In this deck from the Stanford HPC Conference, Thomas Thurston from WR Hambrecht Ventures presents: Surprising Mechanical Lessons About Predicting Innovation Success.
"For a century, corporate innovators, entrepreneurs and venture capitalists have had to rely on their instincts when deciding which strategies to pursue, and where to invest for growth. Now data science is turning human instinct on its head with powerful decision technologies that are giving rise to counter-intuitive discoveries about market behavior and predicting innovation success. Learn how venture capital firm WR Hambrecht is using big data and machine learning to better identify growth opportunities, predict new business success and rewrite the rules of innovation."
Thomas Thurston is a Partner and Chief Technology Officer at WR Hambrecht Ventures. For more than 50 years WR Hambrecht and predecessor Hambrecht & Quist have provided billions of dollars in early growth capital to more than 500 companies worldwide. Examples include Adobe, Amazon, Apple, Genentech, Google, Intel, Pixar and Salesforce. Today the firm guides its investments with an advanced computing system that provides unique insights into private markets; including an ability to pinpoint disruptive companies around the globe. This computing system is called MESE® (sounds like “peace”). Formerly, Thomas used data science to guide growth investments at Intel and led joint R&D between Intel and Harvard, with Professor Clayton Christensen, on the use of algorithms to guide innovation and growth efforts. He also helped found a business in the high performance computing industry that was acquired. He holds a BA, MBA, Juris Doctor and was a Fellow at the Harvard Business School.
Watch the video: https://youtu.be/GehBHi83Yt4
Learn more: https://www.wrhambrecht.com/
and
http://www.hpcadvisorycouncil.com/events/2020/stanford-workshop/
Sign up for our insideHPC Newsletter: http://insidehpc.com/newsletter
Take control of your SAP testing with UiPath Test Suite
Surprising Mechanical Lessons About Predicting Innovation Success
1.
2. Bill Hambrecht
For more than 50 years WR Hambrecht and predecessor
Hambrecht & Quist have provided billions of dollars in early
growth capital to more than 500 companies.
The firm guides its investments with an analytics system called
“MESE®” (sounds like ‘peace’).
3. Important decisions
High risk and uncertainty
Little or no reliable data
THE CHALLENGE
99% OF BUSINESSES ARE NOT PUBLICLY TRADED
4. 4
Small decision sets (often n=1)
Ad hoc decision criteria
Personality-driven
Constrained by geography
5. >
>
>
MACHINE LEARNING AND SIMULATION
Actuarial diligence to better calculate the odds of a
businesses surviving or failing within the investment horizon.
DATA HARVESTING AND ANALYTICS
See across global markets to identify which
companies are winning or losing at any time.
MARKET EXAPTATION SIMULATION ENGINE (MESE®)
6. MARKET TRACTION
• Traction = quantitative evidence of
market demand. Used to estimate
market cap.
• Created by digitally mining and
weighting unstructured data from select
digital sources.
GROWTH INDICATOR
• Rate at which a business’s market
traction is growing or in decline.
Uber
Zim
ride
NuRide
Gtrot
TaxiFinder
Avego
PickupPal
Carticipate
RideFinders
RideShark
Ridesharing
January, 2011
Uber identified as market leader when it had only
raised $1.5M in a Seed round
ApeelSciences
Reed
W
ax
Am
azein
IGIW
ax
Proinec
JBTFoodtech
Stafresh
M
onosol
Naturew
ax
Calwax
Fruit and Vegetable Coatings
June 1, 2015
Apeel Sciences valued at $5.08M in a Seed round
Valued at $420M by 2018
Biocartis identified when it had only raised $14.8M
in a Series A round. Valued over $675M by 2018
Biocartis
GalapagosNV
AkonniBiosystem
s
JubilantLifeSciences
Quidel
Accelerate
Diagnostics
HTG
M
olecular
M
indRay
Genetesis
Abingdon
Health
Utah
M
edicalProducts
Molecular Dx & Biomarkers
October, 2009
Airbnb
RedAw
ning
Hom
eAw
ay
SpareRoom
FlipKey
VRBO
Room
ster
Peer-to-Peer Lodging
October, 2010
Airbnb identified as market leader when it had only
raised $640K in a Seed round
Visibility into companies that otherwise
disclose little or no data about their
financial or commercial status
SEE ACROSS MARKETS. PINPOINT WINNERS.
7. 7
Where can the fewest
resources have the
broadest impact?
How will change in one
part of the market impact
others?
Where should specific
businesses invest for
growth?
8. ACTUARIAL DILIGENCE
SIMPLE EXAMPLE
Strategy SurvivalRate,GrowthType
BetterPerformance 64%Survival. IncrementalGrowth
MarketPosition
Incumbent
Business
BetterPerformance 14%Survival
NewEntrant
Low-EndorNew-Market 66%Survival. HighGrowth
9. WHICH TECHNOLOGY
WILL WIN?
9
WHEN PREDICTING WHICH
TECHNOLOGY WILL WIN,
COMPANIES
NOT TECHNOLOGIES,
ARE THE RIGHT UNITS OF ANALYSIS
THE BEST TECHNOLOGIES OFTEN LOSE
AND MEDIOCRE TECHNOLOGIES OFTEN WIN
10. STEP 1
Big Data
• Scouting and
screening
• Who is in the
market?
• Who is winning?
STEP 2
ML and Simulation
• Actuarial diligence
• What are the
odds?
• How can they
improve?
STEP 3
Human Due Diligence
• Qualitative issues
• Team
• Technical
• Regulatory
STEP 4
Portfolio Management
• Ongoing monitoring
• Follow-on decisions
• Pivots and support
• Exits and liquidity
HUMAN DECISIONS ENHANCED BY TECHNOLOGY
12. 12
SURPRISING MECHANICAL LESSONS ABOUT
PREDICTING INNOVATION SUCCESS
1. Big data, ML and Simulation are exposing flaws in traditional
opportunity scouting:
2. Due diligence has been void of actuarial diligence
a. ex. new entrants should avoid “better performance” strategies
3. When predicting which technology will win, companies
(not technologies), are the right units of analysis
TRADITIONAL TECHNOLOGY-ENABLED
a. Small decision sets (often n=1) a. See across markets, pinpoint winners
b. Ad hoc decision criteria b. Pattern recognition
d. Personality-driven c. Data-driven
d. Constrained by geography d. Global
13. >
>
>
This presentation does not, nor is it intended to, contain investment advice. There are no guarantees that any results discussed in
this presentation will be achieved and there is no guarantee that the views and opinions expressed in this presentation will come
to pass. Different types of investments involve varying degrees of risk, and there can be no assurance that any investment will be
profitable. Comparison of all benchmarks to other indices or references to any stocks are for comparative and educational
purposes only. All models, graphs, charts or other demonstrative images or formulas in this presentation have limitations and are
difficult to use, requiring experience and expertise. Investors should be cautious about any and all investment recommendations
and should consider the source of any advice on investment selection. Various factors, including personal or corporate
ownership, may influence or factor into an expert's investment analysis or opinion. All investors should conduct their own
independent research into individual investments before making a purchase decision. Information presented herein is subject to
change without notice and should not be considered as a solicitation to buy or sell any security. Past performance is no
guarantee of future results.
Disclaimer