Submitting an NSF grant proposal is no small feat—it’s a journey of planning, revising, and learning. After reflecting on the feedback I received for my submission, I’ve compiled some lessons that can help others avoid common mistakes and create stronger proposals.
Reviewers want to know what happens after the grant ends. If you’re creating a course, explain how it will stay part of the curriculum. If you’re building a tool, outline how it will be maintained or shared (open-source, partnerships, etc.). Think beyond the grant period and show how your project will make a lasting impact.
Suppose your project involves creating an augmented reality (AR) application for teaching mechanical engineering concepts. In your proposal, you can state:
"After the funding period, the AR application will become a core component of the [Course Name]. (1) To ensure its long-term viability, we will establish a collaborative agreement with the university's digital innovation lab to manage updates and integrate new features based on user feedback. (2) To promote accessibility, the app’s source code and development documentation will be made available on a dedicated GitHub repository under an open-source license, enabling other institutions to customize it for their courses. To further its reach, we will partner with community colleges and underrepresented minority-serving institutions to pilot the app in their introductory engineering courses and collect feedback for iterative improvements."
Your reviewers may come from different fields, so don’t assume they know your jargon. Take time to explain key terms and concepts upfront. This helps ensure everyone understands your proposal the way you intend.
Suppose your project involves integrating a computational fluid dynamics (CFD) simulation into undergraduate engineering labs. In your proposal, you can state:
"The project incorporates computational fluid dynamics (CFD), a branch of fluid mechanics that uses numerical methods to solve and analyze problems involving fluid flows. For clarity, CFD in this context refers to simulations that model the behavior of fluids in motion, such as airflow over an aircraft wing or water flowing through a pipe. These simulations rely on discretizing the governing Navier-Stokes equations, which describe the motion of fluid substances. By simplifying complex physical phenomena into manageable models, CFD enables students to visualize and analyze real-world engineering problems, bridging theoretical knowledge with practical application."
To ensure clarity throughout your proposal, you can include a dedicated section in the introduction for key terms and definitions. For instance, here is what I did for one of my proposals:
Collaborations can strengthen your proposal, but only if they’re specific. Don’t just list your partners—describe how they’ll contribute. What will they bring to the table? How will they help achieve your goals? Concrete examples make these partnerships feel real and impactful.
Suppose your project involves developing a blended learning module for environmental engineering that incorporates real-world data from local ecosystems. In your proposal, you could highlight collaborations like this:
"This project leverages a partnership with the Generic Research Center (GRC), a leader in ecological data collection and analysis. IERI will provide access to their database of water quality metrics collected from over 50 sampling sites across the state. These datasets will form the foundation for the student exercises, enabling hands-on learning with authentic, real-world data. Additionally, Dr. John Due, Director of Educational Outreach at GRC, will co-develop a set of virtual lab modules integrating this data with our learning platform. Her expertise in translating research into classroom tools will ensure the materials are pedagogically sound and aligned with learning outcomes."
A good proposal doesn’t just describe what you’ll do—it also explains how you’ll know if it’s working. Include details about how you’ll measure success, both as you go (formative evaluation) and at the end (summative evaluation). For example, compare results from before and after your project to show its impact.
Suppose your project involves implementing an innovative blended learning approach to improve problem-solving skills in engineering students. In your proposal, you could write:
"To measure the success of this project, we will employ a comprehensive evaluation plan incorporating both formative and summative assessments. The following components will ensure rigorous and reliable evaluation:
- Validated Instruments: The Engineering Design Self-Efficacy Scale (EDSES) will be used to measure changes in students' confidence and skills in tackling complex engineering problems. This instrument has been extensively validated and ensures consistency in assessing the targeted outcomes.
- Constructs and Descriptors: We will track improvements in key constructs such as critical thinking, collaborative engagement, and problem-solving accuracy. These descriptors align with the program's learning outcomes and will be mapped to specific project activities.
- Formative Evaluation: Mid-semester surveys and focus groups will be conducted to gather student feedback on the blended learning modules. This feedback will inform iterative adjustments to the curriculum, ensuring it meets student needs effectively.
- Summative Evaluation: At the end of the semester, we will compare pre- and post-project data, including performance metrics on a standardized engineering concept inventory and qualitative analyses of student reflections.
- External Evaluator: Dr. Jane Roe, an expert in STEM education evaluation, will serve as the external evaluator for this project. She will oversee the data collection process, validate findings, and provide an independent report on the project’s impact. This ensures objectivity and enhances the credibility of our evaluation approach.
NSF cares about how your project benefits society. Be specific about who will benefit (students, educators, industries) and how. For example, will you create free online resources? Share data or tools with other schools? Involve underrepresented groups? Don’t just say your work will make an impact—show how.
Reviewers appreciate when you think ahead about potential challenges. What if your technology doesn’t work as planned? What if student participation is lower than expected? Include a contingency plan that shows how you’ll adapt and keep the project moving forward. For example:
Suppose your project involves implementing a community-driven design course that requires students to co-develop solutions with local organizations. In your proposal, you might include the following:
"While the project is designed to foster collaboration between students and local nonprofit partners, we recognize potential challenges, such as fluctuating availability of community members or difficulty in securing consistent engagement from stakeholders. To address these risks:
- Backup Activities: If community partner participation declines, we will supplement the course with case studies based on real-world community issues, using anonymized data from previous partnerships.
- Alternative Data Collection: In the event that field visits are not possible, students will engage in virtual interviews with experts and use publicly available datasets to complete their design analyses.
- Reallocation of Resources: If a partner withdraws entirely, project teams will pivot to working with alternate organizations already identified during the planning phase. We have pre-established contacts with three additional local nonprofits as contingency collaborators.
A clear timeline is essential. What will you do and when? What will you deliver at each stage? Spell it out so reviewers can see your plan is realistic and achievable.
Be transparent and detailed in your budget justification. Reviewers need to understand how each dollar supports your project. Whether it’s for equipment, student support, or travel, explain why it’s essential and how it ties into your goals.
Avoid using terms like "nontraditional" or "underserved" solely because they appeal to NSF. If your proposal does not genuinely aim to address these populations, it will be obvious. Be honest and clear about your intentions. This example shows how to genuinely address population-specific challenges without relying on "beautiful" terms as placeholders.
Suppose your project focuses on developing a mentorship program for first-year engineering students. Instead of overusing terms like "underserved" without substantiation, you could write:
"This project targets first-year engineering students, with a particular focus on those who face challenges in adjusting to rigorous STEM coursework. Our preliminary analysis of enrollment data reveals that students from rural communities often enter with limited exposure to advanced math and science courses, leading to higher attrition rates. By partnering with local high schools and offering a structured mentorship program, we aim to bridge this gap. While our program may benefit 'underserved' populations, such as rural students or first-generation college attendees, our primary focus is on creating a scalable framework to support all students facing academic transitions, regardless of background. This honest and focused approach ensures that the program addresses real needs rather than superficially targeting specific groups for broader appeal."