Why are we here? Philippe Baumard, Ph.D. Associate Researcher, UC Berkeley Institute of Business & Economic Research (I.B.E.R) [email protected]Visiting Professor, Stanford University School of Engineeing Dept of Aeronautics and Astronautics [email protected]Copenhagen Business School Executive MBA San Francisco Retreat 2008 Friday, November 14th 2008
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Why are we here? Philippe Baumard, Ph.D.
Associate Researcher, UC BerkeleyInstitute of Business & Economic Research (I.B.E.R)
Ph. Baumard, Institute of Business and economic Research, UC Berkeley / Sgtanford / 2008 – FYEO - Do not distribute
This is what I could try to address today… The so-called « Chasm »
2
Ph. Baumard, Institute of Business and economic Research, UC Berkeley / Sgtanford / 2008 – FYEO - Do not distribute
The chasm = A magma of failing ideas
3
« How do we make sure that this relentless bubble of repetitive failures keep
its existence and momentum »
Ph. Baumard, Institute of Business and economic Research, UC Berkeley / Sgtanford / 2008 – FYEO - Do not distribute
A few scientific findings about innovation
• Finding #1: Innovation is unpredictable• All attempts to create a predictive model of innovation generation have failed• Andrew Van den Ven and his colleagues even empirically proved wrong the famous « action – reaction » loop
(single loop, double loop, Type II learning, etc).
• Finding #2: Creativity, on the other hand, can be managed• Creative individuals are driven by intrinsec motivations• Creativity requires a very high failure intensity and failure tolerance• Not everyone can be creative, but larger numbers increase the probabilities!
4
1 + 2Let’s create a place where a lot oftalented people can repeatedlyfail at being creative again…
Ph. Baumard, Institute of Business and economic Research, UC Berkeley / Sgtanford / 2008 – FYEO - Do not distribute5
Ph. Baumard, Institute of Business and economic Research, UC Berkeley / Sgtanford / 2008 – FYEO - Do not distribute
Part I
Why the Bay Area?
6
Ph. Baumard, Institute of Business and economic Research, UC Berkeley / Sgtanford / 2008 – FYEO - Do not distribute
Simply put:
• On this small rectangle, you have the same total market capitalization that the total of all French Stock Market cap. (2008)
7
Ph. Baumard, Institute of Business and economic Research, UC Berkeley / Sgtanford / 2008 – FYEO - Do not distribute
What is so specific about the Silicon Valley?
• “What distinguishes Silicon Valley is not its scientific advances or technology breakthroughs. Instead, its edge derives from a “habitat” or environment that is tuned to turning ideas into products and taking them rapidly to market by creating new firms.”
• “The enduring competitive advantages in a global economy lie increasingly in local things — knowledge, relationships, motivation — that distant rivals cannot match. This role of location has been long overlooked, despite striking evidence that innovation and competitive success in so many fields are geographically concentrated.”
Michael Porter
• The Silicon Valley constitutes a world unmatched cluster where the cycle from idea to product is on average 30% faster than in any other clusters.
Ph. Baumard, Institute of Business and economic Research, UC Berkeley / Sgtanford / 2008 – FYEO - Do not distribute
InventorsInventors are crucial to IP licensing and valorization are crucial to IP licensing and valorization in the Silicon Valleyin the Silicon Valley
(1) Jensen, R. and M. Thursby (2001) “Proofs and Prototypes for Sale: The Tale of University Licensing” American Economic Review(2) Lowe, R. (2003) “Entrepreneurship and Information Asymmetry: Theory and Evidence from the University of California” Carnegie Mellon Working Paper 2003-09(3) Jansen, C and H. Dillon “Where do the Leads for Licenses Come From?” AUTM Journal
0
0.1
0.2
0.3
0.4
0.5
0.6
Inventor(s) Licensingprofessional
Companysolicitation
ResearchSponsor
Unknown
Lead sources of licensed technologies3
Survey results show inventions are rarely more than ideas and concepts at time of disclosure
79% of inventions require considerable development (29% have no prototype, 48% have a lab prototypes) 1
71% require ongoing inventor involvement to develop the invention
Extreme cases of inventor involvement result in inventor start-ups 2
Finding licensees lies on community networkingCompanies funding the research often do not license, sometimes by university’sor founders’ rules
Challenge fall to licensing professionalsSuccess of IP licensing in the Valleyis based on interpersonal networking(from companies to licensing professionals and inventors)
Ph. Baumard, Institute of Business and economic Research, UC Berkeley / Sgtanford / 2008 – FYEO - Do not distribute
The IP dynamic in California:From Universities to Business
• Largest Silicon Valley’s firms are often based on university technologies– Hewlett-Packard (Stanford)– Biotechnology: Genentech (Stanford, UC San Francisco), Chiron (UC Berkeley)– Internet software: Lycos (Carnegie Mellon), Inktomi (UC Berkeley), Google (Stanford)– University of California had two very early successes in biotech, and the Cohen-Boyer patent
created much of the industry– More than 202 of current leading biotech companies started at UC Berkeley!
• Between 20-30% of Californian Universities’ licensees are start-ups– Data suggests that ~67% of these startups involved the inventor at founding– Emerging California “pockets of entrepreneurship”: Santa Barbara (materials and electrical
engineering), Irvine (pharmaceutical development)– Clusters near inventors
• California Universities hold 2 out of 5 top positions in patent rankings (2002-2008)– 1) Massachusetts Institute of Technology 134– 2) California Institute of Technology 102– 3) Stanford University 96– 4) Wisconsin Alumni Research Foundation 87– 5) Johns Hopkins University 86
Ph. Baumard, Institute of Business and economic Research, UC Berkeley / Sgtanford / 2008 – FYEO - Do not distribute
What are the core advantages of the Silicon Valley in terms of IP development?
Hard economies• +/- Supply chains (possibility of IP application)
• ++ Labor pools (largest engineer cluster in the US)
• ++ Specialized services (large IP licensing professionals community)
• ++ R&D and technology (Intel, Cisco, etc.)
• ++ Capital (largest US capital/employee)
Production
• - - Costs (the SV has high salary rates)
• -- More options (current trend is outsourcing in developing countries where the education level is rising quickly)
• +/- Quicker responses (short IP to prototype cycles, but farther from production clusters)
Soft economies• ++ Association (strong IP community)• ++ Networking (contact with inventors)• ++ Tacit learning (frequency of contacts)• ++ Knowledge leaks (strong university-
Ph. Baumard, Institute of Business and economic Research, UC Berkeley / Sgtanford / 2008 – FYEO - Do not distribute
Part II
Lessons learnedabout innovatingwhile in the Bay
12
Drawing fromK. Von Braun, aged 16
Ph. Baumard, Institute of Business and economic Research, UC Berkeley / Sgtanford / 2008 – FYEO - Do not distribute
1 – Changing the way we do « innovation »
• Get it early, very early on– Idea generation takes place at « lab work » level. When it has been
published, it’s already too late.
• Experiment with final-end customers, suppliers and competitors– This is what the Bay Area does the best: co-opetition, cooperating with
rivals, being obsessive about customer behavior, habits, human factors.
• Scale up as fast as you can– Ramp up exaggerately and cultivate acceptance for early failure
13
Lab Work – Idea generationPhD Seminars
Working Paper Submission
Conferences Publication IP Prototype Market
6 months to 2 years (depending on domains) 1 year (depending on performance) 1,5 year (depending on support)
PhD studentResearchers
PhD AdvisorResearchersLabs Libraries
Journal EditorsConference ReviewersLab Open Research Seminars
Databases Universities IP OfficesIP licensing professionals
Ph. Baumard, Institute of Business and economic Research, UC Berkeley / Sgtanford / 2008 – FYEO - Do not distribute
2 – Even the most innovative have inertia
• Original early schools of innovation are already outdated– Servicization is already an outdated solution to commoditization
– The era of dominance of mechanical engineering in design is gone, despite the Iphone
– Designing services innovation cannot be accomplished coming from mechanics, machines, ergonomics and human factors. That’s not the way great cooks turn into Gault & Millaut’s chefs. At least, not yet
• Etc, etc, etc
• The shared talent of the Bay Area is its continuous listening– Dogmas do not survive for long in the Bay
– Dogmatic thinking, strong convictions, « worldviews » are rapidly washed away by VCs’ disbelief, critical peers, unforgiving market place
– Hence, everybody listens very carefully
14
Ph. Baumard, Institute of Business and economic Research, UC Berkeley / Sgtanford / 2008 – FYEO - Do not distribute
3 - As much as you can, trash these:
15
Problem solving
Deduction
Induction
Imitation
Formal Heuristics
Technological trajectories
Ph. Baumard, Institute of Business and economic Research, UC Berkeley / Sgtanford / 2008 – FYEO - Do not distribute16
4 - As much as you can, promote these:
Csíkszentmihályi, Mihaly (1996). Creativity : Flow and the Psychology of Discovery and Invention. New York: Harper Perennial
Anomalies
Constraint learning
Creativity is “something between discovering and witnessing”. R. Moog
“Engineering is always improvisation” R. Moog
evolutionary, incremental
and accidental
“It’s impossible to have a disembodied intelligence”
Rodney Brooks
Fast,
Cheap
And out of control
Ph. Baumard, Institute of Business and economic Research, UC Berkeley / Sgtanford / 2008 – FYEO - Do not distribute17
5 – Step back, and nurture “cold cases”
• Disruptive functionalities must be greatly significant– With an increasing commoditization, new entrants or incumbents
must offer “disruptive functionalities” with an absolute cost advantage
• JaJah vs. Traditional call-back system (+00010..): No significant advantage (<5% price difference)
• Business model resilience relies on a core proprietary core process
– Google popularity indexation, integrated to desktop search– Skype super node + P2P technology (cross-platform, clarity of
voice)
• Symbiotic entries are more likely to change industry dynamics
– Itunes/Ipod/RIAA vs. Archos– Early antagonistic entries threaten sustainability (e.g. JaJah, EQO,
etc).
Ph. Baumard, Institute of Business and economic Research, UC Berkeley / Sgtanford / 2008 – FYEO - Do not distribute
6 - The best innovation is probably not yours, but your ecosystem is….
• The “Dilemma Zone” is counter-intuitively the best ground for strategic rupture
– Diminish the awareness of the competition– Benefit from the causal ambiguity (of resources dependency + strategic intent)– The best way to develop a services ecosystem is to use dynamic freedom to let
the customers articulate and defend the desired ecosystem (e.g. YouTube)
Ph. Baumard, Institute of Business and economic Research, UC Berkeley / Sgtanford / 2008 – FYEO - Do not distribute
Synthesis
• Shorter cycles from inventor’s creativity to prototyping and IP– The density and frequency of relationships between inventors, universities,
industries, and IP communities is accelerating the pipeline
• Failure rate is unavoidable, but not unlearning from failures is unforgivable
– The “secret” of the Bay Area is that it is a very large pond with very smart fish species that die very quickly
– Like “Little Nemo in Slumberland”, failing is not a dramatic matter as far as the inventive production level is maintained at a high intensity
• Access to weak signals and strategic learning preemption– From researchers’ first working papers to first claims, the average pipeline can
reach 9 months– Intense collaboration between university labs and industrials allow for detection of
early signals and preemptive learning strategies
• More integration, more last stages, more industrialization– Current economic conditions fosters late stage VC investment– IP activity is likely to concentrate on applications, interoperability, industrialization
(processes) and cross-platforms, cross-networks technologies
Ph. Baumard, Institute of Business and economic Research, UC Berkeley / Sgtanford / 2008 – FYEO - Do not distribute
Food for thought
20
Ph. Baumard, Institute of Business and economic Research, UC Berkeley / Sgtanford / 2008 – FYEO - Do not distribute
Sutton – Weird Ideas that Work
1. Hire “Slow Learners” (of the organizational code). 1-1/2 Hire People Who Make You Uncomfortable, Even Those You Dislike.
2. Hire People You (Probably) Don’t Need3. Use Job Interviews to Get Ideas, Not to Screen Candidates4. Encourage People to Ignore and Defy Superiors and Peers5. Find Some Happy People and Get them to Fight6. Reward Success and Failure, Punish Inaction7. Decide to Do Something That Will Probably Fail, Then Convince Yourself
and Everybody Else That Success is Certain8. Think of Some Ridiculous or Impractical Things to Do, Then Plan to Do
Them.9. Avoid, Distract, and Bore Customers, Critics, and Anyone Who Just Wants
to Talk About Money10. Don’t Try to Learn Anything from People Who Seem to Have Solved the
Problems You Face.11. Forget the Past, Especially Your Company’s Successes.
Sutton, Robert I. 2002. Weird Ideas that Work: 11-1/2 Practices for Promoting, Managing, and Sustaining Innovation. New York: Free Press.