Relevant Attributes of Start-ups: For What Should We Look?

The attributes examined for each group of companies follow in the tradition of Gartner (1985). After reviewing many earlier studies, he suggested that entrepreneurship has four dimensions: (1) qualities of the founder, (2) the new organization's structure and strategy, (3) the industry, market, and business environment of the venture, and (4) the processes followed in forming the venture. Gartner believed that interactions among the dimensions determine a venture's performance. A great deal of empirical work has considered how to define and measure relevant aspects of each dimension.

Other researchers have examined how the four dimensions (usually stated in ways that apply to existing businesses) interact and influence business performance. One particularly comprehensive study (Baum, 1995) develops the structural model shown in Figure 1, and tests it on a sample of companies in the architectural woodworking industry. Baum's model has excellent explanatory power (92 percent of the variance in growth) for that industry. He finds three strong, direct influences on growth: (1) the industry-relevant competencies of the CEO, (2) the CEO's motivation for and belief in growth, and (3) a strategy of innovation, quality, and personal selling. Competencies and motivation are themselves powerful influences on strategy. Further in the background "traits" (e.g., tenacity) influence motivation, strategy, and general competencies (e.g., organization skill)--which in turn also influences motivation. The personal qualities of the CEO emerge as the dominant influence on company growth. Those qualities affect growth directly as well as heavily influencing business strategy, another important determinant of growth. By comparison, industry structure and a company's organizational structure are minor influences.

Baum's study is cross-sectional so the mechanisms leading to success are not directly studied. But in the relatively mature, localized architectural wood industry one envisions the more skilled and motivated CEO's gaining market share from their less competent, less motivated competitors through well-known methods--innovative products, customer responsiveness, the personal touch and hard work.

The specific measures and relationships in Baum's model will probably vary depending on the industry and the age of the companies under study. Forces causing variation in the structural relationships include:

• "Best practices" may be better known and observed in certain industries. To the extent that all participants observe best practices a sample will show little variation regarding a practice and it will not serve to discriminate between success and failure--even though it may be very important. Such practices could include, e.g., hiring highly experienced and motivated managers, or using cross functional design teams.


Baum's Structural Model of Company Performance

(Total Effects Shown within Boxes)

• Optimal strategies may depend on industry structure, e.g., growth rate, stage of life cycle, or economic sector (Kunkel & Hofer, 1993). The importance of strategy vs. execution may vary as well.

• In rapidly changing industries a company may have a choice between high risk, "bet the company," strategies that maximize potential growth, and more flexible strategies that offer a higher likelihood of organizational survival. For any given industry a study of survival may exhibit different findings than a study of growth.

Despite the likelihood that the relationships in the architectural wood industry will not hold universally, Baum's study shows that comprehensive models with carefully designed constructs can have a powerful explanatory ability. It suggests the need to examine other industries with comprehensive models, and to examine the mechanisms by which entrepreneurial companies succeed and fail.

Types of Start-ups

Any entrepreneur must have some "model" of a start-up company in order to guide his/her actions. If one universal, simple model existed the task of giving a new company a good start would be relatively easy. But researchers have generally believed there are multiple types of start-ups--that no universal model will suffice. Woo, Cooper and Dunkelberg (1991) tested the validity of a taxonomy that classified start-ups into "craftsman" based, and "opportunist" based. They found that, depending on the attributes one considered, a sample of 550 start-ups shifted between the groups. That is, that dichotomy did not capture the differences among the companies.

Gartner, Mitchell, and Vesper (1989) apply Gartner's (1985) four dimensional framework to a sample of 106 start-ups, and identify eight types. They conclude that their eight types were valid but that a sizable fraction of all start-ups did not fit within the eight.

Given that a single, correct taxonomy has not yet been discovered, and that clearly more than one type of start-up exists, we will begin with an intuitive distinction, one that is used by practitioners, and examine whether companies grouped by those distinctions have distinctly different qualities. The groups are as follows (those expected to be least similar are the most widely separated):

• Large-I: venture capital backed, high-tech.

* Large-II: venture capital-backed, not high-tech.

• Small-II: not venture capital-backed, high growth.

• Small -I: not venture capital-backed, not high growth.

Over 99 percent of new companies nationwide fit the "small" categories. Reynolds (1993) describes the characteristics of a sample of 1,424 such companies, some of which were highly successful. None received venture capital backing. The group "Small-II" roughly matches the top 2 percent of his sample, labeled by him as "high performers."

The four groups fit our research questions: (1) the distinctions, if any, between small and "large" start-ups; (2) whether and how those differences influence outcomes, and (3) whether groupings based on outcomes identify shared attributes better than groupings based on initial characteristics.


Top of Page
Previous Page | Main Menu | Next Page
Return to 1996 Topical Index

1997 Babson College All Rights Reserved
Last Updated 5/1/97 by YuBei Teng.

To sign-up for the Center for Entrepreneurial Studies' publication lists,
please register with the
Entrepreneurship WebTeam.