"The best writing can't turn a bad idea into a good grant proposal, but bad writing can turn a good idea into an unfunded proposal." This quote adorns the cover of the outlines I distribute to people who attend my workshops. A fundamental principal of grantsmanship is that no amount of sublimely good writing can fool reviewers into funding a project that is based on a poor - or poorly developed - idea. A fundable project requires both a good idea and a good - that is, well thought out and well written - grant application.

Innovation has always been viewed as a plus by reviewers - as long as the novelty is not beyond their intellectual capacity.

But recently "innovation" was officially added to the National Institutes of Health (NIH) review criteria. Ellie Ehrenfeld, director of the Center for Scientific Review (CSR; formerly the Division of Research Grants) at NIH, points out that "establishment of [the new] explicit review criteria should provide a guide for reviewers, as well as information to applicants as to the basis on which their applications will be evaluated. The overall intent is to focus the review on the importance and potential impact of the project rather than on experimental details. The inclusion of innovation is one of five rating criteria is intended to signal investigators that creativity is valued, and to encourage them to submit creative proposals at a time when limited resources have led study section members to favor safe science."

A good proposal idea must also tweak the reviewers' imagination. Donna Dean, director of the Division of Physiological Systems, CSR, NIH, says that investigators must energetically convey their enthusiasm and sense of excitement about their work and the new research directions they are planning. If the researcher can concretely and clearly describe what is known, and then present a logical leap forward into the unknown, there is great potential for capturing the reviewers' interest.

All other things being equal, many funding agencies favor hypothesis-driven ideas over non-hypothesis-driven proposals. Proposals likely to be labeled "descriptive" are highly likely to remain unfunded. Sharply focused ideas are likely to fare better than diffuse, all-encompassing proposals. It is part of grantsmanship lore that "safe ideas," likely to yield good results are more likely to get funded than high-risk ideas. However, NIH does ask reviewers to flag proposals as high risk or high impact. With recent revision of review criteria at the National Science Foundation to include "creativity," and at the NIH to include "innovation," perhaps there will be increased funding of riskier proposals that are likely, if successful in their outcome, to have a major impact on a given field. Paul K. Strudler, scientific review administrator of the Radiation Study Section in the CSR, NIH, says, "It's hard to make any sweeping judgments in this first cycle of meetings in which the new criteria are being used. But . . . during [Study Section] discussions, 'significance' and 'innovation' sometimes served as counterweights in debates in which 'approach' had become the primary concern."

How to come up with - and develop - an innovative proposal idea?

First, be conversant with the literature; you are not likely to get funded for reinventing the wheel.

Second, understand the nature of the gaps in knowledge in your given field.

Third, be aware of what scientists in related fields, perhaps even in seemingly unrelated fields and in other contexts,, have done and are doing. A fresh view of an old set of data may sometimes be more insightful than new data. It is no accident that many of the major contributions to modern biology were made by physicists. It is often the people with the "outsider" perspective who frequently unearth hidden problems that have insidiously evolved over time from initially valid procedures. And those thinkers who understand the interdependence of knowledge and have the insight to synthesize ideas culled from different disciplines who sometimes make the most innovative contributions. As a speaker I once heard, said: "The major advance in transportation did not occur by making a better track or better wheels, but by exploring a new concept that man could fly."

Fourth, don't rely just on the mainstream journals in your field: relevant material and interesting ideas are sometimes published in seemingly unrelated and more obscure areas. Go to hear speakers of note in fields other than your own. Search far afield for information that may superficially seem irrelevant but which, on closer examination, may tie into your area of interest. Ideas often come from clever people in fields unrelated to your own. Supposedly Kékulé, after hearing about Darwin's theory of evolution, dreamt about monkeys dancing around in a ring, which gave him the idea for the six-membered ring structure for benzene. Keep your eyes, ears, and mind open.

Fifth, discuss your ideas with trusted colleagues in your own and in other fields. Brainstorming with others often helps to bring your own thoughts into focus. Of course, it is best to discuss your ideas with the brightest people available. But sometimes, discussing your thoughts aloud to an unresponsive person - or even a blank wall - can help to crystallize your vision.

If you are strapped for ideas, investigate the mandate - and, if available, the long range plans - of a relevant funding agency. The NIH National Eye Institute, for example, has published 5-year plans from time to time, delineating what types of research the Institute will support in the subsequent five years. But don't rely only on the printed word; talk to program officers at the agency to determine more specifically the particular aspects of a research area about which the agency is most enthusiastic. The "schmooze" factor can be very important - but double check information obtained verbally. It's also wise to keep track of what the community considers to be "good" ideas. Read the Hot Papers section of The Scientist each week and keep track of what sorts of projects garner honors - but be influenced selectively.

Once you have the good idea in place, half the battle is won. But the good idea alone is not enough. Colette Freeman says "Although a good idea can serve as a useful introduction to a research proposal because it will pique the reviewers' interest, it has to be solidly packaged with data and methods. And to assure that the reviewers recognize the novelty and the significance of an application, the author should spell it out for them." A track record, as evidenced by ample publications in well-reviewed journals, a history of concentration in a particular research field/on a particular problem, and a clear, concise, well-written proposal that is responsive to the instructions of the potential funding agency, are also helpful in getting funded.