Journal of Business Venturing 19 (2004) 767 – 785

Planning for the market: business planning before marketing

and the continuation of organizing efforts
Scott Shanea,*, Frédéric Delmarb,1
a
Department of Economics, Weatherland School of Management, Case Western Reserve University,
11119 Bellflower Road, Cleveland, 44106 OH, USA
b
Stockholm School of Economics, Center for Entrepreneurship and Business Creation, P.O. Box 6501,
SE 11383 Stockholm, Sweden

Received 1 May 2003; received in revised form 1 October 2003; accepted 1 November 2003

Abstract

Drawing on goal setting theory, we argue that writing business plans before undertaking marketing
activities should enhance the continuation of venture-organizing efforts. We examine 223 new venture-
organizing efforts initiated in the first 9 months of 1998 by a random sample of Swedish entrepreneurs
and show that those organizing efforts in which entrepreneurs completed business plans before talking
to customers and beginning marketing or promotion had a lower hazard of termination than other
organizing efforts.
D 2003 Elsevier Inc. All rights reserved.

Keywords: Planning; Entrepreneurs; Marketing

1. Introduction

Should entrepreneurs complete business plans before engaging in marketing activities, like
talking to customers or initiating promotion efforts? Several authors have argued that writing
a business plan offers little value to entrepreneurs (Bhide, 2000). Specifically, prior

* Corresponding author. Tel.: +1-216-368-5538; fax.: +1-216-368-5039.

E-mail addresses: sas46@cwru.edu (S. Shane), frederic.delmar@hhs.se (F. Delmar).
1
Tel.: +46-8-736-9356; fax: +46-8-31-2785.

0883-9026/$ – see front matter D 2003 Elsevier Inc. All rights reserved.
doi:10.1016/j.jbusvent.2003.11.001
768 S. Shane, F. Delmar / Journal of Business Venturing 19 (2004) 767–785

researchers argue that writing a business plan (1) interferes with the efforts of time-
constrained entrepreneurs to undertake more valuable firm-organizing actions like initiating
marketing and promotion (Carter et al., 1996); (2) gives entrepreneurs a potentially harmful
illusion of control over information, such as that obtained from customers (Mintzberg, 1994;
Weick, 1979); and (3) leads to decision-making errors in estimating such things as customer
needs (Kahneman and Lovallo, 1993).
However, these arguments conflict with the basic principles of goal setting theory, which
hold that planning improves most subsequent human action (Locke and Latham, 1990).
Goal setting theory would suggest that writing business plans should enhance marketing
activities.
In this study, we empirically examine the effect of writing a business plan—a document
that summarizes how an entrepreneur will create an organization to exploit a business
opportunity (Stevenson and Van Slyke, 1985)—before undertaking marketing activities on
the termination of new ventures. We define a new venture as an effort by an entrepreneur
or team of entrepreneurs to create a new independent organization. We analyze a unique
data set capturing the life histories of 223 new ventures initiated by Swedish entrepreneurs
in the first 9 months of 1998 and followed over the subsequent 30 months. We control for
the effects of the entrepreneurs’ human capital, the new venture strategy, venture develop-
ment, and the industry. We show that the new ventures for which the entrepreneurs
complete business plans before talking to customers and beginning marketing or promotion
have a lower hazard of termination than other new ventures during their first 30 months of
life.
Several theoretical and practical issues suggest the importance of examining the relation-
ship between business planning and marketing activities on the termination of new ventures.
Initial contact with potential customers and the initiation of marketing and promotion efforts
are major milestones in new venture creation (Venkataraman, 1997). In addition, rapid
initiation of marketing and promotion and early contact with potential customers facilitate
the new product or service launch (Schoonhoven et al., 1990). Furthermore, faster
establishment of customer relationships facilitates access to capital (Venkataraman et al.,
1990). Finally, marketing activities are a necessary condition to the development of new
ventures.
This study provides two valuable contributions. First, we use goal setting theory to
explain why writing a business plan is a useful precursor to marketing activities. We then
test this explanation on a sample of new ventures. Our effort fills a void in the literature,
which currently offers no theoretical explanation (and consequently no empirical support)
for any explanation of why undertaking business planning before undertaking marketing-
related organizing actions should be beneficial. Second, we offset a recent action bias in the
entrepreneurship literature. Because business planning imposes an opportunity cost on the
time of entrepreneurs (Gifford, 1992), one school of thought holds such activities are not
worthwhile (Bhide, 2000; Allinson et al., 2000). By demonstrating that writing business
plans before taking marketing-related organizing action reduces the likelihood that these
efforts will be terminated, we hope to offset this bias and show the value of business
planning.
 S. Shane, F. Delmar / Journal of Business Venturing 19 (2004) 767–785 769

2. Theory development

One school of entrepreneurship researchers argues that undertaking business planning

before engaging in marketing activities will hinder the development of new ventures. First,
given time constraints, some researchers have argued that business planning precludes
entrepreneurs from engaging in marketing activities that demonstrate the reality of a new
business to customers and other stakeholders (Carter et al., 1996). Second, other researchers
have posited that the downside risk from incorrect marketing activities is too small to justify
prior planning. As Bhide (2000, p. 57) explains, ‘‘entrepreneurs do not have much to lose
from an erroneous forecast of. . .market size. . .. They usually do not put much capital at risk.’’
Third, other researchers have stated that entrepreneurs do not need to plan because their
intuition allows them to know how to effectively serve their chosen market without planning
first (Allinson et al., 2000). Fourth, still other researchers have proposed that uncertainty and
time constraints make planning ineffective in meeting market demand in new ventures (Bird,
1988).
However, this set of arguments conflicts with the basic principles of goal setting
theory, which suggests that undertaking business planning before undertaking marketing
activities will enhance the performance of new ventures. Mental processes guide human
action (Bandura, 1986, 1997; Locke and Latham, 1990; Mischel and Shoda, 1995;
Pervin, 1989). One of the most important of these mental processes is goal setting
because most human behavior involves forethought about desired future states and the
ways to achieve those goals (Bandura, 1986). As Locke and Latham (1990, p. 3) explain,
‘‘in purposeful action, it is the individual’s idea of and desire for the goal or end that
causes action.’’
Planning facilitates the integration of goals into people’s behavior (Bandura, 1997).
Planning helps people to form instrumental thoughts about a desired future (Nuttin, 1984).
Moreover, plans identify the necessary skills and information to achieve goals (Simon,
1997), as well as providing a framework for actions to lead to goal achievement (Miller et
al., 1960).
Plans are particularly useful when tasks are fuzzy or uncertain, and the decision maker
cannot rely on experience or habit (Campbell, 1988). Forming a plan mediates the relation-
ship between intention and actions because the plan specifies where and when one should act
to achieve the intended goal (Gollwitzer, 1999). When complex activities are undertaken
under uncertainty, people find it difficult to figure out how to best achieve their goals.
Planning helps people to understand the relationship between action and performance,
thereby mitigating misdirected effort (Campbell, 1988).
Although much planning occurs solely in mental form, goal setting theory also holds that
written planning improves human action. First, writing a plan clarifies goals and permits
people to set more specific objectives, which facilitate the achievement of those goals (Locke
and Latham, 1990; Rousseau, 1997). Second, writing a plan enhances the ability to learn
information necessary to act upon the plan (Langer and Applebee, 1987) by focusing the
decision maker’s attention on the most appropriate tasks (Locke and Latham, 1990). Third,
writing a plan allows a decision maker to better analyze complex activities in which many
770 S. Shane, F. Delmar / Journal of Business Venturing 19 (2004) 767–785

factors interact.2 Fourth, writing a plan helps a decision maker to communicate information to
others about how to pursue actions (Simon, 1997), an important dimension of achieving goals
in complex activity.
Despite the importance of planning to goal setting theory, researchers have not explored
the value of planning in new venture creation. In particular, they have not explored the
sequence between planning and more action-oriented venture-organizing activities. This
study examines the effect of the sequence between planning and marketing-related firm-
organizing actions on the termination of new ventures.
When creating a new venture, people can engage in marketing-related firm-organizing
actions, such as talking to customers, with or without first writing a business plan. The choice
to plan first imposes an opportunity cost on the entrepreneur’s time. Unlike the organizing
activities of initiating marketing or talking to customers, planning itself does not bring a new
organization into existence. As a result, the time spent planning limits the amount of time that
the entrepreneur can spend on firm-organizing actions.
The population of entrepreneurs varies in their approaches to the organizing process. Some
entrepreneurs complete business plans before engaging in marketing-related firm-organizing
activities, whereas others do not. This variation provides the basis of our test of undertaking
business planning first. We develop specific hypotheses about why completing a business
plan before undertaking two marketing activities—talking to customers and initiating
marketing and promotion—reduces the likelihood that a new venture will be terminated.3

3. Hypotheses

Writing a business plan prior to talking to customers reduces the likelihood that a new
venture will be terminated for four reasons. First, writing a business plan allows an
entrepreneur to improve his or her efforts to gather information from customers about their
needs. For example, by developing a business plan that specifies what type of product the
new firm will produce, the entrepreneur can better determine what questions about product
features would best meet customer needs.
Second, by writing a business plan, the entrepreneur can make better use of customer
feedback because the business plan provides a framework in which it can be used. For

2
Decision making about complex tasks is difficult because human beings lack the cognitive capacity to
mentally compare the potential values that result from analyses based on multiple variables (Simon, 1955). As a
result, they engage in a variety of decision-making biases. By allowing people to manage greater amounts of
information, to better connect that information, and to incorporate feedback into decision making, written planning
makes analysis possible in ways that are not possible otherwise (Tolchinsky and King, 1980). Moreover, writing a
plan reduces decision-making biases, such as framing effects (Miller and Fagley, 1991), by leading people to
justify the logic behind their decisions (Ablum-Heath and Di Vesta, 1986) and by allowing for broader
consideration of alternatives than decision making in situ.
3
Our argument does not imply that firm founders will foresee all the consequences of their planned goals nor
does it imply that they will achieve all goals that they set through planning. Rather, we only argue that the benefits
of planning before action will exceed the costs.
 S. Shane, F. Delmar / Journal of Business Venturing 19 (2004) 767–785 771

example, using a business plan to project hiring of new employees allows an entrepreneur to
incorporate feedback from customers into the overall venture model. This allows the
entrepreneur to obtain a more accurate understanding of the complex relationship between
hiring and customer feedback than would be the case without writing a plan.
Third, writing a business plan specifies the action steps to be taken to solicit feedback from
customers, thereby facilitating its acquisition. For example, writing a business plan helps
entrepreneurs to select the target customers to talk to and determine how and when they
should be contacted. For this reason, writing a business plan reduces the waste of time and
effort targeting inappropriate targets in inappropriate ways.
Fourth, writing a business plan facilitates the entrepreneur’s efforts to communicate
information about his or her business. Because the feedback gained from customers depends
on the assumptions made by the entrepreneur, the acquisition of accurate information from
customers is facilitated by accurate communication of the new venture’s strategy and product
offerings to potential customers. Written planning facilitates the entrepreneur’s understanding
of the logic of his or her strategy for venture creation and product offering, thereby enhancing
his or her ability to explain it to others. These arguments lead to the following hypothesis:

H1: The completion of a business plan prior to talking to customers will lower the
likelihood that the new venture will be terminated.

The completion of a business plan before beginning marketing and promotion will reduce
the likelihood that the new venture will be terminated for four reasons. First, writing a
business plan facilitates the process of gathering necessary information. For example, an
entrepreneur might consider different marketing efforts, such as the amount and type of
advertising to undertake, and writing a business plan identifies which approach to take
before initiating efforts. By writing a business plan that outlines the new venture’s business
model, the entrepreneur can choose a more appropriate approach to marketing and
promotion than would be the case if the entrepreneur attempted to make the choice without
a plan.
Second, writing a business plan improves the process of specifying the correct resources
(in nature and quantity) for marketing and promotion by integrating marketing information
with information about other aspects of the firm-organizing effort. For example, the
marketing plan might demand certain approaches to hiring sales staff or prototype develop-
ment, creating a complex dynamic relationship among marketing, hiring, and product
development. Because human beings have a hard time analyzing complex, dynamic
relationships in their heads, they make better estimates of these problems through written
analysis (Simon, 1997). Therefore, by developing a business plan prior to initiating
marketing and promotion, entrepreneurs can engage in marketing and promotion more
efficiently and effectively.
Third, writing a business plan specifies the action steps necessary to market the product or
service. For example, if the business plan indicates that the entrepreneur should first produce
a prototype for demonstration to customers before attempting to sell the product, the
entrepreneur can make his or her sales effort more efficient by waiting to undertake the
772 S. Shane, F. Delmar / Journal of Business Venturing 19 (2004) 767–785

demonstration of the product to the customer until after it has been produced. By helping an
entrepreneur to plan the marketing process, writing a business plan facilitates an entrepre-
neur’s effort to identify the correct action steps to market a new product or service, reducing
the waste of time and effort on inappropriate approaches to marketing.
Fourth, writing a business plan enhances communication about the venture. To create a
new venture, the entrepreneur needs to explain his or her vision to other people (Baum,
1994). Writing a business plan facilitates the transmission of an entrepreneur’s vision by
facilitating its understanding and by making its communication clearer and more specific
(Van de Ven, 1980). These arguments lead to the second hypothesis:

H2: The completion of a business plan prior to beginning marketing and promotion will
lower the likelihood that the new venture will be terminated.

4. Methods

4.1. Design and sample

A test of the effects of completing business plans before undertaking marketing-related

firm-organizing activities on the likelihood that a new venture will be terminated imposes
several difficult data collection requirements. First, because business plans and firm-
organizing activities are undertaken very early in the venture-organizing process (but not
instantaneously at the point at which opportunity exploitation begins), exploring the
relationship between the two variables requires the collection of data on new ventures
longitudinally from the initial point of opportunity exploitation forward in time. Second,
examining this research question requires the collection of real time data because
entrepreneurs rationalize the organizing process when asked about it retrospectively.
Third, archival sources cannot be used to generate a sample for such an investigation
because these sources do not record the existence of organizing efforts until new firms
have been legally registered or have begun production (Aldrich, 1999; Carroll and
Hannan, 2000) producing survivor bias (Aldrich and Wiedenmayer, 1993). Fourth, data
on new ventures must be obtained from initiation until termination or censoring to avoid
left-censoring, which will generate biased parameter estimates in event history analyses
(which are necessary to analyze these types of research questions) (Tuma and Hannan,
1984). Fifth, random sampling of a known population is necessary to generalize our
results to that population.
Overcoming these obstacles requires a large sample research effort in which researchers
first identify members of the general population who are creating a new venture at a particular
point in time. This approach allows for the identification of a sample of new ventures that
represents the population of all new ventures at the point at which an opportunity is first
exploited. Therefore, in the first 9 months of 1998, we first randomly sampled by telephone
35,971 Swedes between the ages of 16 and 70 and asked them to participate in our study. We
received agreement from 30,427 individuals (84.6%).
 S. Shane, F. Delmar / Journal of Business Venturing 19 (2004) 767–785 773

To determine which individuals were starting a new venture, we then asked the
respondents if they were creating a new business either alone or as part of a team. To
mitigate differences in the perception of the definition of a new business, we provided a
definition of a new business to the respondents. To be inclusive, this definition included a
variety of new businesses including farms, consultancies, and home-based businesses.
If the answer to the screening question was affirmative, we subsequently asked the
respondent if the effort to create the venture was on behalf of an existing organization. If the
answer to this question was negative, we then asked if the respondent was a member of the
venture team rather than a consultant or investor. As a result of this process, we found 453
people who were starting an independent business in 1998.
Because we were concerned that many of the respondents might have started the new venture
before we began our observation period, we also asked the respondents to tell us the month and
year when they first began work on the new venture. We defined starting work on the new
venture as any action taken in the pursuit of the opportunity that they had identified (e.g.,
gathering inputs). In this way, we differentiate between people who have merely thought about
starting a new venture and people who have taken action to create a new venture.
We limit our empirical analysis to the cohort of 223 new ventures that were first started
between January and September 1998, the period during which we conducted our initial
survey to avoid selection bias. Those individuals who had started new ventures in prior years,
but were still in the process of establishing a venture, are not representative of the new
ventures for the cohort of new ventures for those years. Other individuals who had also
initiated new ventures in those earlier years, but had completed the effort either positively or
negatively, would indicate they were not in the process of starting a business when surveyed
in 1998. As a result, our screening questions capture all of the new ventures initiated in 1998
but only ‘‘long-in-process’’ organizing efforts from earlier years. Because ‘‘long-in-process’’
organizing efforts would not represent the population of new ventures started in earlier years,
their inclusion would bias our sample and they are therefore excluded.
We follow these 223 new ventures over the next 2 1/2 years through the use of a biannual
telephone survey in which the entrepreneurs served as key informants.4 The response rate for
the successive survey waves was high: 90.5% at 6 months, 91.9% at 12 months, 98.5% at 18
months, and 96.1% at 24 months.
Our looks like the typical set of new ventures established in Sweden. The sample includes
software ventures, manufacturers, farms and other agricultural ventures, home-based ven-
tures, consultancies, and other business service ventures. However, the ventures can be easily
classified in a small number of categories. Roughly half (46.6%) pursued low-technology

4
Although we ask the respondents questions about both the independent and dependent variables, common
method variance is not a problem here for two reasons. First, respondents are asked the questions about the
independent variables at different points in time. As a result, the individual-level response tendency would have to
be stable over time for common method variance to bias estimators, a much less likely scenario than if a cross-
sectional single time point design is employed. Second, many of our independent and dependent variables
objective measures, which are much less likely to be affected by common method bias than continuous subjective
measures, such as Likert-type scales of attitudes.
774 S. Shane, F. Delmar / Journal of Business Venturing 19 (2004) 767–785

service opportunities; whereas 39% pursued high-technology services, such as software

design, and 14% were manufacturing efforts.
For approximately half of the sample (48%), at least the start-up team included someone
who had previously started a company; and for over three-quarters of the sample (76.2%), the
team included someone with at least 1 year’s experience in the new venture’s industry.
We observe the ventures during the period when most organizing activities take place.
Virtually none of the ventures (2%) completed their business plans during their first month of
life. Approximately 83% had no employees in the first month. Only one venture (0.4%)
received funds from an external source in the first month of life. Roughly 53% of the ventures
did not complete product development during their first month of life, while 65% of the
ventures did not obtain inputs during that period.

4.2. Analysis

We use event history models to analyze our data because these models allow us to treat
those new ventures that have not yet been terminated as censored in the month of last
observation, which is important to overcome a methodological problem that plagues research
on entrepreneurship. It is impossible to define a period of time that all observers will believe
is long enough to indicate that a new venture will not be terminated or can be considered to
have completed the organizing process. New ventures that are 1 month, 1 year, 10 years, or
100 years old all face a risk of termination and also may not have completed the organizing
process. Because the category ‘‘continuing’’ is always contingent and no time period can be
established to ensure that termination is no longer a risk, statistical techniques that account for
the censoring of observations need to be employed. Logistic regression or other techniques
that fail to take censoring into consideration generate biased results that depend on the length
of the observation period.
We examine the hazard of termination of the new ventures over time using piecewise
exponential hazard rate models with robust clustering on each new venture. Each month is a
spell in the life histories of the organizing efforts. We select the piecewise exponential
specification of the event history model because it allows the hazard rate to vary with age
(with the assumption that they are constant within each piece) without demanding parametric
assumptions. The age pieces we use are: less than 12 months, 12 to 24 months, and greater
than 24 months.5

4.2.1. The dependent variable

Our dependent variable is the termination of the new venture. We focus on this dependent
variable for two reasons. First, continuation of the organizing effort is a necessary condition
for all other activities in new ventures. A new venture can achieve no other performance goal
(achievement of first sale, positive profits, or the acquisition of financing) if it has been

5
We conducted supplementary analyses with other distributions to confirm that our results are not artifacts of
the distribution chosen. Our results are qualitatively the same with different distributions, suggesting that the
choice of distribution is not substantive in our case.
 S. Shane, F. Delmar / Journal of Business Venturing 19 (2004) 767–785 775

terminated. Therefore, explaining termination of new ventures is a necessary characteristic of

any theory of entrepreneurship. Second, no ‘‘positive’’ measure (e.g., achievement of first
sales, acquisition of financing, positive profitability) can provide an accurate test of the
continuation of the new venture because these milestones are not terminal events.6 Efforts to
define the continuation of new ventures by the achievement of positive milestones require
implicit assumptions as to the time horizon during which that milestone should be achieved.
Such assumptions are quite problematic, as certain milestones are never achieved during the
lives of many new ventures. For example, pharmaceutical firms frequently acquire new
biotechnology firms before the latter ever create new products. Moreover, new biotechnology
firms can take over a decade to bring a new product to market, with the achievement of that
milestone often occurring long after the new company has gone public.7
We define termination of the organizing effort as termination by all members of the venture
team because venture teams are often quite fluid, leading a venture to proceed with only part
of the group that initiated the effort. Therefore, defining termination solely by the decision of
one member of the team to terminate would confound termination of the venture with the
decision of the respondent to terminate participation in the new venture.
We identify the termination event by asking respondents at each survey wave whether
everyone pursuing the venture has terminated and if so in what month. If the new venture was
terminated by all members of the team pursuing it, we code the new venture with a ‘‘1’’ for
termination in the month identified by the respondent. Otherwise, we code this variable as
‘‘0.’’ Those organizing efforts not terminated at the end of 30 months were treated as
censored. We predict termination as a function of several time-varying and time-constant
covariates described below. By the end of 30 months, 82 of the new ventures in our sample
had been terminated.

4.2.2. The predictor covariates

To create our predictor variables, we ask respondents at each wave of the survey if each of
the three activities (completed a business plan, talked to customers, and initiated marketing
and promotion) had been undertaken. If they answered yes about an activity, we asked them
what month that activity was first undertaken. We coded each activity as ‘‘0’’ in each month
that the activity had not been undertaken and ‘‘1’’ in each month that the activity had been
undertaken. For example, of the 223 ventures, 13 had completed a business plan at 6 months,
56 had completed a plan at 12 months, 132 had completed a plan at 18 months, 143 had
completed a plan at 24 months, and 147 had completed a plan at 30 months.
We then constructed two dummy variables ( 1, 0, and + 1) to indicate in each month
whether the venture had a completed business plan but had not undertaken the other activity

6
While a new venture organizing effort that has achieved first sales can be terminated subsequently, a
terminated venture organizing effort cannot achieve first sales subsequent to termination.
7
Readers should note that the ‘‘not terminated’’ category includes significant variation in the status of the
venture organizing effort, including such things as still trying, in hiatus, or succeeded in getting a firm established
as defined by the founder’s cognition, achievement of first sales, or entering the effort into a formal business
registry. What this category has in common is that the venture organizing effort has not been terminated.
776 S. Shane, F. Delmar / Journal of Business Venturing 19 (2004) 767–785

( + 1), had neither completed the business plan nor undertaken the other activity (or had
undertaken both activities) (0), or had undertaken the other activity but not completed a
business plan ( 1). We do this by subtracting the monthly variable for each of the organizing
action variables from the completed business plan variable.
Completed plan before talking to customers. This variable took a value of + 1 for each
month that the new venture had a completed business plan but had not yet talked to
customers; a value of 0 for each month that the venture either had a completed business plan
and had talked to customers or had neither a completed business plan nor had talked to
customers; and a value of 1 for each month that the venture had talked to customers but did
not have a completed business plan.
Completed plan before initiating marketing and promotion. This variable took a value of
+ 1 for each month that the venture had a completed business plan but had not yet initiated
marketing or promotion; a value of 0 for each month that the venture either had a completed
business plan and had initiated marketing and promotion or had neither a completed business
plan nor had initiated marketing and promotion; and a value of 1 for each month that the
venture had initiated marketing and promotion but did not have a completed business plan.

4.2.3. Control covariates

4.2.3.1. New venture development. We control for three dimensions of new venture
development: number of activities undertaken, number of employees, and receipt of external
capital.
Number of activities undertaken. Commitment theorists (e.g., Salancik, 1977) believe
that the more activity a person takes toward a course of action, the less likely that they will be
to abandon that course of action. This argument suggests that the more firm-organizing
activity an entrepreneur engages in, the less likely the termination of a new venture will be
(Aldrich, 1999; Reynolds and White, 1997). We measure the level of firm-organizing activity
with a time-varying count of firm-organizing activities adapted from a list of firm-organizing
activities that prior researchers (Carter et al., 1996) have found to be undertaken by many
entrepreneurs. The activities include obtaining required tax documents to create a new
venture in Sweden; registering the venture with government authorities (a requirement for
starting a business in Sweden); asking for funds from financial institutions or other people;
obtained inputs and raw materials, seeking intellectual property protection, such as a patent,
copyright or trademark; seeking licenses or permits for the venture; and initiated sales of the
product or service. Each of these activities is updated monthly by asking the respondents at
each survey to indicate if these activities had been undertaken and if so in what month. The
variable ranges from a minimum of 0 to a maximum of 7.
Number of employees. Larger organizing efforts are less likely to fail than smaller
organizing efforts (Aldrich and Auster, 1990; Bruderl and Schussler, 1990; Carroll and
Hannan, 2000; Hannan and Freeman, 1989). Therefore, we include a time-varying count of the
number of employees, measuring a part-time employee as one half of a full-time employee.
Received external capital. Arguing that new ventures face capital constraints, Holtz-
Eakin et al. (1994) and Taylor (2001) found that the receipt of external capital reduced the
 S. Shane, F. Delmar / Journal of Business Venturing 19 (2004) 767–785 777

hazard of venture termination. We control for the receipt of external capital with a time
varying dummy variable indicating each month after which the venture has received external
capital.

4.2.3.2. Human capital. We control for three dimensions of human capital: the founders’
start-up experience, founders’ industry experience, and venture team size.
Founders’ start-up experience. Entrepreneurs who have started more companies pre-
viously are less likely to terminate their new ventures (Bruderl et al., 1992) because they have
greater knowledge of the firm-organizing process. We measure prior start-up experience as a
time-invariant count of the number of prior firms founded across the venture team.
Founders’ industry experience. Entrepreneurs with more industry experience are less
likely to terminate their new ventures (Bates, 1990; Schoonhoven et al., 1990) because they
have greater knowledge of the key characteristics of their industry. We measure industry
experience as a time-invariant count of the number of years of experience in the new
venture’s industry across the team venture team.
Venture team size. New ventures undertaken by larger teams are more likely to be
terminated (Bruderl and Preisendorfer, 1998). We measure venture team size as a time-
invariant count of the number of members of the initial venture team.

4.2.3.3. Venture strategy. We control for five dimensions of the venture opportunity and
strategy: whether the venture involved the purchase of an ongoing firm, and the importance of
having more attractive products, better price, serving those missed by others, and a good
location.
Purchased venture. New ventures initiated through the purchase of a business should be
less likely to be terminated than greenfield ventures (Gimeno et al., 1999). We use a time-
invariant dummy variable of one to indicate the purchase of an existing venture.
Attractive products. New ventures whose founders perceive they have more attractive
products or services than other ventures should be less likely to be terminated (Shepherd et
al., 2000). We measure the entrepreneur’s perception that it is important to have more
attractive products and services than its competition by employing a time varying scale in
which ‘‘1’’ equals ‘‘insignificant,’’ ‘‘2’’ equals ‘‘marginal,’’ ‘‘3’’ equals ‘‘important,’’ and ‘‘4’’
equals ‘‘critical.’’
Price competition. New ventures that compete on price are more likely than other new
ventures to be terminated (Reynolds and White, 1997). We measure the founders’ perception
of price competitiveness with a time invariant scale in which ‘‘0’’ equals ‘‘no importance,’’
‘‘1’’ equals ‘‘marginal importance,’’ ‘‘2’’ equals ‘‘moderate importance,’’ and ‘‘3’’ equals
‘‘critical importance.’’
Serving those missed by others. New ventures that exploit opportunities missed by other
firms will be less likely to be terminated than other new ventures (Reynolds and White,
1997). We measure this variable with the founders’ time invariant response to the question
‘‘how important is serving those missed by others to be an effective competitor?’’ Responses
were ‘‘1’’ equals ‘‘insignificant,’’ ‘‘2’’ equals ‘‘marginal importance,’’ ‘‘3’’ equals ‘‘moderate
importance,’’ and ‘‘4’’ equals ‘‘critical importance.’’
778 S. Shane, F. Delmar / Journal of Business Venturing 19 (2004) 767–785

Superior location. New ventures that obtain superior locations will be less likely to be
terminated than other new ventures (Reynolds and White, 1997). We measure this variable
with the founders’ time invariant response to the question ‘‘how important is having a
superior location to be an effective competitor?’’ Responses were ‘‘1’’ equals ‘‘insignificant,’’
‘‘2’’ equals ‘‘marginal importance,’’ ‘‘3’’ equals ‘‘moderate importance,’’ and ‘‘4’’ equals
‘‘critical importance.’’

4.2.3.4. Industry. The new ventures in our sample fall into three primary categories: low-
technology service firms, high-technology service firms, and manufacturing firms, leading us
to include two industry dummy variables. The first dummy, high-technology service, takes a
value of 1 if the new venture is a high-technology service venture; otherwise it takes the value
of 0. The second dummy, low-technology service, takes a value of 1 if the new venture is a
low-technology service venture; otherwise it takes a value of 0. The omitted category is
manufacturing.
We also control specifically for four dimensions of industry: number of firms, average firm
age, market size, and exit rate.
Number of firms. Industry concentration is a barrier to the entry that might lead
entrepreneurs to terminate their new ventures (Geroski, 1995). Therefore, new ventures are
less likely to be terminated in industries composed of more firms (Acs and Audretsch, 1990).
We control for the number of firms in the industry with data from Statistics Sweden8 that
measures the number of new firms in the new venture’s five-digit Standard Industrial Code in
that year.
Average firm age. Industry maturity is a barrier to the entry of new firms that might lead
entrepreneurs to terminate their new ventures (Utterback, 1994). Therefore, new ventures are
less likely to be terminated when the average age of firms in an industry is younger (Carroll
and Hannan, 2000). We control for average firm age in the industry with data from Statistics
Sweden that measures the average age of firms in the new venture’s five-digit Standard
Industrial Code in that year.
Market size. Market size attracts entrepreneurial entry (Geroski, 1995). Therefore, new
ventures are less likely to be terminated if they take place in industries with higher levels of
sales (Romanelli, 1989). We control for the level of sales in the industry with data from
Statistics Sweden that measures the revenues of firms in the new venture’s five-digit Standard
Industrial Code in that year.
Exit rate. High firm exit rates indicate an industry that has reached its carrying capacity
(Carroll and Hannan, 2000). Therefore, new ventures are less likely to be terminated if they
take place in industries with lower exit rates (Aldrich, 1999). We control for the exit rate of
the venture’s industry with data from Statistics Sweden that measures the percentage of firms
in the new venture’s five-digit Standard Industrial Code that exited in that year.

8
Statistics’ Sweden’s business register includes all firms operating in the legal economy and is updated every
2 weeks with information from the Swedish tax authorities. We create the industry variables by dividing the
business register data into five digit standard industrial codes and updating the measures annually using data for
1998, 1999, and 2000.
 S. Shane, F. Delmar / Journal of Business Venturing 19 (2004) 767–785
Table 1
Descriptive statistics and the correlation matrix

Mean S.D. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
< 12 months 0.48 0.50 1.00
12 – 24 months 0.36 0.50 .71* 1.00
Industry exp. 16.88 22.76 .04* .02 1.00
Start exp. 2.61 11.79 .02 .01 .10* 1.00
Team size 1.98 1.16 .02 .01 .54* .00 1.00
Purchase 0.10 0.31 .03* .02 .12* .04* .01 1.00
Serve missed 2.59 0.76 .01 .01 .03* .04* .03* .02 1.00
Price comp. 2.33 0.75 .03* .02 .07* .05* .05* .02 .15* 1.00
Attractiveness 2.48 0.95 .04* .02 .03* .01 .09* .01 .21* .11* 1.00
Location 2.10 0.94 .01 .00 .08* .08* .09* .11* .21* .09* .15* 1.00
Average age 6.57 2.60 .03* .00 .09* .05* .01 .16* .07* .03 .08* .01 1.00
Exit rate 1.21 5.31 .13* .09* .03* .00 .09* .06* .08* .01 .05* .03* .34* 1.00
Market (mil) 160.08 213.75 .04* .01 .16* .04* .01 .03 .02 .13* .01 .10* .01 .12* 1.00
Firms (000) 1.87 2.22 .02 .00 .03 .02 .08* .03* .00 .12* .01 .06* .14* .10* .65* 1.00
Activities 3.36 1.91 .43* .27* .16* .01 .12* .14* .01 .09* .02 .04 .09* .08* .10* .02 1.00
Employees 1.15 2.92 .06* .05* .02 .00 .04* .02 .02 .05* .08* .03 .02 .01 .06* .02 .12* 1.00
Finance 0.20 0.40 .28* .17* .06* .02 .00 .17* .00 .09* .07* .06* .00 .02 .06* .06* .48* .13* 1.00
Plan/customer 0.19 0.61 .02 .00 .05* .03 .05 .01 .08* .00 .14* .10* .03 .04 .03 .02 .11* .03 .04 1.00
Plan/marketing 0.30 0.56 .15* .10* .03 .06* .00 .09* .07* .05* .01 .02 .02 .01 .01 .07* .08* .07* .05* .50* 1.00
High-tech. ser. 0.39 0.49 .01 .01 .10 .08* .12* .22* .00 .05* .03* .16* .11* .13* .17* .17* .09* .04* .02 .04* .01 1.00
Low-tech. ser. 0.47 0.50 .02 .01 .02 .07* .05* .23* .03* .11* .03* .14* .00 .05* .19* .12* .13* .05* .10* .04* .05* .76* 1.00
*P < .05.

779
780 S. Shane, F. Delmar / Journal of Business Venturing 19 (2004) 767–785

5. Results

Table 1 presents the descriptive statistics. Table 2 shows the event history models to predict
the likelihood of termination as a function of order that completing a business plan and

Table 2
Piecewise exponential models to predict venture failure
Variable Model 1 Model 2 Model 3 Model 4
hazard (S.E) hazard (S.E) hazard (S.E) hazard (S.E)
Age pieces
Less than 12 months 2.36 (1.11)# 2.41 (1.14)# 1.99 (0.92) 2.13 (0.99)#
12 – 24 months 1.37 (0.70) 1.34 (0.69) 1.31 (0.67) 1.30 (0.67)

Firm development
Activities undertaken 0.73 (0.06)**** 0.74 (0.06)**** 0.72 (0.06)**** 0.72 (0.06)****
Number of employees 0.47 (0.26) 0.50 (0.24) 0.47 (0.23) 0.49 (0.23)
Received external finance 1.29 (0.59) 1.18 (0.55) 1.30 (0.57) 1.22 (0.55)

Human capital
Start-up experience 0.77 (0.08)** 0.76 (0.08)** 0.76 (0.07)** 0.76 (0.07)**
Industry experience 1.01 (0.01) 1.00 (0.01) 1.01 (0.01) 1.00 (0.01)
Venture team size 0.99 (0.10) 1.02 (0.11) 1.03 (0.10) 1.04 (0.11)

Venture strategy
Purchased firm 0.22 (0.16)* 0.22 (0.18)# 0.25 (0.19)# 0.24 (0.19)#
Missed by others 0.81 (0.11)# 0.75 (0.10)* 0.78 (0.10)* 0.75 (0.10)*
Price competition 1.23 (0.19) 1.25 (0.19) 1.19 (0.18) 1.23 (0.19)
Attractive products 1.05 (0.12) 1.11 (0.13) 1.03 (0.12) 1.07 (0.13)
Superior location 1.06 (0.14) 1.10 (0.15) 1.10 (0.14) 1.12 (0.15)

Industry
Average firm age 1.06 (0.05) 1.06 (0.05) 1.06 (0.05) 1.06 (0.05)
Number of firms 0.99 (0.00) 0.99 (0.00) 0.99 (0.00) 0.99 (0.00)
Industry sales 1.00 (0.00) 1.00 (0.00) 1.00 (0.00) 1.00 (0.00)
Exit rate 1.01 (0.01) 1.01 (0.01) 1.01 (0.01) 1.01 (0.01)
High-technology service 1.08 (0.35) 1.14 (0.38) 1.10 (0.37) 1.14 (0.38)
Low-technology service 1.17 (0.37) 1.29 (0.43) 1.23 (0.39) 1.31 (0.43)

Order of activities
Plan before customers 0.44 (0.10)**** 0.54 (0.13)**
Plan before marketing 0.46 (0.11)*** 0.59 (0.15)*
Log likelihood 206.45 199.30 200.20 196.98
v2 78.11**** 93.60**** 89.58**** 93.26****
The analysis file contains 5093 firm-month observations, 223 cases, and 82 terminations.
*P < .05.
**P < .01.
***P < .001.
****P < .0001.
#
P < .10 (two-tailed tests.)
 S. Shane, F. Delmar / Journal of Business Venturing 19 (2004) 767–785 781

undertaking marketing activities takes place. Model 1 shows the base model in which the
control variables alone are included. Model 2 shows the effect of completing a business plan
before initiating marketing and promotion. Model 3 shows the effect of completing a business
plan before talking to customers. Model 4 shows the effect of completing a business plan
before initiating marketing and promotion and completing a business plan before talking to
customers.
We describe the results from the full model (Model 4). Overall, the model is significantly
different from zero (v2 = 93.26, P < .0001). Several of the control variables predict termination
in ways consistent with previous research. Each prior start-up founded by the venture team
reduces the hazard of termination by 24%. In addition, firms purchased from others are 76%
less likely to be terminated than greenfield new ventures. Furthermore, each scale score of
importance of serving customers missed by others reduces the hazard of termination by 25%.
Finally, each additional venture-organizing activity undertaken by entrepreneurs reduces the
hazard of termination by 28%.
The results also support both hypotheses. Consistent with Hypothesis 1, completing a
business plan before talking to customers reduces the hazard of termination by 46%.
Consistent with Hypothesis 2, completing a business plan before initiating marketing and
promotion reduces the hazard of termination by 41%.
One criticism of our results is that they confound terminations for personal and
organizational reasons. This distinction is important because termination for organizational
reasons provides a stronger test of our hypotheses than overall termination because it is an
indication of a failed organizing effort. For this reason, we reanalyzed our data to predict
separately the termination of the 45 ventures that did so for ‘‘organizational’’ reasons. We
conducted this analysis by asking the respondent to explain the major reason for terminating
the venture. Examples of reasons given were ‘‘lost interest,’’ ‘‘got another job,’’ ‘‘did not
have the time,’’ ‘‘unable to fund,’’ ‘‘team conflict,’’ ‘‘competition too strong,’’ or ‘‘not a
good idea.’’ We coded the reasons for termination into two categories: entrepreneur-specific
personal reasons (e.g., ‘‘got another job’’) and venture-organizing-specific reasons (e.g., ‘‘not
a good idea’’). We then reran our analysis on the subset that provided organizational reasons
for termination. Our results are qualitatively the same for this regression as for the overall
model.

6. Discussion

We argued that completing a business plan before initiating marketing activities would
reduce the hazard of termination of new ventures. We analyzed a unique data set capturing the
life histories of 223 new ventures initiated by Swedish entrepreneurs between January and
September 1998 and followed over the subsequent 30 months. Controlling for a variety of
factors that capture the entrepreneurs’ human capital, venture strategy, the development of the
venture, and the venture’s industry, we show that new ventures are less likely to be terminated
if the entrepreneurs complete business plans before initiating marketing and promotion and
before talking to customers.
782 S. Shane, F. Delmar / Journal of Business Venturing 19 (2004) 767–785

Our approach enhances our confidence in our findings. Our sample accurately represents
the population of Swedish new ventures initiated between January and September 1998,
allowing us to generalize to the broader population of new ventures in Sweden. In addition,
the sample faces no selection bias because all of the new ventures are observed from the point
at which work on them was initiated until they were terminated or were censored after 30
months. Finally, our data do not suffer from hindsight and recall bias because they examine
planning and firm-organizing actions over the period during which they are actually taking
place.

6.1. Theoretical implications

Our results provide implications for several strands of entrepreneurship research. First, this
study provides empirical support for the school of entrepreneurship research that argues that
explaining what entrepreneurs do is a useful area of investigation (Aldrich, 1999; Carter et al.,
1996; Katz and Gartner, 1988). Theorists have proposed that the evolution of new ventures
will be influenced by the activities that entrepreneurs undertake during the organizing process
(Aldrich, 1999). As Carter et al. (1996, p. 163) argue, ‘‘What entrepreneurs do in their day-to-
day activities matters. The kind of activities that nascent entrepreneurs undertake. . .and the
sequence of these activities have a significant influence on the ability of nascent entrepreneurs
to successfully create new ventures.’’ Our results are consistent with this perspective.
Second, our results suggest that the order in which firm-organizing activities are under-
taken also matters to the evolution of new ventures. New ventures do not come into existence
instantaneously in the form of established organizations (Gartner, 1985). Rather, they are
created through organizing processes that take place over time after the entrepreneur has
initiated efforts to exploit an opportunity. Some researchers have argued that the order in
which these different organizing activities are undertaken does not influence the probability
that the venture will continue (Carroll and Hannan, 2000; Hannan and Freeman, 1989). In
contrast to this argument, our results show that the hazard of termination of new ventures is
influenced by the order in which planning and marketing activities take place.9
Third, our results indicate that the action bias of some of the entrepreneurship literature is
inconsistent with empirical evidence about new venture formation. The predictions of goal
setting theory (Locke and Latham, 1990) hold with respect to the value of planning to the new
venture-organizing process. Completing business plans before undertaking marketing activ-
ities reduces the hazard of termination of new ventures. This result is important because some
entrepreneurship scholars have argued that entrepreneurs are better off taking action without
planning first. For instance, Carter et al. (1996, p. 154) explain, ‘‘Behavior such as buying
facilities and equipment might be a more significant indicator to others that a nascent business
is real than undertaking a behavior such as planning. Buying facilities may show others that

9
We suspect that the benefits of the order of firm-organizing processes are not limited to the relationship
between planning and marketing activities but is also affected by the order of other processes. We believe that
future researchers should empirically test Hannan and Freeman’s (1989) assertion that the order of firm organizing
subprocesses does not matter.
 S. Shane, F. Delmar / Journal of Business Venturing 19 (2004) 767–785 783

the entrepreneur has made a significant commitment to creating a new business compared to
what might be a less public demonstration of commitment like planning.’’ Our results suggest
that planning is an important precursor to action and that entrepreneurship researchers would
do well to consider this relationship in theory development.

6.2. Methodological implications

Our study also provides useful methodological contributions to research on new ventures.
We show how researchers can examine the effects of the order of venture-organizing
activities on the hazard of termination of organizing efforts by comparing the timing at
which different activities are undertaken. We believe that the use of an ordering methodology
will help to examine more subtle questions about the evolution of new ventures than is the
case if researchers look at only the presence or absence of characteristics. In particular, we
believe that the examination of the order and timing of activities is particularly useful in
explaining the effects of those processes that do not take place instantaneously when the
organizing effort is first initiated but instead are undertaken over time.
Finally, we believe that our study provides an example of how researchers can examine
venture organizing in ways that overcome the problems of existing research on new ventures.
By looking at the process of new venture creation longitudinally, we show how researchers
can avoid having to assume (unrealistically) that static characteristics present at the time of
sampling explain the formation process. We also show how the collection of data in real time
overcomes the hindsight and recall bias present in retrospective studies of new venture
formation.

6.3. Implications for practitioners

The study provides practical implications for people interested in creating new ventures.
Our results demonstrate that entrepreneurs should complete business plans before talking to
customers or initiating marketing and promotion. Although many entrepreneurs may want to
move directly to marketing activities upon the identification of an opportunity, our results
show that there is a better approach. By developing a business plan first, an entrepreneur will
reduce the likelihood of termination of his or her new venture.

6.4. Limitations

Our study has an important limitation. We measure only the ordering of marketing-related
firm-organizing activities. We have no information about the quality of the completed
business plans or marketing-related firm-organizing actions. This limitation was necessitated
by the complexity of measuring organizing activities longitudinally and the absence of prior
evidence on this question. Nevertheless, the evidence presented in this study suggests the
importance of considering the concept of sequencing. Because both the order and the quality
of business plans and organizing actions likely influence the termination of new ventures,
future research should examine both order and quality in the same study.
784 S. Shane, F. Delmar / Journal of Business Venturing 19 (2004) 767–785

Acknowledgements

Both authors contributed equally to the writing of this paper and are listed reverse
alphabetically. The research design owes an intellectual debt to the Panel Study of Business
Start-ups undertaken by the Entrepreneurial Research Consortium, a temporary voluntary
association of 30+ U.S. and non-U.S. universities. The study was financed by the Knut and
Alice Wallenberg Foundation, the Swedish Foundation For Small Business Research, and the
Swedish National Board for Industrial and Technical Development. We would like to thank
Dan Cable, Riitta Katila, and Atul Nerkar for helpful comments on an earlier draft of this
paper.

References

Ablum-Heath, M., Di Vesta, F., 1986. The effect of conscious controlled verbalization of a cognitive strategy on
transfer in problem solving. Mem. Cogn. 14, 281 – 285.
Acs, Z., Audretsch, D., 1990. Innovation and Small Firms. MIT Press, Cambridge.
Aldrich, H., 1999. Organizations Evolving. Sage, London.
Aldrich, H., Auster, E., 1990. Even dwarfs started small: liabilities of age and size and their strategic implications.
In: Staw, B., Cummings, L. (Eds.), Research in Organizational Behavior, vol. 8, pp. 165 – 198.
Aldrich, H., Wiedenmayer, G., 1993. From traits to rates: an ecological perspective on organizational foundings.
Adv. Entrep. Firm Emerg. Growth 1, 145 – 195.
Allinson, C., Chell, E., Hayes, J., 2000. Intuition and entrepreneurial behavior. Eur. J. Work Organ. Psychol. 9
(1), 31 – 43.
Bandura, A., 1986. Social Foundations of Thought and Action Prentice-Hall, Englewood Cliffs, NJ.
Bandura, A., 1997. Self Efficacy: The Exercise of Control. W.B. Freeman and Company, New York.
Bates, T., 1990. Entrepreneur human capital inputs and small business longevity. Rev. Econ. Stat. 72 (4),
551 – 559.
Baum, J., 1994. The Relation of Traits, Competencies, Vision, Motivation, and Strategy to Venture Growth. PhD
Dissertation, University of Maryland.
Bhide, A., 2000. The Origin and Evolution of New Businesses. Oxford Univ. Press, New York.
Bird, B., 1988. Implementing entrepreneurial ideas: the case for intention. Acad. Manage. Rev. 13 (3), 442 – 453.
Bruderl, J., Preisendorfer, P., 1998. Network support and the success of newly founded businesses. Small Bus.
Econ. 10, 213 – 225.
Bruderl, J., Schussler, R., 1990. Organizational mortality: the liabilities of newness and adolescence. Adm. Sci. Q.
35, 530 – 547.
Bruderl, J., Preisendorfer, P., Ziegler, R., 1992. Survival chances of newly founded business organizations. Am.
Sociol. Rev. 57, 227 – 242.
Campbell, D., 1988. Task complexity: a review and analysis. Acad. Manage. Rev. 13 (1), 40 – 52.
Carroll, G., Hannan, M., 2000. The Demography of Corporations and Industries. Princeton Univ. Press, Princeton.
Carter, N., Gartner, W., Reynolds, P., 1996. Exploring start-up event sequences. J. Bus. Venturing 11, 151 – 166.
Gartner, W., 1985. A conceptual framework for describing the phenomenon of new venture creation. Acad.
Manage. Rev. 10 (4), 696 – 706.
Geroski, P., 1995. What do we know about entry? Int. J. Ind. Organ. 13, 421 – 440.
Gifford, S., 1992. Allocation of entrepreneurial attention. J. Econ. Behav. Organ. 19, 265 – 284.
Gimeno, J., Folta, T., Cooper, A., Woo, C., 1999. Survival of the fittest? Entrepreneurial human capital and the
persistence of underperforming firms. Admin. Sci. Q. 42, 750 – 783.
Gollwitzer, P., 1999. Implementation intentions. Am. Psychol. 54 (7), 493 – 503.
Hannan, M., Freeman, J., 1989. Organizational Ecology. Harvard Univ. Press, Cambridge, MA.
 S. Shane, F. Delmar / Journal of Business Venturing 19 (2004) 767–785 785

Holtz-Eakin, D., Joulfaian, D., Rosen, H., 1994. Sticking it out: entrepreneurial survival and liquidity constraints.
J. Polit. Econ. 102 (1), 53 – 75.
Kahneman, D., Lovallo, D., 1993. Timid choice and bold forecasts: a cognitive perspective on risk taking.
Manage. Sci. 39 (1), 17 – 31.
Katz, J., Gartner, W., 1988. Properties of emerging organizations. Acad. Manage. Rev. 13 (3), 429 – 441.
Langer, A., Applebee, A., 1987. How Writing Shapes Thinking. National Council of Teachers of English, Urbana,
IL.
Locke, E., Latham, G., 1990. A Theory of Goal Setting and Task Performance. Prentice-Hall, Englewood Cliffs,
NJ.
Miller, P., Fagley, N., 1991. The effects of framing, problem variations, and providing rationale on choice. Pers.
Soc. Psychol. Bull. 17, 517 – 522.
Miller, G., Galanter, E., Pribram, K., 1960. Plans and the Structure of Behavior. Holt, London.
Mintzberg, H., 1994. The Rise and Fall of Strategic Planning. Free Press, Boston.
Mischel, W., Shoda, Y., 1995. A cognitive – affective theory of personality: reconceptualizing situations, dispo-
sitions, dynamics, and invariance in personality structure. Psychol. Rev. 102, 246 – 268.
Nuttin, J., 1984. Motivation, Planning and Action. Erlbaum, Hillsdale, NJ.
Pervin, L., 1989. Goal Concepts in Personality and Social Psychology. Erlbaum, Hillsdale, NJ.
Reynolds, P., White, S., 1997. The Entrepreneurial Process. Quorum Books, Westport, CT.
Romanelli, E., 1989. Environments and strategies of organization start-up: effects on early survival. Admin. Sci.
Q. 34, 369 – 387.
Rousseau, D.M., 1997. Organizational behavior in the new organizational era. Annu. Rev. Psychol. 48, 515 – 546.
Salancik, G., 1977. Commitment and control of organizational behavior and belief. In: Staw, B., Salancik, G.
(Eds.), New Directions in Organizational Behavior. St. Clair Press, New York, pp. 1 – 53.
Schoonhoven, K., Eisenhardt, K., Lyman, K., 1990. Speeding products to market: waiting time to first product
introduction in new firms. Admin. Sci. Q. 35, 177 – 207.
Shepherd, D., Douglas, E., Shanley, M., 2000. New venture survival: ignorance, external shocks and risk reduc-
tion strategies. J. Bus. Venturing 15, 393 – 410.
Simon, H., 1955. A behavioral model of rational choice. Q. J. Econ. 69, 99 – 118.
Simon, H., 1997. Administrative Behavior. Free Press, New York.
Stevenson, H., Van Slyke, J., 1985. Pre-Start Analysis: A framework for thinking about business ventures.
Harvard Business School Case Study Product Number 9-386-075.
Taylor, M., 2001. Self-employment and windfall gains in Britain: evidence from panel data. Economica 68,
539 – 565.
Tolchinsky, P., King, D., 1980. Do goals mediate the effects of incentives on performance? Acad. Manage. Rev. 5,
455 – 467.
Tuma, N., Hannan, M., 1984. Social Dynamics: Models and Methods. Academic Press, San Francisco.
Utterback, J., 1994. Mastering the Dynamics of Innovation. Harvard Business School Press, Boston.
Van de Ven, A., 1980. Early planning, implementation, and the performance of new organizations. In: Kimberly,
J., Miles, H. (Eds.), The Organizational Life Cycle, pp. 83 – 133.
Venkataraman, S., 1997. The distinctive domain of entrepreneurship research: An editor’s perspective. In: Katz, J.,
Brockhaus, R. (Eds.), Advances in Entrepreneurship, Firm Emergence, and Growth, vol. 3, pp. 119 – 138.
Venkataraman, S., Van de Ven, A., Buckeye, J., Hudson, R., 1990. Starting up in a turbulent environment: a
process model of failure among firms with high customer dependence. J. Bus. Venturing 5, 277 – 295.
Weick, K., 1979. The Social Psychology of Organizing. Addison-Wesley, Reading, MA.