A Comparison of Methods and Sources for Obtaining Estimates of New Venture Performance

Candida Brush, Babson College, USA

(with Pieter A. Vanderwerf)


Measuring the performance of new ventures is of interest because they are a major source of job creation and because improvement in performance is critical to their survival and growth. However, collecting data on the performance of new ventures is often difficult due to a lack of historical information and accessibility. This article presents conclusions from a literature review of 34 current empirical studies from the entrepreneurship field that measured some aspect of performance and describes the results of an exploratory study that tested empirical variation across two methods of data collection and three sources of information used in measuring the performance of new ventures.

Sixty-six recently formed (4–6 years old) manufacturing firms in Massachusetts were identified and queried on different aspects of performance. The sample was split: one group being surveyed by telephone, the other by mail. Besides the self-report by the venture, two additional sources of performance information were used: an archival source and a competitor that had been identified by the new venture. Measures of performance information included those most frequently used by researchers, such as annual sales, number of employees, return on sales, growth in sales, and growth in employees. Both subjective and objective measures were employed. Correlational tests and regressions were used to compare measures, sources, and methods for reliability.

Results show that sales figures obtained from archival sources and direct questioning of new ventures were highly and significantly correlated. Competitors proved to be a reliable third source in that the performance estimates they made were highly correlated with the estimates reported by the new ventures themselves, but their estimates sometimes differed widely in absolute value. Both mail and telephone methods rated well in obtaining complete responses, even though the response rate for the telephone was twice that of the mail responses.

The main implications for researchers are consideration of the trade-offs in terms of cost, time, and accuracy for various methods of data collection and sources of performance information. This research does not attempt to prescribe which measure to use; rather, it offers the results of empirical tests that suggest which methods and sources might be used to collect the information.

For the new venture owner/manager, this research suggests that competitors of new ventures are often knowledgeable of the sales and profitability of new ventures.


*What is a DOI?