It is beyond the scope of this post to discuss all the nuances of the SBIR program. However, my experience reviewing scores of proposals has taught me a few lessons about what is likely to get, or not get, funded. Most of the proposals I have looked at have been Phase I (6 months and $150,000) in the topical areas of chemical and biological sensors, but I think many of these lessons will apply to other proposals as well. If you work for a small business soliciting funding in this program, or you are an academic researcher looking to kick-off a new business venture with an investment from Uncle Sam, pay close attention.
- Include some preliminary data, any data, even if it is not yours, in the proposal. I know that preliminary data is downplayed in SBIR proposals, compared to standard NSF or NIH grants; however, it is important to present some information to show that your concept is not just a cocktail napkin doodle. By the way, simulations can augment experimental data but cannot replace it.
- Provide a detailed description of the device, its materials, and how it will be fabricated. Every sensor has both receptor (how the target analyte is picked out of the sample) and transducer (how this recognition is converted to a usable signal) components. Make sure you adequately describe each, even if you are using standard receptors such as antibodies.
- Make sure the fabrication has a pathway to manufacturing. If you have an elegant concept that can only be demonstrated in the laboratory, you are not ready to submit an SBIR proposal. There must be a product that comes out of this sometime down the road.
- Show a basic understanding of the characteristics and parameters of importance in your application area. If you are going to claim that your device will beat all the competition by orders of magnitude in sensitivity, you must also address the effects on selectivity.
- Write your research plan with sufficient detail. Describe how you will do the experiments, what are the samples and how many, how you will collect and analyze the data, what are the expected outcomes, and what alternative approaches exist in case of problems. Also, in my opinion, developing your plan for Phase II should not be part of your Phase I effort.
- Create a research team with diverse expertise. If you are an electrical engineer who has developed a new sensing widget which you plan to use to detect cancer biomarkers, you should have someone on your team with experience in biochemistry or clinical medicine.
- Have letters of support from potential collaborators, commercial partners, and customers. The more letters you have indicating what a great idea this is, how much they look forward to working with you, and what a need this fills in their specific market, the better.
- Make the figures big enough to read the axis labels. This may be obvious, but you would be surprised by how many people try to save space with tiny figures and then leave some of the 15 page limit unfilled.
- Don’t spend so much ink on commercialization plans. Maybe other reviewers look at this more carefully than I do, but for me it is more important to read the details of your technical description and research plan so that I am satisfied of your capability to do the work. If a reviewer is not convinced of your technical competence, it will not matter how much revenue you plan to earn.
- Create a budget and time commitment that is reasonable for the limited funding and project length. It is good to be ambitious, but most reviewers are also researchers who know when you have bitten off more than you can chew. Also, while you would like to keep all the funding to pay your own salary, it is much more believable if you diversify the budget to pay a team of researchers with specific roles, rather than putting the entire burden on one individual.