Prediction of new firms performance: some practices that increase the odds of success
Abstract This study developed an ordered probit model to predict small and medium size enterprise performance in
Chile as failed, mediocre, or successful,
which is more robust than the commonly used binary
probi t and logi t models. The model variables that explain and predict levels of performance are: use of the
Internet, starting with adequate working capital, having good financial and accounting records and controls, using specific
plans, having higher levels of education, partners, parents owning a business, and marketing the firm. The model is
significant (p = .000) and all the variable coeffients, except education, were significant. The sample includes 403 small
businesses classified as: 158 failed, 101 mediocre, and 144 successful firms. The model can benefit both the would be and
current entrepreneurs; those who assist, train and advise them; those who provide capital and supplies for their ventures;
public policy makers; entrenpeurship and small business educators and researchers. Implications focus on improving
success and avoiding failure of small businesses, and public policy recommendations in Chile are discussed.
Keywords Success . Failure . Performance Prediction . Internet . Working capital . Financial information . Record
keeping . Planning . Education . Partners . Parents . Marketing. Ordered Probit.
JEL Classifications M13 . C21 . J18 . M10 . A14
1 Introduction and Literature Review
There is little doubt that new business ventures introduce a dynamic element into the economy and can make an
important contribution to development (Fritsch, 2008). Entrepreneurs have a very important function in the economy;
they engender relatively much employment creation, productivity growth, and produce and commercialize high-quality
innovations (van Praag and Versloot 2007). However, most firms die in the first few years (Cressy 2006). Therefore, to
increase the likelihood of creating a successful business is a main issue for those who dare to bear the risk of starting a
new venture, and understanding why firms fail and succeed is crucial to the stability and health of the economy (Carter,
Williams, and Reynolds 1997; Pompe and Bilderbeek 2005).
Every entrepreneur starts a venture with hopes of succeeding. However, it can be found that each year in the
USA there are more than half a million startups, but almost the same number close, although not all closings are failures
(Bates 2005). In Chile, less than 42% of small businesses survive five years and less than 50% survive 10 years (Cabrera,
De la Cuadra, Galétovic, and y Sanhueza 2002).
To date, the research found in Chile is focused only to certain economic sectors, or to the use and incorporation
of a particular tool as some specific technology. It is also possible to find some studies regarding small firm efficiency
and in depth studies regarding their characteristics (Cabrera, et al. 2002; Silva, Majluf, and Paredes 2006; Yeyati and
Micco 2007), but there is no reference with regard to the specific causes that lead these companies to succeed and fail.
Suárez (1994) found that the successful companies are those that understand transformation as an integral process, and
Page 1 of 19 
Convention