full paragraph for more context, for me this makes it seem very unlikely that he ever said the “if I had asked my customers what they wanted they would have said faster horses.”

Even before that time I had the idea of making some kind of a light steam car that would take the place of horses–more especially, however, as a tractor to attend to the excessively hard labour of ploughing. It occurred to me, as I remember somewhat vaguely, that precisely the same idea might be applied to a carriage or a wagon on the road. A horseless carriage was a common idea. People had been talking about carriages without horses for many years back–in fact, ever since the steam engine was invented–but the idea of the carriage at first did not seem so practical to me as the idea of an engine to do the harder farm work, and of all the work on the farm ploughing was the hardest. Our roads were poor and we had not the habit of getting around. One of the most remarkable features of the automobile on the farm is the way that it has broadened the farmer’s life. We simply took for granted that unless the errand were urgent we would not go to town, and I think we rarely made more than a trip a week. In bad weather we did not go even that often.

See also “Interview Prospects to Find Unmet Needs, Persistent Problems, and Goals at Risk

+ + +

Startup Stages Overview Video

Written by Sean Murphy. Posted in 1 Idea Stage, 2 Open for Business Stage, 3 Early Customer Stage, 4 Finding your Niche, 5 Scaling Up Stage, skmurphy, Startup Stages

This is Sean Murphy for SKMurphy, Inc.  I want to talk to you about our startup stages model and understanding that risk reducing milestones that separate each stage.

We break the startup journey into five stages.  In each stage you will explore different options and converge on a key risk reducing milestone. Starting from idea or formation, moving through open for business, early customers, finding your niche,  and scaling up.

In the Idea and Formation stage you are searching for a customer need, a problem solution fit, and a team. You know you are done when you have a problem that energizes a team that is mutually accountable and jointly capable.

In the Open for Business stage you are searching for a business model and proof of value. This allows you to formalize the team commitment and make firm offers at a price.

In the Early Customer stage you are searching for proven value, and the proof is that customers actually pay you, not just tell you that they will pay, they actually pay. That’s only possible once you are set up and are able to transact business.

In the Finding Your Niche stage you now have a target customer type that you are going to select from customers that you have already done business with who reference each other’s buy decisions.  You need to learn the domain language of that specific customer set and learn how to find similar customers.

In the Scaling Up stage you are now looking for product scale. This requires that you have a repeatable scalable process, that you can identify additional niche markets , and identify additional opportunities. What was heroic has to become routine. Now you are adding employees who are specialists.

In each stage we see this same pattern of exploration and convergence. You are going to look at many options and then converge on key solutions.

Thank you, this has been Sean Murphy for SKMurphy, Inc. We help technology firms find early customers and early revenue.

Founder Story: Luc Burgun, EVE

Written by Sean Murphy. Posted in EDA, Founder Story

This originally appeared in my “Entrepreneurial Engineer” column in EETimes as “No longer a startup, EVE aims for top tier of EDA players” on Mar-29-2011. I have added some additional hyperlinks in this version.

Dr. Luc Burgun is co-founder and CEO of EVE. He has more than sixteen years of experience in EDA in both engineering and executive management positions. Prior to co-founding EVE, he was R&D Director for Meta Systems, a French company acquired by Mentor Graphics in 1996 that specialized in hardware emulation systems. Dr. Burgun holds a Ph.D. degree in Logic Synthesis from the University of Pierre and Marie Curie in Paris and has been granted six patents. The following interview took place over E-Mail; hyperlinks have been added to offer some additional context.

Structuring Business Relationships: Employees, Contractors, Alliances, Partners and Co-Owners

Written by Theresa Shafer. Posted in Events, skmurphy

Relationships can be very instrumental in the success of a consulting practice. When building a practice, consultants need to build business relationships with a variety of potential partners. Understanding the power of these relationships — and establishing trust and confidence — is one of the keys to growing your business effectively.

Sean Murphy is the April speaker at PATCA. Sean will share important tips and considerations for choosing partners to help grow your business. He will discuss a range of working relationships including employees, contractors, alliances, partners and co-owners.

April 12, 2012 6pm

Register at

This session will explore:

  • Differences between employees, contractors, alliances, partners and co-owners
  • Defining the relationship roles
  • What partners need from you
  • How to pitch to a partner
  • Tips for structuring a deal
About PATCA:

PATCA is a nonprofit professional association of independent consultants and principals who work in small consulting practices. Since its founding in 1975, PATCA members have been highly respected for their professionalism, integrity, objectivity and competence. They include experts in all aspects of business and technology and serve clients in many industries throughout the U.S. and the world. PATCA offers a free Post a Project referral service for clients searching for an expert consultant. It also provides business leads, project referrals, and consulting education to its consultant members.

CTO Mastermind Open House

Written by Theresa Shafer. Posted in Consulting Business, Silicon Valley

Saturday March 24, 9-11am

Ground Floor Silicon Valley, located at  2030 Duane Avenue, Santa Clara,

We are launching new Mastermind groups in response to several requests from entrepreneurs who wanted to form an advisory board of peers with a deeper understanding of each other’s businesses and shared accountability.

Come to the meeting and see if you feel comfortable with the other folks that we invite and we will work out times and locations. There will certainly be one group that meets on weekends, there may be others that meet on a workday.

The difference between these mastermind meetings and a Bootstrapper Breakfast meeting is that anyone is welcome to drop in to a breakfast, this will be the same group meeting and holding each other accountable for goals and commitments. Over time, because these entrepreneurs are more or less in the same stage of their business and meeting multiple times they will get to know each better than the average breakfast attendee.

There is no charge for this open house but if you decide to join a facilitated small group there is a small monthly subscription.  Want to be notified of future open houses join Bay Area Mastermind meetup.

Linkedin, Twitter, and GMail Need a Read Only Password

Written by Sean Murphy. Posted in Tools for Startups

I have been exploring the use of several analysis applications that could access my LinkedIn account, my twitter account,  and my GMAIL account. They want to help me leverage private information that requires my account password.

So far so good, except that LinkedIn, Twitter, and Gmail don’t appear to have any provision for just granting read only access.

Small Wins Enable Larger Wins

Written by Sean Murphy. Posted in skmurphy

“Product-market fit happens first in sub-segments. Understanding the core value proposition for hyper-sub-segmented markets is where you’ll find strong market signals.  You can’t go big without winning small first.”
Brant Cooper “Don’t Think Big. There, I Said It.

The principle of small wins is critical to winning the trust of enterprises prospects making them your customer and in penetrating new markets.  We blogged about our approach in “Crossing the Chasm: Look for a Niche in a Lot of Pain” with a reformulate representation of Geoffrey Moore’s “bowling alley”:

Bowling AlleyLook for niche markets who are in a lot of pain. If people are in enough pain they will change their behavior and risk adopting something new.  After entering niche markets, we can move technology up and out by using the ones in the most pain as reference case studies to the others. Also notice you start with the smallest niche market. This will allow you to make your early mistakes on a smaller market. It also buys you time and expertise to develop a whole product.

“You aspire to great things?
Begin with little ones.”
St. Augustine

Honor Customer Commitments To Avoid Poisoning the Well

Written by Sean Murphy. Posted in 4 Finding your Niche, skmurphy

In a small way, every talent acquisition poisons the well for future, bootstrapped startups. It erodes the confidence of users and potential customers. People put their company blog on Posterous, they add their business to GoWalla, they gave AdGrok a few hours of their time, etcetera, etcetera.

I’m not saying I would turn down the offer. But I fear the long-term effect of all these acqui-hires is my potential customers saying “No thanks. I doubt you geeks will be around in 18 months” when I market to them.”
Erik Dungan (HN: callmeed) on “Posterous Acquired by Twitter” HN Thread (bold added)

An acquisition hire means that the original service is shut down or turned into an open source project but no longer maintained by the founders.  In a thread on the Google acquisition of AppJet (EtherPad) on Hacker News  commenter pvg nicely summarized the moral hazards in the transaction:

Any early-adopter of a start-up product accepts the risk that the company behind the product might be unsuccessful. Customers like that bet, in part, on the notion that despite the statistically long odds, the company is making every conceivable effort to stay alive and succeed.

The ‘buyout-and-the-product-dies’-type exit introduces a sort of divided loyalty and misalignment between the goals of the founders and the goals of their customers – if you’re unsuccessful you might fail and we might suffer an abrupt service termination but at the same time, if you’re quite successful, we might also suffer an abrupt service termination.

My observation on HN at the time was based on the original announcement that the service would be shut down within 6 weeks.

If only they had interviewed at Google and joined the team they wanted to be on a year ago.

These transactions trade on the goodwill of early adopters. And they make it harder for other startups as potential early adopters start to assume that it’s better to wait for what Google releases instead of investing time in a product that will be scrapped either if the company fails or is successful and acquired.

[Given that many of the AppJet team came from Google] My point is that if they wanted to build their own company they should remain committed to the product they built, and find a better support model for current customers/users than shutting down without notice. I do not begrudge them making money at all.

But one reason that they have “millions of dollars worth of Google stock” is because they offered a service that people adopted and paid for. I think they have more of an obligation to customers and users than the initial announcement indicated and I worry that not taking better care of customers in the transition makes it hard for other startups.

Google ultimately relented and open sourced the code, with this announcement currently archived on )

Google recognizes the value of the EtherPad code base and has released the code as open source. For more information, see  EtherPad project on Google CodeEtherPad discussion group

This open source release has already led to many efforts to foster further development and provide EtherPad-like services. If you are looking for a service based on the EtherPad software, or want to run your own EtherPad server, see the following links (not affiliated with Google, use at your own risk):

This is not an “aint it awful” post.  It’s a suggestion to take care in the commitments that you make to customers, especially early customers who help you create your brand,  so that you honor only honor the letter of any agreement but also the spirit.  If the opportunity to solve customer problems does not get you up in the morning it’s unlikely that a “big bag of money” will let you move to a better lifetime.

More fundamentally, as I wrote in “Overnight Success

  1. If you define success as making a lot of money quickly you should go into sales and cut out the middleman.
  2. You can buy one lottery ticket and make a lot of money. You can buy many lottery tickets every day of your life and never recover the cost of your lottery tickets.
  3. Most of the time the opportunity for “overnight success” is sold by folks who are interested in making a profit on your dreams without actually fulfilling them.
  4. Of all the sources of funds for an early stage venture, revenue is the most compelling demonstration of traction. Too many entrepreneurs view fund raising as an accomplishment in and of itself.

The challenge with a startup–like many other things in life–is that you need to integrate many different inputs, your own hopes and fears among them, and negotiate a working consensus with your co-founders, suppliers, partners, and customers to be successful.

Updated Tue-Mar-13 to fix link to In a thread on the Google acquisition of AppJet (EtherPad)
h/t to Syed Naimath (@SNaimath)

Related Post Wed-Mar-14 Reginald Braithwaite‘s post “Dear Landlord” closing paragraphs:

If I move in, I’m committing my business to your place. I don’t want to read a six-page letter telling me what a great ride it’s been and how much fun you had building this place and how much you’re getting to sell out, and oh yes, the loading dock is open 24 hours to help me move my stuff the hell out before you bulldoze.

Architecture models often have these tiny plastic figurines that look like people walking around. If I’m supposed to move in before you sort out your “monetization strategy” and “exit plan,” This isn’t an office and I’m not a tenant. It’s a model of an office and you’re asking me to be the plastic figurine sitting at the foamcore desk.
Thanks, but no thanks. I’m done with that.

Self-Publishing Often Marks The First Generation of New Knowledge

Written by Sean Murphy. Posted in Events, skmurphy

The first time I heard the word samizdat was at an ACT UP meeting. The term was applied to the vast stacks of photocopies that every member picked up on the way into the Monday-night meeting: treatment guidelines, drug studies, bureaucratic analyses, action plans, contact lists, and announcements of events ranging from performances and gallery openings to house parties and memorial services. This collection was never referred to as anything other than “the table” (even though it usually spread over two or three), a twelve- or eighteen-foot-long banquet of paper down both sides of which several hundred gay men and lesbians, nearly indistinguishable in their Doc Martens and Levi’s and sloganed T-shirts, bent their spiky or shaved heads and served themselves and one another with the ordered geniality of an Amish wedding. I was an intellectually pretentious but under-educated twenty-two-year-old who didn’t want to admit he was unfamiliar with a term that had the clear, clannish peal of jargon, the ignorance of which marked him out as neophyte or, worse, interloper. In fact I heard the word as “same-as-that,” which led me to think of it as an assertion of status: though these stapled stacks of paper, most written by people with no political background or scientific or journalistic training, lacked the credentials and durability of proper books, they were nevertheless the real library of AIDS, and the bound books that trickled out of traditional publishing houses were the table’s supplements rather than the other way around.

Dale Peck “Same-As-That” [PDF][registration required] Harpers March 2012

I spent the morning at the Pycon 2012 poster session. There were projects or products related to weather forecasting, data driven journalism, drug discovery, non-SQL databases, robots, data driven activism, open source volunteer enlistment and management, gridbeam, etc.. Running in the same convention hall as a job fair for python programmers. This enabling technology and related products and open source projects are being applied to engineering and scientific problems in a variety of industries and disciplines: defense, oil and gas, animation and movie making, financial analysis, pharma, and engineering.

Coupled with my experiences at Strata and Big Data Camp that I detailed in “A Picture is Worth a Thousand CPU Hours” it’s clear to me that high performance computing–databases, parallel algorithms and infrastructure, and visualization all in service of substantially more complex analysis–is in a very productive ferment. I think this goes well beyond the Hadoop centric cloud computing models that gained early prominence that I have focused on for the last four years o so:

The closest I have come to identifying a second waypoint–after Hadoop–for navigating this rapidly evolving landscape is Evangelos Simoudis‘ “Insight as a Service” formulation:

Founder Story: Linc Jepson, 74ze

Written by Sean Murphy. Posted in Founder Story, skmurphy

This originally appeared in my “Entrepreneurial Engineer” column in EETimes as “Linc Jepson’s 74ze leverages Russian and American engineering talent to persevere” on Jan-18-2011. I have added some additional hyperlinks in this version.

Linc Jepson studied Electrical Engineering at Tufts and after he graduated with his BSEE he was drawn to Silicon Valley’s technology boom in 1997. He worked as an RTL design engineer at several startups including Auravision, Broadlogic, and Believe and in 2002 he travelled to Eastern Europe for two years where he worked for a startup microprocessor company). He returned to Silicon Valley in 2004 to co-found 74ze, a design services firm that leverages Eastern European engineering talent. I sat down with him after a recent Bootstrappers Breakfast to learn more about his entrepreneurial journey; what follows is an edited transcript of our conversation with hyperlinks added for context.

Q: You have a broad base of experience, what kind of work do you enjoy?

I enjoy the fast-pace of small companies, where there is not only the opportunity but the necessity to do all kinds of work. And that work has an immediate and significant impact. Early in my career I was doing clean-slate development, performance research, upgrading and debugging modules of long-gone developers, verification, 3rd-party IP integration, you name it. I didn’t realize it at the time but it was great training for consulting.

Q: Why did you pick Eastern Europe when you left the Valley?

During the 2001 downturn in Silicon Valley I decided to indulge my wanderlust a bit and to get a better feel for the foreign labor market. I had previously worked with a team of engineers in India. I had studied Russian for about four years in high school and college and had a seedling of an idea in mind about starting a services company that would have a portion of its team abroad.

In 2002 I moved over to Minsk and then Moscow for about two years. I brushed up on–OK I greatly improved on–my Russian and found a job doing RTL design in Zelenograd, Russia. Interestingly, the company had been founded by Chinese/Malay investors. I created a CompactFlash controller and a touchscreen interface. I was also making contacts and continuing to think about an outsourcing team.

I had confirmed that there is some exceptional talent there.

Q: How did you decide on outsource engineering as a business model?

We consider 74ze (pronounced 74 Zee) a services firm, first and foremost. We happen to use remote experts when it benefits a project. About a third of our projects never utilize our remote team. In school I studied International Relations as well, for a BA. Working with remote teams allows me to pursue another passion.

While I wasn’t keen on the mercenary perception that many folks have of contract engineers, I knew that there could be improvements made in the outsourcing model. Specifically, most engineers tend to shy-away from verbal or face-to-face communication. It seems easier to be non-confrontational and swap a few emails instead of picking the phone or Skype, but you can delude yourself into thinking that all will be okay. This can be particularly dangerous when you are an outsider of a company. These communication issues are often compounded by cultural issues. Merge a few factors such as a contractor being foreign and also more junior than the client’s counterpart, as is often the case with remote teams, and you often find someone who too often will dodge asking a short, potentially embarrassing question, and try to compensate by putting in more hours.

The fallout from labor globalization can be overcome with stronger communication and more global management experience. It’s not achievable in all stages of development, but someone who work effectively  not only with other engineers in the same building but also with engineers in Krakow or Kiev is growing in value.

Q: How did you get started?

In 2004 I moved back to the US. We incorporated 74ze in Delaware, and then in Russia a few years later, after we had traction. One of our first two jobs was a local one. We had 2-3 people, all US citizens, on site. We only did minor work abroad, such as some analog modeling. The other early job was for a company in Europe. The job was a referral from an existing relationship. That job was done completely remotely. We only met the technical people from the client at a trade show, a few months later, where they demonstrated with our product.

Q: If you are doing RTL design and verification what EDA tools do you use?

An immediate obstacle that we ran in to when we started was the high cost of EDA tools. Right off the bat, this started knocking us out of the running for various contracts. When I was living in Zelenograd, I had used Aldec’s HDL verification tools and had a very good experience with them. After not using them for a few years, because we were using customer tools in the US, one of the guys on our team proposed that we incorporate Aldec’s tools into our next bid.

Over the years, we’ve built up a relationship with Aldec and have grown to rely on their tools quite a bit. It’s unfortunate that we didn’t get much traction in conversations with the other EDA vendors, but I think that we have found a good solution with Aldec Riviera-PRO and a solid partner for moving forward.

Q: Can you give me a brief overview of where the company is today?

We’ve shifted more towards verification in the last few years, both in terms of the jobs we’ve taken and also in terms of a conscious decision to solidify our roots in that space. While most of our early jobs were in or involved some design, and we continue to work in this field, we see that the continuing abstraction of testing is creating a very large opportunity. We’ve been honing our SystemVerilog skills in the last few years and have been doing more verification than anything else, lately.

While we are still a mix of full-time and contract-based engineers, we are a pure engineering team with next to no marketing or sales overhead. This means less overhead cost for the client, but also that we don’t have the sheen of the larger companies. When we meet with a prospective client, they meet the lead engineer in the first meeting.

The economy interfered with our plan to cycle a few younger engineers into our team. Our team has a minimum of 12 years of experience, but I expect that we’ll have an internal project running as a testing ground for some junior engineers later this year.

Q: What are the two or three things that you have been able to accomplish that you take the most pride in or satisfaction from?

We increased our role significantly with two clients over a 2-3 year relationship. With one of those companies, we joined to develop a few blocks from spec and ultimately took on a much larger amount of design work as well as key roles in integration and top-level verification. We helped carry them through full chip functional sign-off and final timing closure. The chip worked the first time. The CTO of one of those companies, Mark Indovina, became an advisor.

We nailed the deadline to get a prototype RTL processor core developed for another client, enabling them to meet the deadline for a multi-day meeting with one of their key clients in Japan. We had to make two flavors, optimizing for power and performance constraints.

I am happy that we have grown to the point where it made sense to incorporate in Russia; growing our team there was a significant milestone.

Q: What has been the biggest surprise? What was one key assumption you made, perhaps even unconsciously, that has caused the most grief?

Sales. Selling is tough. I came back from Eastern Europe and a bit arrogantly thought that as a skilled group of engineers, we would break into the contractor market easily. I have learned that selling is a process. I am not a salesperson in the least. My father was and I didn’t have much respect for the field, until I tried it.

Q: What development, event, or new understanding since you started has had the most impact on your original plan? How has your plan changed in response?

I thought that I would be spending a lot more time in Eastern Europe than I have been. We have a senior team there that runs smoothly. The last time I was there I was trying to hire another engineer for a project and was unsuccessful. Salaries were skyrocketing.

We are more local and on-site and less offshore-centric today than I initially envisioned.

Q: What suggestions do you have for entrepreneurs?

Be willing to deviate from your plan; not necessarily from your objective. I think we can (I did) envision the future too much or too specifically, such that you have a hardened mold in your mind and of how something will turn out and then work like hell to fill that mold with your progress. Sometimes there is less resistance to your goal down a parallel path.

Q: Thanks for your time.

Linc Jepson is a long time attendee at Bootstrapper Breakfasts®, and has a brief video interview up at “Linc Jepson: the main reason I come to the Bootstrapper Breakfast

Bug Reports vs. Business Plans

Written by Sean Murphy. Posted in 1 Idea Stage, 4 Finding your Niche

“Unhappy users in our industry don’t continue to file bug reports; they start writing business plans.”
Michael “Mac”  McNamara talking about the EDA Industry

I was reminded of this remark that Mac made at an SDForum event several years ago as I was reading the “Who Are User Entrepreneurs?” study which summarizes findings on innovation, founder characteristics, and firm characteristics released in February of 2012. I was alerted to it by a press release from Kauffman released yesterday “Nearly Half of Innovative U.S. Startups Are Founded by ‘User Entrepreneurs‘”

Here are some interesting passages from the study, with some further commentary mixed in.

What is User Entrepreneurship? User entrepreneurship is defined as the commercialization of a new product and/or service by an individual or group of individuals who are also innovative users of that product and/or service. A user entrepreneur tends to experience a need in her life and develop a product or service to address this need, often before founding a firm. As a result, user entrepreneurs are distinct from other types of entrepreneurs in that they have personal experience with a product or service that sparked innovative activity and in that they derive benefit through use in addition to financial benefit from commercialization.

I would suspect that these entrepreneurs bring domain knowledge and often an ability to offer differentiated services based on their own inventions allowing them to bootstrap.

We find key differences among users who founded firms around innovations meant for use in a previous job or business (professional-user entrepreneurs) and users who founded firms around innovations meant for personal use (end user entrepreneurs). […]

The differences may have as much to do with education and socio-economic background and the causality may run in the other direction.

Professional-user entrepreneurs appear to have more experience along a number of dimensions than do other entrepreneurs in both the full sample of firms and the subset of firms conducting R&D in their first year of operations. Although the founders are, on average, the same age, they report higher educational attainment and more years of industry work experience, are more likely to have founded a firm before, and are more likely to have founded a firm in the same industry before. Their firms are less likely to be founded at home, less reliant on self-financing, more likely to receive venture capital financing, more likely to have revenues— and, among firms with revenues—generate higher revenues and are more likely to possess patents and trademarks than both the full sample and subset of firms conducting R&D. […]

There are certainly many “change agents” who improve the robustness and viability of the firms they work at (but didn’t found). Also called intrapreneurs or bricoleurs. These folks may set out on their own to start a new company as well.

End-user entrepreneurs appear to have a demographic profile distinct from the full sample of firms and the subset of firms conducting R&D in their first year of operations. End-user entrepreneurs are more likely to be members of minority groups: they are more likely to be female; more likely to be American Indian, Alaskan Native, or Black; and less likely to be Asian. Their firms employ fewer workers, have lower revenues, are more likely to be founded at home and operate from home five years after founding, are more heavily self-financed five years after founding, are less likely to receive bank financing, and are more likely to possess patents than are entrepreneurs in the full sample and subset of firms conducting R&D.

What are the implications

  • In fast moving fields, especially when you are selling to businesses, good user relationships are essential for encouraging enhancement suggestions that are viable and , if ignored, will lead to new competitors springing up.
  • The harvest of insights from early users can be as important as the revenue and the testimonials a business relationship generates.
  • Delivering the initial version of your product as a service looks like it may be a marker for success. Other techniques for “selling the result”  instead of the product may be equally potent (e.g. selling the holes in the wall where the customer wants them instead of selling a drill).

A Picture Is Worth a Thousand CPU Hours

Written by Sean Murphy. Posted in 1 Idea Stage, Events, Rules of Thumb, skmurphy

Last week was a thought provoking one for me with Big Data Camp on Mon February 27,  a “Great Demo” workshop on February 29, and a tour of the Strata 2012 exhibit hall on March 1. I encountered either examples or stories of visualizations that required hundreds to thousands of CPU hours to create at all three events.

Thanks to the commodization that cloud computing is now driving you can spend $100 and get 1,000 CPU hours, the equivalent of a dozen racks of computers in a data center for a day depending upon how many chips and cores you can fit usefully into an enclosure. It’s roughly equivalent to running one computer for a six weeks to get an answer.

A lot of these visualizations involve streaming data, realtime processing,  or graph calculations where one size does not fit all:  SQL and Hadoop are not going to be the pervasive computing building blocks.  I think we are in for a period of intense experimentation and search for new technologies (or perhaps old algorithms that can be given the computing horsepower they need to get rapid solutions.

A lot of weird new data stores, visualization techniques, and parallel processing methodologies are going to find problem architecture niches and become familiar tools in the next few years.

“When the going gets weird, the weird turn pro.”
Hunter S. Thompson

Three questions for those playing the home game

  • What’s the strangest useful visualization you have seen in the last year?
  • What picture is worth a thousand CPU hours–or about $100 at a dime an hour on Amazon–to you?
  • What models for “Insight as a Service” do you see emerging?

Related Blog Posts

Pretotyping – Techniques for Building the Right Product

Written by Sean Murphy. Posted in 4 Finding your Niche, Books, Rules of Thumb, skmurphy

Alberto Savoia defines pretotyping as determining that you are “building the right product before you invest in building your product right.” His book “Pretotype It” (Second Edition available as a Free PDF or on Kindle for $0.99) lists a set of seven techniques for pretotyping on pages 39-40. This post analyzes and elaborates on the techniques from the book (bold text is from the book) and then offers five additional ones that should be included.

Seven Basic Pretotyping Techniques

  1. The Mechanical Turk – Replace complex and expensive computers or machines with human beings.
    Also known as

    • starting with a service
    • wrapping a thick protective blanket of consulting around your product so that no one is hurt by it
    • selling the holes not the drill
    • Wizard of Oz (pay no attention to the man behind the curtain).
    • Flintstoning (Fred Flinstone’s feet powered his “car”).
    • Manualating (a backward formation from automating)
    • the concierge method
  2. The Pinocchio – Build a non-functional, “lifeless”, version of the product.
    Useful for form and fit validation. Jeff Hawkins famously carried around a block of wood to get an appreciation for what a PDA might feel like.
  3. The Minimum Viable Product (or Stripped Tease) – Create a functional version of the product, but stripped down to its most basic functionality.
    A basic approach for any bootstrapper – make sure you have the simplest offering that customers are willing to buy before you worry about adding features (and delaying time to break even revenue).  In reading this Savoia is using the Marty Cagan MVP model “smallest possible product that has three critical characteristics: people choose to use it or buy it; people can figure out how to use it; and we can deliver it when we need it with the resources available – also known as valuable, usable and feasible.”
  4. The Provincial – Before launching world-wide, run a test on a very small sample.
    Start in a niche. When in doubt zoom in for traction.
  5. The Fake Door – Create a fake “entry” for a product that doesn’t yet exist in any form.
    I am not a fan of this except in very limited circumstances for B2B markets as it can be very corrosive to the trust required to built a long term business relationship. And at least with software products for business, a longer term relationship is normally intrinsic to the customer’s calculation of the value of your offering. If you start to erect “Potemkin village” products that have too many false fronts or facade items in your menus and options prospects may doubt the entire offering.
  6. The Pretend-to-Own – Before investing in buying whatever you need for your product, rent or borrow it first.
    Find a way to use tooling or equipment before committing to  a significant purchase.
  7. The Re-label – Put a different label on an existing product that looks like the product you want to create.
    Often a more complex product can have menu items deleted or entire branches of a menu tree pruned to explore whether this is a market for a simpler offering. At Cisco we didn’t stuff two connectors on a four port router and changed the paint job to create a “lower cost” model until the box could be re-designed.

Five I Think Should Be Added

  • The holodeck – simulate the effect of a product on a workflow: understand where the next bottleneck is to determine how much benefit eliminating one or more steps (or reducing one or more category of error) will actually yield. This is the default method for “system on a chip” design approaches but I suspect we will see more service workflow simulations as a part of the development of new service offerings in the future.
  • Family Tree – verify that manual implementations exist for what you plan to automate, has someone written an Excel macro (or an EMACS macro)  to solve the problem. Are people already following a checklist to prevent a category of errors? Replacing workarounds involves less behavior change (at least in terms of a customer’s view of the real problem) than getting them to try something without antecedents.
  • “What’s On Your Mind” – understand the customer’s view of the problem and the constraints your solution has to satisfy before proposing one.  This normally requires an active curiosity about the customer’s perception of their needs.  This is not the same as asking them for features and implementing them without considering the deeper implications.
  • Picnic in the Graveyard – do research on what’s been tried and failed. Many near misses have two out of three values in a feature set combination correct (some just have too many features and it’s less a matter of changing features than deleting a few). If you are going to introduce something that’s “been tried before” be clear in your own mind of what’s different about it and why it will make a difference to your customer.
  • Want Ad – ask customers to write up a job description with a focus on “results to be achieved” by your product. Clayton Christensen calls this the “jobs to be done” model for a new produce (See also Chapter 3 from Innovator’s SolutionWhat Products Will Customers Want to Buy

Savoia Adds “One Night Stand”

In workshops given after the second edition was published Savoia has added a new technique: The One Night Stand. Primarily aimed at retail innovation it says you can create “a complete service experience without the infrastructure required by a permanent solution.  Here are some details from the  “Pretotyping Cheat Sheet” by Leonardo Zangrando (

  • How: Deliver target customers the real experience in an extremely narrow geographic scope and time frame.
  • Why: Avoid large infrastructure investment until validating market interest and actual use.
  • Where: In the same real-life situation where the innovation will be used but with limited time and geographic scope.

Three situations where this is most appropriate:

  1. The solution is-or depends critically upon–an interactive service experience
  2. You expect demand for the offer will be sensitive to the choice of channel, and you need to test a number of possible customer interception points
  3. You want to validate a large homogeneous market before scaling up

I think this is an intelligent elaboration on what was called “The Provincial” in the second edition but it’s particularly appropriate where a specialized facility can be replaced in a trial for a temporary setup (e.g. a tent in the parking lot of an existing store, a stall in a farmers market, a rented facility in preference to building your own before you have determined there is a need).

Related Articles and Blog Posts

Building Codes For Software

Written by Sean Murphy. Posted in skmurphy

“I think it’s still a long time before we’ll have building codes for software.”
Rich Skrenta “There is no building code for software”

A building code is a function of the municipality issuing building permits for what is essentially custom construction. Products (and product liability calculations) are driven by tort liability, which is managed by insurance and mechanisms like Underwriters Laboratory testing and certification. It’s also managed at a Federal and state level by legislation

Even though most software licenses have strong limits on any warranty or liability I suspect that we will see both insurance-driven and legislative “design codes” imposed on software as it becomes as woven into the fabric of our lives as wood and sheet rock. The “software content” of not only airplanes but automobiles, power tools, appliances, and many other products is increasing each year.

Traditions are solutions to problems that we forgot we had. Most of the building codes are “rules of thumb” to prevent injury or loss of life, hardwired into legal mechanisms to prevent folks from cutting corners to save money and imposing a downstream risk.

Rich Skrenta is probably  right that it will be a while before application software becomes subject to “building codes” but embedded software, and not just for medical implants and airplanes, may transition in a matter of a decade or two.

Two good books that bear at least tangentially on this topic are “How Buildings Learn: What Happens After They are Built” by Stewart Brand and “House” by Tracy Kidder (also author of “Soul of the New Machine“). Kidder offers a number of examples for why it’s hard to develop an innovative design for a house that is actually livable.

Sketching The Likeness Of An Imaginary Business

Written by Sean Murphy. Posted in 1 Idea Stage, 3 Early Customer Stage, Books, skmurphy

Startups originate in the mind of an entrepreneur, often as the result of observing something that seems odd, or is the result of juxtaposing two or three seeming unrelated or even incongruous ideas. The first challenge the entrepreneur faces turn his insight into something others can critique and improve upon: to show them sketches of the imaginary business and ask for feedback.

Founder Story Sam Wurzel, Octopart

Written by Sean Murphy. Posted in Founder Story, skmurphy

This originally appeared in my “Entrepreneurial Engineer” column in EETimes as “Octopart helps nearly a half a million people find the part they need every month” on Mar-3-2011. I have added some additional hyperlinks in this version.

I originally made a note to blog about Octopart in October of 2007 and again in April of 2008 as I used them to research part information for various client projects. The interface has evolved over the years but the site offers a very clean and information rich way to search for parts and part information, aggregating content from a number of sites into a single coherent view. The company was founded by Sam Wurzel, Andres Morey, and Harish Agarwal in 2006. All three had a common background in physics and they have brought a level of rigor along with a hacker perspective to part selection that has created a useful and innovative part selection site.

When they announced in January that DigiKey was allowing them to include their catalog in the Octopart parametric search results I realized I needed to do an interview with them. I was able to talk to Sam Wurzel, what follows is an edited transcript of our conversation with hyperlinks added for context. I have included more on the founders’ backgrounds after the main interview to give readers who are interested a window into their diverse backgrounds and low key humor.

Q: Can you talk a little bit about your background

All of us have a background in physics, and physics is what brought us together. Andres and I became friends while studying physics in college and Harish and Andres became friends while studying physics in grad school. Our areas of research were all different; I was working on plasma physics, Andres was working on experimental cosmology and Harish was working on biophysics.

Q: Can you talk a little bit about what led you to found your company, what was the problem that motivated you?

In 2005 I read Paul Graham’s essay “How to Start a Startup” and it really changed my perspective that I could start a technology company. I was in graduate school in Boulder but it was becoming clear that the academic path was not the right one for me and I had been looking for alternatives.

Soon after that I sent Andres the link, and we started throwing around ideas for startup companies. In the spring of 2006 I got a phone call from Andres. He was having trouble finding a low temperature capacitor for his experiment and suggested that we build a database of electronic parts and make it easily searchable on the web.

Q: Are there any entrepreneurs in your family or who you interacted with when you were growing up?

Both of my grandfathers were entrepreneurs in their own way: one owned a pharmacy and the other owned a car dealership.

Q: So both ran businesses that had complex inventory management issues?

It’s funny, I hadn’t really thought about it before, but now that you ask I remember working in the parts department of my grandfather’s car dealership. Maybe that gave me some insight into the problems that Octopart solves for engineers and electronics hobbyists.

Q: How did you get started?

In mid 2006, we started writing code and learning about web technologies in the little spare time we had. After working in the lab all day, we would come home and write code on a Linux server that I bought at a yard sale for $50. We would often work until 3 or 4 in the morning. By that fall we had a working prototype and applied to Y Combinator. We got some seed funding from them and incorporated the company at the end of 2006. We launched the site in March of 2007.

Q: Can you give me a brief overview of where the company is today?

Today we’ve grown to serve over 440,000 unique visitors per month who are searching for electronic parts. We list the inventories of over 50 distributors, including some of the largest distributors in the industry.

We also have a relationship with Cadence: our aggregated part information is published in the OrCad Component information system.

Q: Octopart is a powerful search tool: what’s the business model? Are you profitable?

Our business model is connecting part buyers with distributors, and the distributors pay for that traffic. We also do display advertising targeted to the electronics industry. We are profitable.

Q: What are the two or three things that you have been able to accomplish that you take the most pride in or satisfaction from?

From a technical standpoint, we’re very proud of the back-end system we’ve built to handle the incoming data feeds and the front end system to serve up fast responses to user queries. From a product standpoint, we’re proud that our users find Octopart useful. Getting emails from users who love the site is great.

Q: What has been the biggest surprise? What was one key assumption you made, perhaps even unconsciously, that has caused the most grief?

When we started Octopart, we were sure that within 6 months, we would have all the major distributors signed up and we would be overwhelmed with users. In fact everything takes longer than we expect it to. That includes building technology, building relationships and getting users. On the surface, it seems like the problems involved in part search are straightforward: get the data, build a system to keep track of it, and build an intuitive frontend interface. But each of those problems have subproblems, and each subproblem needs to be iterated on quite a bit.

Q: What development, event, or new understanding since you started has had the most impact on your original plan? How has your plan changed in response?

Surprisingly, the business model and the design of the site today has not changed that much since we first conceived it. The biggest difference between the original vision and where we are today is the time it took us to get here. We still focus on two critical challenges: getting good data from many sources and correlating it into an integrated view of a part, and offering our users an intuitive and powerful interface for finding the parts that they are looking for. Both are hard problems and although we have made a lot of progress I wouldn’t consider either of them to be fully solved.

Q: You must be doing something right if almost half a million people visit every month looking for part information. Thanks for your time

I have included biographical information supplied by Octopart on the founders as I found it very interesting reading.

Sam Wurzel
Sam graduated from Brown University with a Bachelors degree in physics and engineering. He went to graduate school at the University of Colorado at Boulder where left the PhD program with a Masters degree to work on Octopart.

Sam likes to build things. While a student, he spent alternate summers working in experimental physics labs as research assistant and in bicycle shops as a mechanic. In grad school at CU Boulder Sam joined a newly formed lab testing the design of a fusion plasma confinement scheme which one day might be useful in a commercial fusion reactor.  Although Sam liked the lab work, he realized academia was not a good fit for him. So, he started working on Octopart, and eventually left his PhD program to move to Berkeley to pursue Octopart full time.

At Octopart Sam manages relationships with distributors, writes code to handle their data feeds, and works on techniques to normalize the data arriving from many different sources.

When Sam is not working on Octopart, he enjoys running and reading while on public transportation.

Andres Morey
Andres received his Bachelors degree in physics from Brown University in Providence, RI and attended UC Berkeley for graduate school, leaving the PhD program with a Masters degree in physics to work on Octopart. As a grad student, Andres worked on the IceCube Neutrino Observatory – an experiment that uses neutrino interactions in the ice at the South Pole to map cosmic neutrino sources.

While working on various hardware projects for his experiment, Andres spent a lot of time searching for electronic components. Frustrated with the online search options, he called up his friend and fellow grad student, Sam Wurzel, and suggested that they build an electronic parts search engine. After spending several months working on a prototype they decided to leave grad school to work on the project full time and in November 2006 they incorporated as Octopart, Inc.

Since leaving grad school, Andres has been working on Octopart full time. Andres is responsible for most of Octopart’s consumer-facing features including client-side code and all graphical and UI elements. In the course of working on Octopart Andres discovered that he loves working at a tech startup because it is a platform for solving a series of never ending problems from managing human relationships to squashing obscure Internet Explorer bugs. Andres discovered that he loves coding because of the feeling he gets when he finds an elegant solution to a coding puzzle.

Currently, Andres spends his free time thinking about Octopart.

Harish Agarwal
Harish received his undergrad degree from the University of Illinois at Urbana-Champaign in Engineering Physics and a Masters degree from Cambridge University in Semiconductor Physics. Harish left his physics PhD program at U.C. Berkeley with a Masters degree to join Octopart.

Harish enjoys understanding systems and developing projects that work on top of them. As a graduate student in Jan Liphardt’s biophysics lab at U.C. Berkeley, Harish was tasked with studying nuclear transport in eukaryotic cells. Having come from a physics undergraduate education, this involved hitting the books and pestering kind colleagues for advice and gems of wisdom. This crash course preceded many long days at the bench developing biological protocols and a microscopy system to track nanometer scale cargo transit through the nuclear pore on millisecond timescales.

Harish left academia in the spring of 2007 to join two friends in developing Octopart, a search engine for electronic parts. Having come from a biophysics lab to work on an already launched website without knowing exactly what MySQL stood for, this involved a lot of intense on the job learning. In the past three years, Harish has had the opportunity to work on many nooks and crannies of Octopart, from developing front end user features, to hacking search capabilities into open source search engines.

Octopart has been covered in the last year by

Follow Up: Octopart Acquired in Aug-2015 by Altium

I’m very happy to share big Octopart news today: We’ve signed a definitive agreement to join Altium, a leading provider of design software for the electronics industry. Octopart will become a wholly owned subsidiary of Altium and continue to grow and operate independently from our headquarters in New York City.

First things first: this deal is good for Octopart users. Octopart will be increasing the pace of what we’ve always done: opening up access to part data for design, sourcing, and manufacturing. Our search engine will stay free, fast, and open. Our API will remain available to existing and new users. We’ll continue to provide component data in a clean, straightforward way without any pop-up ads or unnecessary forced registration nonsense.

We’re joining forces with Altium because we think we can innovate and grow Octopart more quickly together than apart. We live in a time when electrical engineers, makers, and hackers have high expectations for their design tools and for component search. Octopart users expect rich content like CAD models and reference designs at their fingertips when doing component selection. Users of PCB design software expect supply chain intelligence at hand when they are designing new products. Bringing Octopart together with Altium will make this possible, and more. We envision a future where going from prototyping to production is a seamless experience and we’re going to work together to make that vision a reality.

From the beginning of Octopart we’ve been encouraged and spurred on by our investors, our families, and our friends, and we are deeply grateful for that. Today Altium joins that group which will only accelerate Octopart’s forward motion. 

Quick Links

Bootstrappers Breakfast Link Startup Stages Clients In the News Upcoming Events Office Hours Button Newsletter SignUp