Notes on “The Art of Proposal Writing”

The following is from The Art of Proposal Writing that was presented SIGGRAPH 2008 on August 15, 2008. The full slides can be downloaded from http://www.cs.csustan.edu/~rsc/S08-proposal-course.pdf.

The only good reason to develop a proposal is that you have something you want to accomplish, and external funding is needed to accomplish it.

You may start with only a general idea of what you want to do, and your first tasks are to be sure that the idea could work

• Is the idea important to the field?
• Is there a potential funding source for the idea?
• Is there support on campus for the idea?

A project starts with the current situation and defines a set of tasks that will create a changed situation. Your task is to give a clear statement of the current and goal states and how you will move from one to the other.

Look at developing your project in several stages. You may be able to find different funding for eachstage

– Preparation (perhaps local startup funds)

– Proof of concept (perhaps local support, perhaps Proof of Concept programs from larger funding sources)

– Full development (national or large regional source)

The success at each stage helps to prepare the project for the next stage.

Funding is developed, not found

• You are funded through a conversation between you and the funder
• Effective funding is a result of telling your story well
• Your story should be about an investment, not a gift

Funding agencies have specific things they want each of their programs to accomplish. These are usually stated in their solicitations or program statements.

How do you learn from a decline?

– Talk with the declining program officer
– Do a careful post-mortem with your development office to see how to do better next time

Before writing a proposal!

• What story are you telling? (a great idea is not enough!)
• What are the technical goals?
• What has been done previously?
• What is the technical approach?
• Why is it better than previous / alternative approaches?
• What is the scientific impact?
• What is the broader impact?

What are the benefits for society?
– Scientific? Industry/Government/Society?
Will the results be disseminated broadly enhance scientific understanding?
– Software available? Testbeds?
What is the educational impact?
– Underrepresented groups? New courses?
Undergraduates? K-12 students?
What new infrastructure for research and education is being created?
– New instrumentation, partnerships, etc.?

Seven deadly sins of proposal writing:

1. Failure to focus on the problems/payoffs
2. No persuasive structure: poorly organized
3. No clear differentiation: competitive analysis
4. Failure to offer a compelling value proposition: potential impact
5. Key points buried: no highlights
6. Difficult to read: full of jargon, too long, too technical
7. Credibility killers: misspellings, grammar, inconsistent format, etc.

Broader Impacts – Scientific & Societal (some notes based on reviews of real proposals)
• The strong support from these industry partners signifies large potential value of the research.
• The proposal could potentially have a big impact for both research and industry.
OR
• Making the software publicly available would strengthen the impact of the work.
• The panel did not find available code from the PI’s previous work.
• A testbed application was needed that could be tracked through the different parts of the proposal.
• A tie to a specific application would have helped demonstrate that their method will outperform
existing techniques.

NSF-Oriented Education Projects
• A project has some specific properties you can use to organize your plans. It
– Solves a specific problem
– Has a goal, a plan, and a rationale
– Has institutional commitment
– Has a careful understanding of the cost
– Has a way to know how well it succeeds
– Has a way to share its results with others
• These give you specific ways to organize your proposal

Some Evaluation Resources:
• NSF’s User Friendly Handbook for Project Evaluation http://www.nsf.gov/pubs/2002/nsf02057/start.htm
• Other tools
– Online Evaluation Resource Library (OERL) http://oerl.sri.com/
– Field-Tested Learning Assessment Guide (FLAG) http://www.wcer.wisc.edu/archive/cl1/flag/default.asp

Some comments (with minor changes) I have written on reviews (Steve Cunningham)
The comments below come from my personal reviews on projects since I left NSF, covering intellectual merit and broader impacts, with a few changes to remove specific project or other references and to trim down longer comments. These are included to show some of the kinds of things that a reviewer may pull out of a proposal so that you can try to avoid them.
Intellectual Merit

The concepts discussed here are of potential interest, but do not yet seem well developed for computer science. These need to be better developed before they can be considered to be “well grounded in research and practice” and used as a basis for curricula. The project has a number of buzzwords strung together, but they do not seem to be well connected.
It would surprise the reviewer if this [project work] hadn’t been done someplace before, but there is no indication that this project has looked for that. Indeed, “literature search” is yet to be done. Although literature is cited in the proposal body, there is no bibliography with the proposal.
The project plan is very open and is based on the assumption that XXX, but it would have been very helpful if any example of such an approach had been cited.
The project faculty are active, but there is no apparent track record for their work with the kind of transformative ideas that would be needed by a project such as this.
The reviewer does not share the optimistic outlook of the proposers. … The question is whether there is enough push in the process to let it succeed. There are many steps involving many individuals and groups, but the narrative does not give any timelines. It’s difficult to see how this much process could give any vibrant, strong products. How could they expect to do all this in two years?
There are many good ideas in this project, and the reviewer is sure that it will serve the institution well. However, the reviewer questions whether this work is appropriate as a XXX project.
Part of the difficulty the reviewer had with this project comes from the fact that it was difficult to see just what the eventual outcome of this project would be.
There are some general statements about results, but little that was specific.The evaluation plan does not contain measurable goals or have methods to collect evidence about the project.
Broader Impact

The project feels weak in broader impact. It has very limited, passive dissemination and no active outreach. Its impacts are local and it’s difficult to see how it would have an interest for anyone else.

The project seems to be entirely local, with the project team coming from the departmental curriculum committee and the only outside input coming from the departmental advisory board. There is no history of working on transformational projects and the cited literature is quite thin. The reviewer was surprised to see no XXX publications or activities mentioned. In general, this project seems to be entirely focused on the local environment.
The dissemination plan is completely passive, which is surprising given the presence of the external board, and there is no mention of plans to extend any results to other institutions.
Dissemination looks vague, based on a range of unspecified journal publications. There are no workshops or other active processes planned to help other institutions develop similar programs.

Student diversity seems to be expected because of being an urban campus and because the faculty are diverse, but no figures are given about the diversity of the students.
It is not at all clear how this project would attract more women or underrepresented groups into computing degrees.