GGI Legacy Site

We have a new look. Please see our new home page.


GGI RapidNews R&D Product Development eZine: Volume 2, Issue 9- November 2, 2001
GGI RapidNews is published monthly


Looking for information on R&D Metrics? Please see our latest Metrics Summit.

2000 Metrics Survey Information: Free downloadable copies of the 2000 R&D Metrics Survey Description and/or Survey Questionnaire are available in our Market Research Reading Room at You will also find free downloadable files for GGI's 1998 Product Development Metrics Survey. The 1998 and most 2000 Metrics Survey results are available for purchase at

2000 Metrics Survey Results: The last section of the 2000 Survey questionnaire explored the commonality of R&D metrics used by R&D and product development across industries, i.e., what metrics are most frequently used.

  • The 2000 GGI R&D Metrics Survey listed 48 "commonly known" R&D metrics, and asked respondents to check off those that were used in their company. 30 of the 48 metrics were in use by 5% or or more of the respondents.
  • Only 3 metrics were used by more than 50% of respondent companies. There was a large gap to the 4th most common metric, which 31% of the companies used.
  • Two (2) of the 3 metrics reported in use by more than 50% of respondent companies are measures required by legal, financial and/or regulatory authorities; or are derived by sorting information required by those regulatory authorities. They top 3 metrics were "R&D Spending As A % of Sales," "Total Patents Awarded/Pending," and the famous 3M metric whose abbreviated description is "Sales Due To New Products."
  • The famous 3M metric first introduced to industry in 1988: "Current Year Sales/Profits Due To Products Released In The Prior N Years" is the only "new" R&D metric in the last 70 years to gain widespread industry acceptance. It was the second most commonly used metric.

There has been little innovation in R&D measurement practices for the past 70 years since regulatory reporting requirements were first introduced in the 1930s. GGI previously surveyed this in 1998 and the results were almost the same. Going forward we should expect, as happened when manufacturing processes during the 1980s and 1990s matured and measures became more common, that R&D management and measurement practices will mature and become more common. We expect in the future that many more R&D metrics will become commonly used across industry.

GGI has been selectively releasing results of the 2000 Survey in most issues of RapidNews during the past year. If you are interested in learning more about about these results, please visit our RapidNews "archives" at

2002 Metrics Survey Beta Test: Our biennial survey will be conducted again in 2002. If you are interested in participating in our "Beta Test" of the initial 2002 Survey design, please contact and let us know. All recipients of RapidNews will be given the option of participating in the 2002 Survey once the Beta Test activity has been completed. We expect to commence this activity in the spring of 2002.

Industrial R&D Patent Distribution: Several issues ago we discussed patent trends and their possible implications for future R&D innovation. By using patent numbers and quality as a R&D metric, several researchers* have found high concentrations of innovative R&D performance in very small numbers of company personnel. They alone appear to be largely responsible for their company's innovation and competitive success. While this concentration of R&D technological inventiveness with few people has long been theorized, to date there has been little empirical evidence to substantiate it in companies.

The authors, in summary, say the following:

1. Industrial R&D is the most important source of product and process innovation.
2. That innovation is decisive to establish company competitive advantage and growth.
3. R&D "people management" is therefore important to technology management.
4. The primary task of R&D management is to identify, use and promote R&D potential.
5. Evaluation of R&D people performance is critical to the realization of that potential.

The authors contend that patent numbers and quality, such as patent citations or foreign country patent applications, are a non-subjective means to determine R&D performance. They are a reliable output measure because, they say, there is an empirically shown positive relationship between expenditures and R&D patent output. Patents indicate R&D results, which have been transformed into economically successful products.

Until recently, few empirical studies of inventor distribution on the company level, rather than in the "field of science" have been carried out. The authors examine the existence of key inventors, analyze contributions to company performance and identify factors which contribute to key inventors becoming key. The differences in this study are reflected in the concern for "time of patent activity" and the "quality of resulting patents," where quality is measured by "technological value" (share of patents granted) and "economic benefit" (share of valid patents, citation ratio and share of US patents). Use of these metrics result in an even greater concentration of quality patents among R&D personnel.

43 German companies were selected among the chemical, electronic and mechanical engineering industries. These industries file the most German patent applications. Company sizes were small, medium and large, based on sales revenues. Medium was 100-500 million DM. Patent data were retrieved from the on-line databases PATDPA at and WPINDEX at

The research findings were as follows:

  • Inventive performance is highly concentrated in the companies studied with the top 10% of inventors 3 times as patent active and 4 times as patent productive as the average inventor.
  • The number of key inventors is very small. On average, they account for about 24% of all patent applications and 35% of all company patenting performance, but in any one company they may be as high as 69% and 86% respectively.
  • The percentage of key inventors decreases with increasing company size. Absolute numbers are, however, higher. Key inventors are more likely found in the small and medium companies and are therefore more important to those company's economic performance.
  • The electronics and mechanical engineering industries are characterized by a higher percentage of key inventors than the chemical industry.
  • Differences between key and normal inventors were significant. However, the difference between share of patents granted between the two was not. Differences were significant for economic patent quality. Key inventors significantly contribute to their company's market competitiveness and profit margins.

The management implications of this research are substantial. The take-aways can be summarized as the need for R&D knowledge preservation, technological process gatekeeping, increasing overall R&D performance and headhunting for key inventors. The importance of key people to technology innovation and market competitiveness emerges as a crucial objective for all R&D managers.

* Holger Ernst, Christopher Leptien and Jan Vitt; "Inventors Are Not Alike: The Distribution of Patenting Output Among Industrial R&D Personnel"; IEEE Transactions On Engineering Management, Vol. 47, No. 2, May 2000

Life Science Salaries Still Grow: "While real average weekly earnings for all US workers rose 3.6%, seasonally adjusted, from August 2000 to August 2001, median salaries for US life scientists increased 7%.

The AAAS (American Association for the Advancement of Science) survey of 19,000 life scientists also found some key issues. This is the most comprehensive survey yet of salaries among US life scientists.

They are among the top earners in the country. Real average earnings for all US workers in August 2001 was $490.54 weekly, coming to $25,508 a year. (An evaluation of average annual pay for all US workers covered by state and federal employment insurance programs set the average higher: $35,296 in 2000).

Academic life scientists earn median salaries of $80,000 in 2001 while non-academics report salaries with a midpoint 20% higher at $96,000.

But women life scientists don't do as well.

Male life scientists in academia earn median salaries that are 32% higher, on average, than salaries for women--$94,000 vs. $72,000. The gap is even larger in non-academic settings. Some of this is due to the fact that more males are further in their career cycle, have worked longer, and are in the high-income field of medicine. Women tend to work more in academia, which pays less.

For physicians and executives, the gender difference can be large. Male CEOs and VPs in non-academic settings earn around $160,000 against $125,000 for women. Male physicians outside academia earn $130,000 while women make about $90,000.

Where's the most money? The survey reports these median salaries for academic vs. non-academic life scientists: medicine, $145,000 academic, $140,000 non-academic; pharmacology, $98,000 and $110,000; biotechnology, $63,000 and $100,000; cancer biology, $85,000 and $88,000; immunology, $84,000 and $88,000; neuroscience, $80,000 and $92,000; biochemistry, $71,000 and $92,000; genetics, $75,000 and $90,000; microbiology, $70,000 and $83,000; and cell biology, $68,000 and $85,000.

Among non-academic life scientists, CEOs and VPs earn the most ($153,000), followed by physicians ($125,000). Directors and assistant directors make about $115,000, managers $90,000, principal investigators, $97,000, scientists, $86,000, and research investigators, $76,000.

Academics who are administrators or full professors earn $112,000 and $108,000, respectively. Non-teaching research personnel earn $42,000, associate professors, $72,000, assistant professors $62,000, and adjunct instructors, $43,000.

Aside from money, what makes life scientists happy? Intellectual challenge was rated as highly important by 79% and autonomy and flexibility in decision making were rated top by 70%. Salary and compensation were rated as a top concern (no more important than geographic location) by 53%.

The survey will be available free to AAAS members through a fully searchable database at the AAAS members-only Website, An analysis of the survey will be freely available to the public at "

Copyright 2001, Frost & Sullivan, New York, NY 10006

MRT Metrics Conference Wrap-Up: The October 15-17 Product Development and R&D Metrics Conference "Quantifying and Justifying Product Portfolio Decisions" held in Chicago was well received. This 6th Annual Metrics Conference has developed a loyal following from industry-leading companies, most of whom are way out in front in their R&D and product development measurement practices. There were outstanding keynote presentations by Dr. Robert G. Cooper and Donald Reinertsen; as well as outstanding presentations by HP, FedEx, CHI Research, Haliburton Energy and Harley Davidson. Subjects included "New Product Generation Metrics for TTM," "The Changing Metrics of US R&D," "An Industry Survey of Key Processes, Metrics and Tools of Collaborative Design" and "Driving Strategic Alignment Through Product Development Portfolio Management."

Brad Goldense, who has been involved in all six years of this conference, presented GGI's recently developed seminar on Proactive and Predictive R&D Metrics (PPRD) in a half-day version that was well received. The chief regret expressed was that the seminar didn't last a whole day.

The National Institute of Standards and Technology [NIST] is in the process of incorporating some of the PPRD frameworks and metrics into a to-be-released guideline and/or standard on R&D measurement. If you are interested in having this seminar conducted at your company, please directly contact Brad Goldense at If you are interested in simply purchasing the Coursebook used for this seminar, please visit GGI's iStore at

Web-Enabled R&D: On December 11-12, 2001 in New Orleans, the International Quality & Productivity Center (IQPC) will present a conference on choosing and implementing technology for cost-effective R&D. Brad Goldense is the conference chairperson. On December 10th, in a preconference Seminar, Brad will present GGI's Proactive and Predictive R&D Metrics seminar. Visit for additional details and registration information.

By attending this conference, you will learn more about how 3M built and continues to grow a digital library for smooth document management; how Coors aggregated and deployed its key documents for company-wide use and more effective R&D; how NASA is bringing down the cost of technology by integrating new tools with existing technology; and how AT&T has developed its process for choosing which new technology will be integrated with its legacy systems for maximum company effectiveness.
RapidNews is an e-mail publication from Goldense Group, Inc (GGI). Its subject matter includes survey findings, company news, book reviews, key industry conferences and R&D information of interest to clients and associates. Please send communications to rn(at) Thank you.