New Venture C ommercialization of C lean Energy Technologies
David S. Miller
MIT Engineering Systems Division
1-877-531-9017 voice/fax
dsm@ alum. mit.edu
John Sterman
MIT Sloan School of Management
30 Wadsworth Street, E53-351, Cambridge MA 02142
617.253.1951 voice 617.258.7579 fax
jsterman@ mit.edu
This paper examines why new ventures founded to commercialize clean energy technologies that
are cost effective and beneficial to adopters have failed to achieve widespread adoption. A new
venture simulation model was developed that models the cash flow, labor force, market,
competition, and product development for a prototypical clean energy technology venture. When
the model is parameterized to correspond to a venture that starts with superior technology at an
attractive price its behavior corresponds to the experience of many of the companies interviewed
for this research. The modeled venture takes many years to achieve profitability due to long
sales cycles, limits to market growth, and the time needed to gain experience producing and
selling its products, and therefore has a high probability of failure. Analysis of the model results
in a set of guidelines for what these ventures, investors, and policy makers should do to increase
their odds of success.
1 Introduction
This paper addresses how to improve the odds of success of new ventures commercializing clean
energy technologies. We are motivated by the importance of reducing the emissions of
greenhouse gases (most notably CO2) from energy production to address the serious risks of
climate change. It is likely that the wide adoption and use of clean energy technologies is
necessary in order to do so. Furthermore, many clean energy technologies are economically
efficient as well as environmentally beneficial. Numerous advantages to end users include lower
and less volatile energy costs and a more stable and reliable energy supply. However, these
technologies have not been as widely adopted as may be presumed from these benefits, and new
ventures formed to commercialize these technologies have failed to do so.
Clean energy technology can be defined as any technology that reduces harmful emissions
resulting from the production and use of energy. Examples of clean energy technologies include
renewable and/or efficient distributed generation (e.g. solar, wind, geothermal, fuel cells,
cogeneration); energy efficiency technologies which enable the use of energy services at lower
cost to users; intelligent energy management; biofuels; and ancillary products and services that
improve the efficiency of power generation and transmission.
We focus on new ventures because only new ventures have been able to commercialize
disruptive new technologies. And only disruptive technologies have the potential to restructure
the current global energy regime. In every other case in which new technology created a new
New Venture Commercialization of Clean Energy Technologies
industry by replacing a standard commonly used technology, such as when electricity replaced
gas lighting, or automobiles replaced horse-drawn vehicles, new ventures led the way. However,
over the last several decades, as the importance and value of clean energy technologies have
become widely accepted, new clean energy technology ventures have not been able to achieve
success and wide adoption for their products and technologies. Why?
There is an extensive body of literature on how and why innovations are diffused, but less
research has been done on what leads to success or failure for new technology ventures. In the
most substantial work to date, Roberts (1991) found that larger investments of initial capital; the
sales experience of the founders; a marketing orientation of the firm; and a strategic focus of the
firm on its core technology and markets were correlated with success. Utterback, Meyer, Tuff,
and Richardson (1992) found that lasting commitment and persistence were critical for
technology ventures and Hilmola, Helob, and Ojalac (2003) found that reducing product
development time was important. Joglekar and Levesque (2006) determined that allocations of
resources to R&D and marketing should account for the anticipated productivity of those
functions, and that a new venture is better off obtaining a single large investment than multiple
smaller ones. However, prior to this research effort, it was not clear whether these results would
be true for clean energy technology ventures, which have not been specifically studied or
modeled before now.
Based on interviews with clean energy entrepreneurs and other stakeholders and on case studies
of clean energy technology ventures, a new venture simulation model was developed that models
the cash flow, labor force, market, competition, and product development for a prototypical clean
energy technology venture. When the model is parameterized to correspond to a venture that
starts with superior technology at an attractive price its behavior corresponds to the experience of
many of the companies interviewed. The modeled venture takes many years to achieve
profitability due to long sales cycles, limits to market growth, and the time needed to gain
experience producing and selling its products, and therefore has a high probability of failure.
Analysis of the model results in a set of guidelines for what these ventures, investors, and policy
makers should do to increase their odds of success. The venture is better off starting with more
sales and marketing personnel and expertise rather than engineers, and should develop no more
product features than are necessary to sell the product. The venture should forego recurring
revenue and instead receive payments up front whenever possible. A single initial equity
investment in the venture is considerably more valuable than a series of investments.
Govemment policies that raise the cost of carbon emissions; reduce barriers and increase
incentives for adoption of clean energy technologies; and subsidize the development of these
technologies can greatly increase the growth of these ventures and the odds of success.
1.1 Barriers to Adoption
Over the course of four and a half years, the principal author conducted over 100 interviews with
clean energy entrepreneurs and a variety of stakeholders related to clean energy ventures. The
stakeholders include the customers of clean energy technology, energy service providers,
investors in the ventures, and participants in policy-making processes related to clean energy
technologies. Interviewees were selected from both established and newly created clean energy
technology ventures; from large and small customers of these products and technologies; and
from a wide variety of sectors of the industry, including distributed generation, demand side
management, renewable energy generation, energy efficient building technologies, and energy
New Venture Commercialization of Clean Energy Technologies
equipment maintenance. Many of the interviewees were recommended by prior interviewees.
Most of the interviews were informal, though notes were recorded for most. Several formal
interviews were also conducted that were based on a sequence of pre-determined questions; these
were recorded on tape.
Numerous factors were identified that affect the adoption of clean energy technologies. These
include regulatory factors such as subsidies for fossil-fuel based energy and/or clean energy
technologies; real time pricing (or the lack thereof) for electricity use; utility interconnection
requirements and surcharges for stranded costs or standby service; siting restrictions for
distributed generation; and carbon taxes or cap and trade regulations, and regulations to promote
energy efficiency. Also important are market factors such as the price of fossil fuels and of
electricity, uncertainty surrounding the economic benefits of new technologies, and the impact of
new technologies on markets. Institutional and behavioral factors, such as the agency problem in
which decision makers do not receive the benefits of adoption, risk aversion, the learning curve
for users to understand new technologies and the effects of word of mouth (or the lack thereof)
regarding new technologies cannot be underestimated. Finally, the technologies themselves need
to work as advertised and to improve over time.
Though a clean energy technology may be economically advantageous, many positive feedbacks
support established energy technologies and the companies that provide them. Figure 1 depicts
many of these loops.
New Venture Commercialization of Clean Energy Technologies
<Word of «——
mouth>
<Adoption of clean
energy tech>
@) +
Es. Ot word of mouth
Perception of benefit
of new technology
<Product attractiveness
of clean energy tech>
+
Perception of risk of
new technology
+
| Regulations
Competition’ Familiarity with wens SS . a ~~ Infrastructure
advertising/ public competition technology Frice 0 support of
education mi existing
ord
“Adoption oF
existing ae
Figure 1: Loops Effecting the Adoption of Clean Energy Technology
Nobody Gets Fired: Existing energy technologies (primarily based on fossil fuel) have been
widely available for many decades and therefore are very familiar to the public and to
commercial enterprises that are heavy energy users. Therefore, when evaluating which energy
technology to use (or to continue using) a decision maker at a firm understands that he will not
be criticized (or fired) if the firm continue to use the same technologies that it has used for many
years, and which all other firms use as well. Furthermore, any new energy technology will be
perceived as risky (it is not tried and true like existing technologies) and the perception of risk
will detract from the attractiveness of the new technology. Therefore, the “safe” decision is to
continue using and purchasing the existing technology, which reinforces its familiarity and
encourages further use in the future.
Public Awareness: The providers of existing energy technology have an incentive to reinforce
the public’s familiarity with their technology and the perception of risk related to new
technology. They also have the financial resources to mount broad advertising campaigns that
tout the benefits (and familiarity) of conventional energy solutions and aggravate the perception
of risks (and fears) of newer clean energy technology. The success of these campaigns bolsters
New Venture Commercialization of Clean Energy Technologies
the adoption of existing energy technologies, providing further resources to mount future
advertising campaigns.
Regulation Capture: Large energy firms tend to make very large political contributions and
exert considerable influence on policymaking and regulations that govern or are related to energy
production and use. They use this influence to shape regulations that favor or lower the cost of
production of their technologies (e.g. subsidies for fossil fuel exploration and development) and
that increase the cost of providing altematives (e.g. onerous interconnection and siting
regulations for distributed generation technologies). These regulations result in increased profit
for these firms, which they, in part, reinvest to shape future regulations.
Learning and Price: Most technologies become less costly to produce over time. Given that
existing energy technologies have been produced and used for many years, firms understand
them well, which reduces the expense of providing them. Negotiating the leaning curve for new
technologies may require a firm to invest in training and possibly new employees, thus adding to
the expensive of adopting a new technology. The lower price encourages further use of the
existing technologies, and inhibits the adoption of the new technologies and the cost reductions
that would allow them to compete better.
Built Infrastructure: One of the reasons that existing energy technology becomes less costly to
use over time is that marginal costs are lower once its supporting infrastructure is built. A
massive infrastructure has been built to deliver electricity throughout the United States through
the centralized grid. Though electricity users pay charges associated with the creation and
maintenance of that infrastructure, its existence has lowered the cost of large-scale fossil-fuel-
generated power. A developer of a large coal-powered plant does not have to worry about the
cost of creating an infrastructure to deliver the power generated by that plant to end users.
However, the developer of a plant meant to produce hydrogen for use in fuel cells must be very
concemed about the cost of infrastructure to deliver the hydrogen to end users. That cost would
severely hinder the construction of such a plant. Therefore, existing infrastructure supports the
expansion of existing technologies which then justify incremental improvements to the
infrastructure and further use of the existing technologies.
Insufficient W ord of Mouth: The reinforcing “word of mouth” loop is often used to explain an
exponential increase in the adoption of a new technology as new users contact potential users
and encourage further adoption, therefore creating even more new users. However, this only
works if there are enough users to spread the word. If there are many factors inhibiting the
adoption of a new technology (as per above) there may not be enough new adopters to encourage
others to use the new technology. A lack of peers using the new technology may further
discourage any new users from adopting.
1.2. Case Studies
We studied three clean energy technology firms in depth to determine the details of their sales
cycles and the particular challenges they faced (as well as the successes they had) in achieving
wide adoption of their products and services. Leaders of these ventures, and others that were
interviewed, found themselves facing much longer sales cycles and much more conservative
prospective customers than anticipated. They found that low prices for conventional energy
decreased the attractiveness of their technology and that regulations hindered the adoption of
their products.
New Venture Commercialization of Clean Energy Technologies
2 Model Development
We developed a simulation model to better understand the factors that most directly determine
the success or failure of a new clean energy technology venture. The model was designed to
help uncover strategies and policies that would increase the odds of success and of wider
adoption of clean energy technologies.
The focus of the model is a firm that starts with an attractive product, but no customers and few
employees. The model tracks the working capital of the firm, the development of features of the
product, the growth (and contractions) of the firm’s labor force, and the status of each of its
prospective and current customers. Figure 2 is an overview of the model highlighting three
sectors: the firm, the market and the competition. Space precludes full presentation of the model,
however complete documentation is available in Miller (2007).
2.1 The Firm
The key parameter for the firm is its working capital. The firm’s working capital determines
how much capability it can develop, and when working capital runs out, the firm fails. The
working capital is increased by investments and by revenue from selling products, and is used
primarily to pay for the cost of goods sold (COGS) and to create and enhance the firm’s
capabilities, primarily through hiring engineering and sales and marketing personnel (the salaries
of these personnel in the model include all non-production operating expenses of the new
venture). The engineering personnel create and enhance the features of the firm’s product. The
sales and marketing personnel expend effort (e.g. direct selling, creation of marketing material,
advertising, etc.) to increase the attractiveness of the firm’s product to the market.
The firm’s working capital is affected by two important loops. One is the “positive cash flow
loop” in which working capital spent to develop products and make them attractive to the market
results in sales and revenue to the firm. This process increases working capital and enables the
firm to make the product more attractive and generate even more revenue. The other important
loop is the “running out of money loop” in which working capital is spent to increase the firm’s
capabilities, and the more capabilities the firm has, the more working capital it needs to spend.
The “running out of money loop” runs in a much shorter timeframe than the “positive cash flow
loop”, creating some of the challenges we will explore in later sections.
The cash flow sector of the model is based on aspects of the financial accounting module in
Oliva, Sterman, & Giese (2003), the product development sector of the model is based on the
inventory management sector described in section 18.1 and figure 19-5 of Sterman (2000), and
the labor sector of the model is closely based on the labor supply chain introduced in Section
19.1 of Sterman (2000).
New Venture Commercialization of Clean Energy Technologies
ri Revenues, === 3 2 2
Firm Va
ee
Positive cash
+ flow
Inflows of
capital Working | ___y__ »—>
Cantal Outflows of
capital
uu 8
@
Running out of
money cocs
coGs
S pendin “|
pending \
Va @s
| R
Pi
FY Ta pt oe Ht culpa
+ Product . Comp Comp | ee
Sales/Marketing Devi Capab Coton ee
ites
\ a
Sales/marketing Prod features “+
efforts
+ rae
Product ee
Attractiveness:
+ Increase In & ig a Fs
i Se ee ee
Total =| Potential > Pprospect Hot | Rate ro Rate
[Population| Prospects Prospects s> Purchasers 2>|A dopters
Prospect
Potential Prospect Hot prospect
ejay Prospect Loss Rate Gets Purchase
Tost ez Hal Loss Rate O88 Ral
Prospect
Lifetime
Lost Prospects
Word of 4R)
Mouth
sp Word of Market
mouth
Figure 2: High Level Overview of Model
New Venture Commercialization of Clean Energy Technologies
2.2 The Market
The market sector is composed of a series of stocks representing prospective customers at
various stages in the sales cycle. This structure is based on an extension of the Bass model
(Bass, 1969). However, rather than focusing only on the stocks of potential adopters and
adopters, the model developed here disaggregates the stock of potential adopters into more
specific stocks including potential prospects, prospects, hot prospects, and purchasers. “Potential
prospects” are firms that are capable of adopting the current version of the product that the
venture has chosen to apply sales effort to persuade them to leam more about the product.
“Prospects” are firms that are capable of adopting the product and have been made aware of the
product by the venture, and have not ruled out adopting it. “Hot prospects” are firms that have
expressed interest in adopting the product and are either actively trialing it or evaluating it in
some other fashion. “Purchasers” are firms that have purchased the product, but have not yet
started using it. “Adopters” are firms that have purchased and are actively using the product.
Also, there is a stock of “Lost Prospects” which are firms that were prospects, but then lost
interest in adopting the product or actively made the decision not to adopt.
The model also takes into account how the stock of potential prospects is replenished from the
total population (what we call “market growth”). Also, in addition to the influence of advertising
and word of mouth, the new model makes it possible to more clearly calibrate the influence of
factors such as price and product features that might make the product more attractive and drive
the adoption cycle. The “word of mouth” loop from the Bass model is still important, but the
significance of Bass’s “market saturation” loop may be lessened if the size of the total population
is large relative to the stock of potential prospects, and if there is a positive rate of market growth
to replenish potential prospects.
The rate of the flow from stage to stage of the prospect chain is a function of the number of
prospects at that stage, the average amount of time prospects remain at that stage, the amount of
sales effort expended, and the productivity of that sales effort:
(1) Advancement Rate = f(Prospects / Avg Prospect Lifetime, Potential Rate from Sales Effort)
(2) Potential Rate from Sales Effort = f(Sales Effort * Productivity of Sales Effort)
(3) Productivity of Sales Effort = f(Max Productivity of Sales Effort, Sales Experience,
Features/Competitor Features, Price/Competitor Price, Marketing Effort, Word of Mouth,
Customer Support)
Prospects that do not advance from one stage to the next within the average lifetime for that
stage become lost prospects. For details of the factors and equations that determine the rate of
prospects moving from each stage to the next, see Miller (2007).
2.3 The Competition
The competition sector of this model includes ways in which the firm’s competitors directly
affect the firm’s behavior and the “competitor” represents an aggregate of all competitors to the
firm. Because the firm under consideration here is a clean energy venture, it is assumed that the
competition is comprised primarily of conventional fossil-fuel-based energy firms such as a
utility selling electricity generated by coal-fired plants.
New Venture Commercialization of Clean Energy Technologies
The competition’s working capital is presumed to be unconstrained compared to the new
venture, and the competition’s costs, capabilities, etc., are exogenous to the model. The
endogenous parameters related to the competition are their prices and features. When the new
venture develops additional features, the competition may respond, usually after a delay, by
developing additional features themselves. Also, if the new venture’s prices are lower, the
competition may respond by lowering their prices. However the model is parameterized so that
the competition has limited ability to adjust their prices, based on the assumption that the
competition cannot control the price of fossil-fuel-based energy. Of course, if the competition is
able to improve their prices or features, the new venture may respond in kind, creating positive
loops of price and feature competition.
2.4 Novel Aspects of Model
The model is based on prior research on the dynamics of technology ventures and the adoption of
new technologies, and incorporates a number of novel attributes:
Market sector for clean energy technologies: General new product diffusion models work well
for goods being sold into a mass market, but do not fully represent the dynamics of adoption of
high value technology products into a conservative customer base. Based on interviews and case
studies of clean energy technology ventures, the “potential adopter” stock was disaggregated into
potential prospects, prospects, hot prospects and purchasers, each of which could be lost before
becoming an adopter. The time delays and most important factors for transition (e.g. price,
features, marketing, word of mouth, customer support) were identified for each stage of
adoption.
Product development sector including intellectual property: Technology ventures often
depend on their ownership of and ability to develop intellectual property that is not easily
appropriable by competitors. Since technology products usually contain both appropriable and
nonappropriable features, a product development sector was developed that takes into account
varying values and development resources needed for appropriable and nonappropriable features
for both the modeled venture and for the aggregate competitor.
Runway: New ventures are often constrained by working capital in ways that larger ventures are
not. New ventures usually do not have the ability to borrow money to cover expenses; a single
hire or layoff could make or break the firm. The simulation model reflects these ventures’ focus
on their runway - the amount of time they have before they run out of capital - and bases hiring
and layoff decisions on this parameter.
Effect of government policies on new clean energy ventures: Most venture simulation models
consider government policies to be outside the boundaries of the model. The model developed
here considered the effect various government policies related to clean energy technology would
affect the modeled venture, and includes parameters that allows one to adjust the existence and
effect of those policies. See below for more information about the policies included in the model
and the effects they have.
New Venture Commercialization of Clean Energy Technologies
3 Base Case Simulation and the Valley of Death
Table 1 presents business projections taken from the investor presentation for a clean energy
technology startup (and is typical for a business plan projection of the ventures examined for this
research). In each of the following scenarios, the “base case” venture is based on attributes of
this and the other startups that were studied for this research.
Year 1 2 3 4 5
Revenues $189 $4,126 $16,712 $32,106 $51,925
coGs $174 $3,535 $8,457 $9,311 $10,413
Gross Margin $15 $591 $8,255 $22,795 $41,512
Operating
Exp $2,324 $3,177 $6,496 $10,316 $14,508
EBITDA ($2,309) ($2,586) $1,759 $12,479 $27,004
Total Installs 6 69 235 435 713
Employees 7 16 31 46 63
($ amounts in 000’s)
Table 1: Business Plan
The base case venture is planning to sell a high value product (cost of over $100,000) into a
conservative market. We assume that the new venture starts out with a product that has better
features at lower cost than competitors, with the bulk of its feature advantage non-appropriable
(e.g. protected by patents). Furthermore, the new venture starts out with at least $3,000,000 of
investment capital, based on management's projections of how much capital is needed, and how
much the investment market is willing to provide this particular management team.
The venture starts with six employees, four focused on engineering and support, and two on sales
and marketing. The engineering-focused employees in the firm have above average experience
(having already developed the product), but the sales employees are at an experience
disadvantage, given that the product has never been sold before. However, the employees learn
and become more productive over time and in particular after working with customers by making
sales and installing their product. There are 100,000 firms that could conceivably adopt the new
product (total population), and initially 100 of them are reachable by the startup and would
consider the prospect of purchasing the new product (potential prospects). The CEO of this
typical firm strives to maintain at least a 25% feature advantage of their products over the
competition and attempts to maintain sufficient working capital to operate by instituting a hiring
freeze whenever they have less than twelve months of capital left at the current burn rate, and
laying off employees as necessary to maintain at least three months of working capital.
The venture whose projections are in Table 1 secured a $4M initial investment and an additional
$1.5M investment in Year 2 when the venture began running out of capital. Given these
investments, and the simplifying assumption that all revenues go directly to working capital in
the year they are recognized, and all working capital is retained, then Figure 3 shows a graph of
the working capital based on the projections in Table 1. Note that this graph looks distinctly like
a hockey stick. Indeed, if we remove delays in the sales cycle, triple the default capability of
10
New Venture Commercialization of Clean Energy Technologies
firms to adopt the technology (and therefore to become prospects), and assume that all engineers
are hired with the same experience as the founding engineers (assumptions in Table 2), then the
simulation model comes close to replicating the pro forma performance (See Figure 4).
$45,000
$40,000 Working Capital ?
$35,000 #
$30,000 Ji
$25,000 i,
$20,000
$15,000 y
$10,000
$5,000
$0 ——*
Year
Figure 3: Projected W orking Capital from Business Plan ($1,000s)
Proforma Working Capital
10M
5M
oto
0 12 24 36 48 60
Working Capital : ProForma
Figure 4: Working Capital from Model with Relaxed Assumptions
Avg Prospect Lifetime 0.1
Avg Hot Prospect Lifetime 0.1
Avg Purchaser Lifetime 0.1
Initial Capab of Firms to Adopt 0.15
Avg Experience of New Eng Hires | 10,000
Table 2: Assumptions Necessary to Replicate Business Plan Projections
11
New Venture Commercialization of Clean Energy Technologies
However, the experiences of the clean energy ventures interviewed for this research do not bear
that expectation out, and neither does the simulation model. Figure 5 shows the simulation
model results of the performance of the venture. The venture does achieve strong profitability,
but only after fifteen years. In the experience of the entrepreneurs and investors interviewed for
this research, most startup companies that have investors to pay back do not get nearly that many
years before they need to start showing results. Hence, the new venture in this example is likely
to fail.
Working Capital
40M
|
30M }
g }
2 20M |
& i
10M [
ofS Ly
0 244 «48 «=672)=— so 96'is120—'—s4d 168) 192) 216
Time (Month)
Working Capital: _BaseCase(3M Invest)
Figure 5: Performance of Venture over 18 Y ears
The “valley of death” refers to a period of time during which a startup company may not have
sufficient capital to grow and is not able to attract new investments, and appears over a wide
range of scenarios for clean energy technology companies. Figure 6 shows a sensitivity analysis
of working capital over the first seven years of the firm’s existence given a uniform distribution
of initial investments between $1M and $10M. Figure 7 shows a sensitivity analysis of working
capital over a uniform distribution of initial production costs and initial features from 50% less to
50% greater than the default values. Note that in all cases, the valley is evident, and lasts at least
four years.
12
New Venture Commercialization of Clean Energy Technologies
_Sens_ Init Invest
50% 75% (NIN 95% (I 100%
Working Capital
10M
8.999 M
7.999 M
6.998 M
5.998 M
4.998 M |
3.997 M
2.997 M
1.996 M
996,400
-4,000
18.2
Time (Month)
Figure 6: Sensitivity Analysis Over Range of Initial Investments
_Sens Price Features
50% 75% NI 95% I 100%
Working Capital
10.50 M
7.875 M
5.249 M
2.622 M
Time (Month)
Figure 7: Sensitivity Analysis over Initial Costs and Features
4 Emerging From the Valley
What is the difference between the state of the venture and its market between the points in time
when the venture starts its dip into the valley and when it leaves? Assuming no new sudden
infusion of capital or breakthrough in technology during that period, what changes allow the firm
to seemingly suddenly become very profitable and rapidly increase its working capital after so
many years of operation?
Table 3 shows the stocks that determine whether cash flow will be negative, neutral or positive.
When we set the initial values of these parameters to their month 180 values from the base case
simulation, the cash flow starts off positive (see Figure 8). This demonstrates that these
13
New Venture Commercialization of Clean Energy Technologies
parameters are sufficient to generate positive cash flow in the model. Further, sensitivity testing
shows that these parameters are necessary, since a significant reduction in the value of any of
these parameters from its Month 180 value results in negative cash flow for at least some period
of time. Table 3 summarizes what happens when any of these parameters are reduced by 50% at
Time 0 of their Month 180 value. The table shows the percent reduction in working capital at
one, three, and five years into the run.
Time (Month) 12 36 60
Accounts Receivable 37% 29% 12%
Engineers 9% 67% 33%
Avg Engineer Experience 65% 83% 45%
Avg Sales Experience 6% 2% 1%
Cumulative Purchases 12% 13% 7%
Potential Prospects 38% 64% 34%
Hot Prospects 44% 40% 21%
Purchasers 21% 14% 7%
Adopters 58% 87% 68%
Features [self,appropriable] 65% 19% -14%
Features [self, nonappropriable] 68% 44% 76%
Features [competitor nonappropriable] -18% 55% 56%
FUD [self,appropriable] 49% 20% 8%
FUD [self,nonappropriable] 66% 54% 42%
FUD [competitor,nonappropriable] -11% 32% 21%
Table 3: Percent Reduction in Working C apital from 50% Reduction in Parameter
Working Capital
24.90 M ] ]
18.68 M j
j 12.47M / |
6.254M A |
/| J
37,156 we | beet
0 24 48 72 96 120 144 168 192 216 240
Time (Month)
Working Capital: Month180 Values at MonthO ——;—3—3—3—3—4
Working Capital :BaseCase(3M Invest)
Figure 8: Performance of Firm with Month 180 Values Set at Time 0
14
New Venture Commercialization of Clean Energy Technologies
This analysis helps to inform us what factors determine when the venture will be profitable.
Working capital is critical to enable the firm to maintain its workforce and produce its products.
A full sales pipeline (potential prospects, prospects, hot prospects, etc.) is necessary for the
venture to close sales and generate revenue. The market for the venture’s products must be
growing to sustain growth of the venture and the venture’s customers must be paying reasonably
promptly for their purchases. The venture needs enough sales people and needs them to be
experienced and effective at selling the firm’s product. And the venture needs engineers who are
effective at maintaining the features of the product and keeping it ahead of competition. The
venture must be able to sell the product at an attractive price and still make a profit. And the
product must generate positive word of mouth in the market.
Unfortunately, a venture cannot start with a full pipeline, positive accounts receivable,
employees experienced in working with customers for the firm’s product, or with positive word
of mouth. The question then is how long it will take for a new venture to reach a sustainable
positive cash flow, and what must happen for this to be achieved.
5 Public Policy Factors
Even though a clean energy technology venture may do everything right, it still may have
difficulty succeeding if government policies discourage adoption. US government policies
currently provide substantial subsidies to the fossil fuel industries, creating substantial barriers to
the adoption of distributed generation and other clean energy technologies (California Energy
Commission, 2000; Lillis, Eynon, Flynn, & Prete, 1999; National Renewable Energy Laboratory,
2000). Coupled with a conservative customer base this presents an uphill battle for any clean
energy technology venture.
Many policies have been proposed to encourage the development and adoption of clean energy
technologies (Barringer & Revkin, 2007; Center for Clean Air Policy, 2006; Stavins, Jaffe, &
Schatzki, 2006; Stern, 2006), and these policies generally fall into three categories:
Carbon Policy: Most climate change or global warming legislation attempts to impose a cost to
the emissions of CO2 (the most common greenhouse gas). The Kyoto Protocol, legislation
recently passed by the state of California, the Northeastern and Mid-Atlantic states’ Regional
Greenhouse Gas Initiative (RGGI), and climate change legislation before the US Senate all
attempt to create CO2 emissions trading systems that would impose costs on companies emitting
COz. Other proposals have suggested simply placing a tax on the emission of CO2. Any of these
regulations would impose a cost on any fossil-fuel-based competition (or on not adopting the
new clean energy technology). For this reason, the model represents a carbon policy as an
increase in the costs of the competition. An increase in the competition's prices due to a carbon
policy enables the new venture to charge a higher price and extract higher profits while retaining
a price advantage, or to sell more easily at the original price.
Subsidy Policy: Another common type of policy is to subsidize the development or purchase of
clean energy technologies. For example, the federal government provides grants to cover a
portion of the research and development costs for some clean energy technologies. An example
is the Small Business Innovation Research Program (SBIR).! The result of this policy is to lower
' See http://www.science.doe.gov/sbir/ for information on SBIR grants for energy technology development
15
New Venture Commercialization of Clean Energy Technologies
the cost of providing the clean energy technology, enabling higher profits for the firm without
raising the price to the consumer.
Increasing adoption capability: The final group of policies either remove regulatory barriers or
provide regulatory incentives for the adoption of clean energy technologies. Examples of
regulatory barriers that can be removed are those that impose high additional costs on companies
that connect and utilize distributed generation. Regulatory incentives provide tax breaks for
companies that implement energy efficiency measures, or tax credits for the development of, for
example, wind farms. These policies increase the number of firms that are capable of adopting
clean energy technologies and therefore increase the rate at which the number of potential
prospects increases.
Figure 9 illustrates a comparison of three policies: a carbon policy, which causes competing
solutions to be 20% more expensive than the base case; a subsidy policy, which reduces
production costs for the new venture by 20%; and a policy that enables 5% more firms to
become capable of adopting the product. Implementation of any of the policies results in
significantly better performance than the base case.
Working Capital
730.83 M
584.67 M
@ 438.51 M
i—|
A 292.35 M
146.19 M - LA
37,156 man Zita]
0 24 48 #72 #96 120 144 168 192 216 240
Time (Month)
Working Capital: _IncrAdoptionCapab 05 + +
Working Capital: CarbonCost_20
Working Capital: Subsidy 20 —3 3 3 3 3 3
Working Capital: BaseCase(3M Invest) —4 + = ~ ; =
Hb
nA
nn
He
Figure 9: Effect of Policies
16
New Venture Commercialization of Clean Energy Technologies
6 Strategies to make a clean energy technology venture successful
We know from prior research, from the sources interviewed from this research, from direct
experience and from analysis of the model that the following three attributes are critical to
success for any new technology venture; management, market, and sustainable competitive
advantage. These factors are already well established in the literature (Eesley & Roberts, 2007;
Porter, 1985; Roberts, 1991; Utterback et al., 1992), and are briefly summarized here:
Right Management Team: Experienced investors state that the first and most important
attribute of any new venture are the talents, experience and attitudes of the management team.
Prior startup experience and sales experience are strongly correlated with success. The
importance of personal characteristics, such as persistence and flexibility in the face of adversity,
and the appropriate need for and use of personal power cannot be underestimated. It is
challenging for an analytical simulation model to reflect the impact of these personal
characteristics, but the model reflects in several ways that greater experience leads to greater
success.
Right Market: Another well established success factor for technology ventures is that the
venture is addressing a market need they understand well in a way that is a good match for the
size and capabilities of the venture, and that the target market has high growth potential. The
simulation model captures this by taking into account the sales and marketing effectiveness of
the venture, the growth potential of the market, and the nature of the sales cycle.
Sustainable C ompetitive Advantage: For a new technology venture to succeed, it needs to offer
a technology-based product that not only meets a market need, but also is different from and
better than competing alternatives at an attractive price. Further, the venture must be able to
sustain these advantages over time in the face of determined and resourceful competition and
establish a good reputation by word of mouth.
It is comforting that the simulation model captures well-known factors for the success of new
technology ventures. But, more importantly, what new insights does the model provide us about
clean energy technology ventures? The model offers some answers to important questions
regarding capital investments, the right mix of employees, product development goals, selling
versus leasing of the product, pricing in relation to competitors’ prices, and the significance of
government policy.
6.1 Capital Investments
Is it better to have a single initial investment of $3M, or three investments of $2M each, at 0, 12
and 24 months?
Investors typically prefer to stage investments over time. And, given this example, most
entrepreneurs would prefer the staged investments totaling $6M over two years to the single $3M
investment. Even if we assume a 20.5% discount rate based on the average long term
performance of early stage venture investments (Thomson Financial/National Venture Capital
Association, 2007), the staged $2M investments have a NPV of $4.18M, which is still
considerably higher than a $3M initial investment. It would take a very high 45% discount rate
for the two altemmatives above to be equivalent on a NPV basis.
At the earliest stages of a new venture, the value of the venture is minimal, and the entrepreneur
must sell the equity of the venture at a relatively low price in order to attract capital. The
17
New Venture Commercialization of Clean Energy Technologies
entrepreneur's need for capital is tempered by a desire not to “give away” too much of the
company. If the venture’s management believes it will be able to attract additional capital after a
year or two of operation, gaining experience, and establishing a presence in the market, the
venture might wait, and sell the equity at a higher price at that time. In this case, putting off
additional investments is preferable.
From the investor's perspective, the initial investment is very risky. The investor may be
intrigued enough by the technology and management team to “put a toe in the water” but will
likely want to keep the initial investment as small as possible. Only after the venture has proven
itself to at least some degree, will investors be more willing to invest additional capital.
It is therefore very common for technology ventures to receive a series of investments over time.
And most entrepreneurs would rationally choose to receive three $2M investments spaced over
two years rather than a single investment of $3M. However, if their firm behaved like the
prototypical clean energy technology venture simulated in the model, they would be wrong.
As can be seen in Figure 10, the model shows that a venture that would succeed with a $3M
initial investment would go bankrupt with three $2M investments spread over two years. The
venture goes bankrupt because it never has sufficient working capital and enough of a runway to
hire the engineers needed to keep the product better than the competition. More importantly, the
venture will never have the sales and marketing resources and experience needed to build up a
strong enough pipeline. A $3M initial investment provides enough working capital over the first
18 months to fund the product development and sales and marketing resources and develop the
experience needed to build up a pipeline that will enable the venture to survive and eventually to
thrive.
Working Capital
13.03 M
ans WEES eee eee
0 20 40 60 80 100 120 140 160 180 200
Time (Month)
Working Capital: 2M Invest_3 a a a a er es oe
Working Capital: BaseCase(3M Invest)
Figure 10: Comparison of $3M Investment vs. Three $2M Investments
Given that clean energy technology ventures take a long time to develop a market, and that labor
and production costs must be paid over that period, clean energy technology companies may
18
New Venture Commercialization of Clean Energy Technologies
require and justify a higher initial investment than other technology companies justify. For
example, software ventures usually have a product that can easily be trialed and adopted if the
customer finds its features and price attractive. These ventures usually do not need years to
develop a pipeline and revenue if they have a product demonstrably better than the competition.
Therefore, there is less risk that a delayed investment will irreparably damage a software venture.
Such a company is likely to perform better with three $2M investments rather than a single $3M
investment.
In contrast, biotech companies take a very long time to develop a market. For them, factors
critical to their success are based on the outcomes of product tests and the decisions of regulatory
agencies that are largely beyond the control of the sales force. A larger initial investment to
build up a sales force may not make the difference between success and failure, and investors are
well advised to reduce their risks by staging their investments.
However, investors who follow a staged investment strategy that is rational for early stage
software or biotech ventures may fail with the same strategy for clean energy technology
ventures. For the energy ventures, the market takes a long time to develop and development of
the market can be proportional to the early stage resources of the venture.
Given that clean energy technology ventures may require a risky larger initial investment, how
do investors decide which ventures are worth the risk? Investors would be well served to
consider the factors detailed in Section 4 that the model shows have the largest effect on the
fortunes of a clean energy technology venture and that may be evident at the start of the venture.
The more a venture can demonstrate that it has a nonappropriable technology that makes its
product attractive to customers, that a large number of prospective customers already exist, and
that its market will grow quickly over time given the resources to develop it, the more that
company may justify a relatively large initial investment. Given the size of the energy market, a
truly innovative energy company with many potential prospects has the potential to grow very
large, rewarding the investment made by the early investors.
6.2 Labor force composition
What is the best balance between the engineering and sales staff?
The base case clean energy technology venture starts with four engineers and two sales persons.
As is typical for technology startups, it is assumed that the engineers played a role in the
development of the product which is now ready for market, and that the sales persons are new to
the firm. Given that this is a technology venture, this would seem a reasonable ratio. The
engineers are needed to maintain the product and develop it further, and to support the early
customers. The sales people still need to learn the market before they become effective.
However, it turns out that this common ratio is suboptimal. If the venture were constrained to
six employees, it would do much better with four sales people and two engineers (See Figure
11). The most important task for the company once its product is ready is to develop a market
and fill the pipeline, and sales resources are needed for those tasks. Only later on, when
customers begin to adopt the product and competitors begin to catch up, are additional engineers
needed to shore up customer support and product development. But in the early stages, once the
venture has a product that is attractive to the market it should maximize sales and marketing
staff, and minimize engineering and product development staff if necessary to do so.
19
New Venture Commercialization of Clean Energy Technologies
Working Capital
486.34 M ra
g fo
= 243.18 M
fal
7
wae Lb
37,156 alae ie el
0 24 48 72 96 120 144 168 192 216 240
Time (Month)
Working Capital :_4Sales2Engineers +——t—1—_1—_1—_4+>—_ 4-1-4
Working Capital :_ 4Engineers2Sales
Figure 11: Sales vs. Engineering Focus
6.3 Product Development Goals
How much better than the competition should the venture strive for its products to be?
In the base case of the simulation model, the simulated venture desires its product features to be
25% more attractive than the competition. In reality, it is difficult to know exactly how much
more attractive a product is than the competition, since each customer will value the features of
the products differently. However, management must decide how much resources to allocate to
product development. An argument can be made that the venture should devote resources so that
its product is at least 50% better than the competition. After all, greater features do lead to more
sales, and many technology ventures focus on maximizing the features and functionality of their
products.
For the simulated venture, that approach would be wrong. In fact, that decision would bankrupt
the company. Conversely, if the venture de-emphasizes product development and only strives
for 10% more attractive features, the simulated venture will be much more successful (see Figure
12 for a comparison of results from striving for 10% better features, 25% better features, or 50%
better features). Naturally, these results depend on the assumption that a 10% differentiation is
sufficient to motivate sales for the product. Working in isolation, the product development staff
cannot know how many features are needed, and the bias is often to develop too much. The new
venture needs to work with current and potential customers to determine which features are
important and which are not. The optimum strategy is to develop only the features that
customers confirm will most differentiate the product.”
2 Note that this is in reference to the improvement of an existing product that customers do or can have experience
with and not the creation of a new product
20
New Venture Commercialization of Clean Energy Technologies
Working Capital
344.03 M
E 172.01 M
A
HK
er Ea
1079 [elebelel el leet ae
0 24 48 72 96 120 144 168 192 216 240
Time (Month)
Working Capital: _1.1xFeatures i—_i—_3,_ =, 2 ES
Working Capital: _1.25xFeatures
Working Capital: 1.5xFeatures —3 i+ 3—3
Figure 12: Comparison of Desired Features
6.4 Selling vs. Leasing
Should the venture prefer up front payments or recurring revenue?
The sample firm we are modeling charges the full price of their product up front, and also
charges a 20% annual maintenance fee as long as the customer remains an adopter. One might
wonder how the firm would fare if it adopted a leasing policy, charging little to nothing up front,
but receiving significantly higher recurring revenue per customer. Assume a very high lease rate
of 30% of the purchase price annually as long as the customer is using the product, in addition to
the 20% maintenance charge, and compare the following two scenarios:
(4) Base Case Revenues = New Adopters * Price + Existing Adopters * Price * 20%
(5) Leasing Revenues = Existing Adopters * Price * (30% + 20%)
Assuming they could find customers to accept this, most entrepreneurs would choose the leasing
model, which yields significant additional revenue per customer over time. Even with a 20%
discount rate over 10 years, the leasing model yields 25% more revenue on an NPV basis. With
a 10% discount rate over 20 years, the leasing model results in an NPV that is 63% higher. The
payback period is only a little over three years. But in the life of a new venture, those three years
are critical, and the choice of the leasing model would be wrong. Under the leasing scenario, the
simulated venture would go bankrupt.
Figure 13 graphs the line between success and failure for the base case venture based on the
percent of the product price paid up front and the percent paid annually as either a leasing or a
maintenance charge. Note that the firm will not succeed unless it charges both. The graph also
shows how much a customer that has a 10% cost of capital would be willing to pay for a lease in
addition to the 20% maintenance charge. Note that the regions of customer preference and
venture success only intersect at the default 100% up front price. For any reasonable cost of
21
New Venture Commercialization of Clean Energy Technologies
capital, the customer would not be willing to pay a high enough annual fee in exchange for a
reduction in the up-front cost to enable the venture to succeed. This is because the implicit
discount rate for the venture is extremely high. Up-front cash is much more valuable than future
payments.
Percent of Price Paid Annually
100%
90%
80% |
10% |
60% } |
50% veoh
40%
a=
_ pee ESSE
“0 see
10% Customer Preference
0%
Success
% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Percent of Price Paid Up Front
Figure 13: Up Front Payment vs. Annual Lease and Maintenance Payment
6.5 Pricing
What percent of the competition’s price should the venture charge?
We assume that an advantage that the new venture has is that it can lear to produce its product
at a lower cost over time, while competitors with much more mature technology have already
reached the end of their learning curve. Therefore, the new venture will have lower production
costs over time and can choose to sell its product at a lower price or to extract higher margins.
Given that lower prices drive additional sales, entrepreneurs often strive to charge as low a price
as possible. This is often a good strategy. In the base case simulation, the venture strives to
charge 25% less than the competition. The model results show that these lower prices result in
higher sales over the first years of the venture’s existence when we compare the base case
against a simulation in which the venture is charging the same price as the competition (Figure
14).
22
New Venture Commercialization of Clean Energy Technologies
Purchase Rate
0
0 2 4 6 8 10 12 14 16 18
Time (Month)
Purchase Rate :_ SamePrice a a a a
Purchase Rate :_BaseCase(3M Invest)
Figure 14: Comparison of purchase rate over 18 months
However, if we assume for the clean energy technology business that other factors (such as
features and word of mouth) play significant roles in a purchase decision, and that relatively low
quantities of the product are sold at relatively high prices and high margins, then the advantages
of a lower price diminish over time. Furthermore, a new venture is likely to lose a pricing war
against competitors with significantly greater resources and cash reserves if the competitors
choose to respond by lowering their prices. Therefore, the simulated venture performs best when
it charges the same price as competition and maximizes its margins.
Figure 15 shows a comparison over the 20 years of the simulation of purchase rate between the
base case, in which the venture charges 25% less than competition when its costs allow it to do
so, and the case in which the venture always charges the same price as the competition. Note
that the increased purchase rate from a lower price over the first years turns out to be temporary.
Counter to what might be expected, after about eight years the purchase rate in the case in which
the venture charges a higher price exceeds the lower price case. Naturally, working capital will
increase at a higher rate when the venture is selling more of its product at a higher price. In this
simulation, the additional resources (more sales persons and engineers) gained from the higher
margins outweigh the increased attractiveness from a lower price. Clean energy technology
entrepreneurs need to keep in mind when pricing their products that sometimes charging a higher
price will ultimately result in more customers.
23
New Venture Commercialization of Clean Energy Technologies
Purchase Rate
40 ot
Lap le
3
io}
ea
E he
0 ae oe grat
0 24 48 72 96 120 144 168 192 216 240
Time (Month)
Purchase Rate: SamePrice —-1—-t—-t—3—4—>$ — 3-443 4,
Purchase Rate: BaseCase(3M Invest) 22-2222 2-2
Figure 15: Charging same price as competitors vs. charging 25% less in base case
7 Effect of Government Policies
The preceding section examined the effect that various management strategies would have in
improving the performance of a clean energy technology venture. This section will explore the
effect of combining these management strategies with the goverment policies described in
Section 5.
Can a clean energy technology venture succeed without government policies in place to
support clean energy technologies?
The answer to this question is both yes and no. If we implement the above management
strategies in the simulation model in an optimal manner by reducing the desired feature ratio
from 1.25 to 1.1, increasing the initial sales force from two to 16, and increasing the target price
from 75% to 100%, the base case venture does significantly better. As shown in Figure 16, these
strategies enable the simulated venture to leave the valley of death sooner, and result in nearly
$1B of working capital by year 20, for an annual IRR on the initial $3M investment of over 33%.
By most measures, that would be considered highly successful.
24
New Venture Commercialization of Clean Energy Technologies
Working Capital
962.66 M
g
& 481.35M
A
pet
37,156 (COC ae zt |
0 24 48 72 96 120 144 168 192 216 240
Time (Month)
Working Capital: MgtPolicies --——1—3—4—_>—_ 3.9 4-4-3 4
Working Capital: BaseCase(3M Invest)
Figure 16: Results of Implementing Management Strategies
However, the venture still has a significant chance of failure during the four years before it
achieves a consistently positive cash flow and begins to rise out of the valley of death. Though
this firm will eventually be very successful, this is by no means obvious by year four. Investors
or entrepreneurs may become disenchanted after facing several years of losses with minimal
revenue, customers or working capital, and give up before realizing increasing profitability in
year five.
We consider the cumulative probability of the investors or entrepreneurs giving up on the
venture based on the accumulation over time of a hazard rate of failure. The hazard rate of
failure is the inverse of the expected life of the venture at any point in time and is a function of
the cash position of the venture, its features compared to the competition, the current number of
prospects compared to the initial prospects, and the length of time the firm has been in operation:
(6) Hazard Rate of Failure = f(1 / Current Ratio, Competitor Features / Features,
Initial Total Prospects / Total Prospects, Time)
(7) Current Ratio’ = f(-Cash Flow From Operations /
(Working Capital + AR Multiple * Accounts Receivable))
As any of the working capital, features or total prospects approach zero, the hazard rates from
these terms will approach infinity (ie. the expected lifetime of the firm will be very small).
Conversely, when cash flow is positive, or the features or prospects have favorable values, the
contribution of the corresponding term to the overall hazard rate will be negative (e.g. better than
normal prospects will increase the expected lifetime of the venture).
3 Tf the firm is bankrupt (Working Capital < 0), the Current Ratio is set to a very small number instead of this
equation, and therefore the Hazard Rate from Current Ratio will be very large, and bring the Cum Prob of Failure to
>=1 (i.e. the firm has failed)
25
New Venture Commercialization of Clean Energy Technologies
Figure 17 presents the cumulative probability of the investors or entrepreneurs giving up on the
venture in the base case or with the optimal management strategies.. Even with the management
strategies in place, the venture still has a significant probability of failure.
Cum Prob of Failure Based on Hazard Rate
2 2 i.
3
0.8 L aia
0
0 24 48 72 96 120 144 168 192 216 240
Time (Month)
Cum Prob of Failure Based on Hazard Rate :_MgtPolicies fe tf + =e +
Cum Prob of Failure Based on Hazard Rate :_BaseCase(3M Invest) —-2——2——2——2—_2.
Figure 17: Cumulative Probability of Failure with Management Strategies
The implementation of the government policies described and analyzed in Section 4 can change
this story. As shown in Figure 18, the venture accumulates twice the working capital and leaves
the valley of death much sooner in the presence of favorable policy than it might with the
management strategies alone. As can be seen in Figure 19, the presence of the government
policies reduces the probability of failure substantially.
26
New Venture Commercialization of Clean Energy Technologies
Working Capital
27.67 M ) 7
LAN. A j
37,156 Rea ee 7; ja
0 24 48 72 96 120 144 168 192 216 240
Time (Month)
Working Capital :_ MgmtStrat+GovtP olicies =
Working Capital :_MgtPolicies
Working Capital :_BaseCase(3M Invest) 3—_ 3-3-3333
Figure 18: Results of Government Policies in addition to Management Strategies
Hh
Hb
Hb
Hb
Hb
He
Cum Prob of Failure Based on Hazard Rate
0.2 4 |
0 24 «48 72 96 120 144 168 192 216 240
Time (Month)
Cum Prob of Failure Based on Hazard Rate :_ MgtPolicies+GovtPolicies —-t——t——_4—_4.——
Cum Prob of Failure Based on Hazard Rate :_ MgtPolicies
Cum Prob of Failure Based on Hazard Rate :_ BaseCase(3M Invest) ——-3——3——_3—_3-—
Figure 19: Probability of Failure with G ovt Policies in addition to M gmt Strategies
t t t t t i t t t
Po
Hh
He
27
New Venture Commercialization of Clean Energy Technologies
A clean energy venture with superior technology and the ideal management strategy can succeed
without government policies in place to support clean energy technology. However, the long
duration of the valley of death suggests a high risk of failure. Government policies reduce the
barriers to success and provide the venture a much higher chance of succeeding and achieving
wide adoption of clean energy technology.
Note, however, that though the combination of strategies and policies reduces the probability of
failure, they by no means assure success. As noted previously the model developed here is
meant to be used as a learning tool, and is not predictive. Though it is possible a real company
could do better than the simulated one, many factors are not taken into account in the model that
could cause a real venture to do worse, and to have a higher probability of failure. These factors
include:
e Macro economic factors, such as an economy-wide recession, or a slowdown in the
industry of the venture’s customers
e Energy market factors, such as a decrease in the price of fossil fuels or other alternative
energy technologies
e A new innovation that is more attractive than the venture’s technology
e New regulations that negatively impact the venture
e Stochastic disruptions in the acquisition of new prospects or customers that significantly
disrupt the firm’s revenue stream
¢ Personnel issues within the venture that cause management and/or employees to he less
effective (e.g., personality conflicts, health problems, etc.)
e Incompetence or theft on the part of management or employees
e Negative word of mouth (whether justified or not)
Clearly, the success of a new venture is never assured. However, the key lesson is that the
combination of the above management strategies and government policies may significantly
increase the odds of success (and the widespread adoption of the technologies) from what they
would have been otherwise. Given that policies make such a significant difference, governments
wishing for new clean energy technology ventures to succeed have a rationale to act. And it is in
the interest of the ventures themselves to exert as much influence as possible on governments to
promote the policies discussed (perhaps by forming industry lobbying groups).
8 Extensions
During the course of this research, quantitative data on over 1,000 clean energy technology-
related ventures was gathered, but the level of detail and quality of the data was too sparse for
much of it to be of use. Research is needed to determine the actual success and failure rates of
clean energy technology ventures based on a better sample of data. It would also be instructive
to gather detailed quantitative and qualitative attributes of these firms, and to establish statistical
correlations between the attributes and the level of success of the firms.
Also, the simulation model developed here could be enhanced in many ways.
e Policies meant to promote the adoption of clean energy technologies may spur additional
competition to the venture being modeled. Competition may expand the market, but may
28
New Venture Commercialization of Clean Energy Technologies
also make it more difficult for the venture to succeed. The model does not address this
interaction.
e Competition in the model could be disaggregated (in particular between fossil-fuel-based
competitors and other clean energy competitors).
e The model does not take into account factors and feedbacks that would limit exponential
growth of the new venture.
e The workforce in the model could be further disaggregated (with potentially separate
stocks for product development, customer support, sales, marketing, management,
administrative). Overtime, burnout and other important factors that shape the
effectiveness of the workforce could be modeled.
e The cash flow sector of the model could be expanded and improved to incorporate more
of the factors important to the balance sheet and income statement of a new venture.
e The existence and impact of equity and debt investments could be more explicitly
modeled.
e The modeling of the impact of policies could be expanded and improved to include other
policies that affect the venture, and to incorporate more of the resulting effects and
feedbacks from the implementation of these policies.
e The modeling of intellectual property (IP) development (non-appropriable features) could
be improved to more accurately reflect the value, costs and time delays inherent to the
development of IP
e The determination of desired sales effort could be improved to better reflect the hiring
decisions for sales and marketing personnel of actual firms.
For every sector of this model, more detail and additional feedback loops could be added and
new estimates could be made for values of the parameters (perhaps based on a more extensive
data set for clean energy technology ventures). However, it must be kept in mind that the model
cannot fully reflect reality. Any improvements should be made with the purpose of learning
about the performance and attributes of these ventures in general and not of predicting the future
for any one. Towards that end, we intend to turn the model into a management flight simulator
for educational purposes.
Though the simulation model here was developed based on data collected from and about new
clean energy ventures, it is quite possible that the lessons learned from analysis of the model can
be applied to other kinds of new ventures. In particular, when not taking the clean energy
policies into account, the model is very likely to apply to the commercialization of any new
energy technology. More generally, lessons from the model with the parameters described here
may apply to a new venture in any industry that faces conservative customers and long sales
cycles. Finally, with a different set of parameters, the model possibly can be used to explore the
commercialization and adoption of any new technology. However, it must be kept in mind that
the model is not meant to be predictive of any particular real company’s experience, and the
lessons learned from analysis of the model will only be as valuable as the parameterization of the
model enables it to be.
29
New Venture Commercialization of Clean Energy Technologies
9 A final word
As noted in the introduction, climate change is one of the most serious challenges of our time,
and the wide adoption of clean energy technologies is critical in order to address it.
Considerable focus has been appropriately devoted to the development of these technologies, to
improving their features, and to reducing their costs to make their wide adoption possible.
However, there must be commensurate focus on strategies and policies to enable the wide
adoption of the clean energy technologies once they are ready. History shows that the wide
adoption of disruptive new technologies is driven by new ventures. It is in the interest of society
to promote strategies and policies that will help clean energy technology ventures successfully
distribute their products and technologies.
Neither the private sector nor the public sector can address this problem alone. Private
investments coupled with optimal management of clean energy technology ventures may fail and
have failed without policies in place that address the impediments to the adoption of clean
energy technologies. The technologies already exist to address climate change, and
entrepreneurs and private investors are committing their resources to promote their adoption.
However, particularly in the U.S., policies must also be put in place to help enable wide
adoption. There is little time to waste.
30
New Venture Commercialization of Clean Energy Technologies
References
Barringer, F., & Revkin, A. C. (2007, January 18, 2007). Bills on Climate Move to Spotlight in
New Congress. New York Times.
Bass, F. M. (1969). A New Product Growth Model for Consumer Durables. Management
Science, 13(5), 215-227.
California Energy Commission. (2000). Distributed Generation: CEQA Review and Permit
Streamlining, (No. P700-00-19).
Center for Clean Air Policy. (2006 ). Cost Effective GHG Mitigation Measures for California,
Summary Report: An Independent Analysis of Measures to Reduce Greenhouse Gas
Emissions in 2010 and 2020 to Meet Executive Order S-3-05.
Eesley, C. E., & Roberts, E. B. (2007). Cutting Your Teeth: Learning from Rare (One or More)
Experiences (No. 4609-06): MIT Sloan Research Paper, Available at
http://ssrn.com/abstract=909615.
Hilmola, O.-P., Helob, P., & Ojalac, L. (2003). The value of product development lead time in
software startup. System Dynamics Review, 19(1), 75-82.
Joglekar, N., & Levesque, M. (2006). Marketing, R&D and Valuation of Startup Firms
Boston University School of Management Working Paper.
Lillis, K., Eynon, R., Flynn, E., & Prete, L. (1999). Federal Financial Interventions and
Subsidies in Energy Markets 1999: Primary Energy (No. SR/OIAF/1999-03): Energy
Information Administration.
Miller, D. S. (2007). New Venture Commercialization of Clean Energy Technologies. PhD
Dissertation, Massachusetts Institute of Technology, Cambridge, MA.
National Renewable Energy Laboratory. (2000). Making Connections: Case Studies of
Interconnection Barriers and Their Impact on Distributed Power Projects (No.
NREL/SR-200-28053): Department of Energy.
Oliva, R., Sterman, J. D., & Giese, M. (2003). Limits to growth in the new economy: exploring
the ‘get big fast’ strategy in e-commerce. System Dynamics Review, 19(2), 83-117.
Porter, M. E. (1985). Competitive Advantage: Creating and Sustaining Superior Performance.
New Y ork: Free Press.
Roberts, E. B. (1991). Entrepreneurs in High Technology. New Y ork: Oxford University Press.
Stavins, R. N., Jaffe, J., & Schatzki, T. (2006). Analyses of California Climate Change Policy:
Are They Too Good to Be True?
Sterman, J. D. (2000). Business Dynamics : Systems Thinking and Modeling for a Complex
World. Boston: Irwin/McGraw-Hill.
Stern, S. N. (2006). Stern Review on the economics of climate change, from http://www.hm-
treasury.gov.uk/independent_reviews/stern_review_economics_climate_change/sternrevi
ew_index.cfm
Thomson Financial/National Venture Capital A ssociation. (2007). Private Equity Performance
Remains Steady In Q3 2006 from
http://www.thomson.com/pdf/financial/news_release_pdfs/Q306_PE_ performance
Utterback, J., Meyer, M., Tuff, T., & Richardson, L. (1992). When Speeding Concepts to Market
Can Be a Mistake. INTERFACES, 22(4), 24-37.
31
