A few weeks back, I asked my followers on Twitter what topics they’d like to see me blog about. One of them suggested that since I do a lot of SBIR proposals at my day job, that maybe I could write an article on advice and lessons-learned for how to write better SBIR proposals.
For those of you not familiar with the term, SBIRs in this case stands for Small Business Innovative Research grants. Basically they’re small R&D contracts that NASA, the DoD and other federal agencies do that only small businesses1 can bid on. Each of these agencies will put out solicitations 1-3 times per year, with a list of topics of interest to the agency. Companies can then submit a proposal for innovative research addressing something in that topic, and if they get selected they’re given typically around $100-150k of money for a six month Phase I feasibility study contract. If that goes well, they can submit a Phase II proposal, and if they win that, it’s a 2-year $750k effort that usually culminates in some level of prototype. To make a long-story short, for bootstrapping aerospace startups, SBIRs if done right, can be a good source of non-dilutive funding, and can help you build up technology expertise as you work to get to a point where you can go after bigger commercial markets.
Anyhow, with that digression aside, I wasn’t sure when I got the advice if I was really competent to write something like this. Sure we’ve written a few and won a few, but we’re not really what I would consider an SBIR farm. I think to-date SBIRs have been less than 1/3 of our total revenue as a company. I ran the numbers tonight, and apparently between NASA and DoD SBIR and STTR2 Phase I proposals, including ones where we proposed as subs, we’re now up to over 43 proposals, with 11 wins. So, I think I can say we no longer totally suck at this, and are in fact probably solidly mediocre.
So… with that stirring endorsement in mind, here’s a few tips I could think of to help you improve your SBIR proposaling fu.
- Get to Know Your Customer: One of the best ways to improve your odds of writing a good proposal is to have spoken with the Topic Manager (TM) or Technical Point of Contact (TPOC) first, to learn more about what they’re really looking for, before the solicitation opens3. DoD makes this easy by having a “pre-solicitation” period where they explicitly list the contact info for the TPOC and give you a month to talk with them before the real solicitation opens. NASA makes it harder by not giving out TM contact info. In order to get to the TM your best bet is to look at old SBIR topics from previous years4. Find one that looks aligned with your interests. Look up which center is listed as the lead center for the topic. It usually stays the same over time. Look up that center’s SBIR Program or Tech Transfer Office, and ask them if they can put you in contact with the TM. Most of the time they will. Once you’re in contact, you can now do some customer discovery research. Find out what problems are keeping them up at night. Find out what specific sub-problems are the ones they’re most interested in solving. Find out what types of solutions they’ve looked at, and what their opinions are about the relative merits of those solution types. Find out if they have a transition path in mind–a government program that needs this, or if it’s more just professional curiosity. Find out about their philosophy on SBIRs and how strategic they try to be–are they just seeking fun ideas to fund? Or do they use the topic as a way to get free money to fund research in specific areas they need for an operational program? Find out if they can recommend any NASA researchers in this area that they work with who’ve written papers in this area. If it’s a DoD presolicitation period, I’d also try to dig to find out if the there is a specific technical approach they’re already sold on. On how many Phase I and Phase II awards they expect to give out for the topic, and things like that. Lastly, after you’ve done your customer discovery and know more about the problems they’re trying to solve, that’s when you can start batting ideas around with them to get their feedback. Doing this properly helps you not only by making sure you’re proposing something relevant, but now the people who’ll be reviewing your proposal at least know your name, and you have a better idea of their biases, technologies of interest, and how they see the competition. But best of all, if you’ve done your homework right you at least know if you have a shot at all of being awarded something. There’s nothing more frustrating than wasting several weeks writing something you think they’ll love, only to find that they consider your idea “non-responsive” to the topic. Ever since I found that the DoD has a pre-solicitation period, I’ve submitted way fewer proposals, but my win rate has gone up substantially. It’s much better to have them shoot a hole in your idea when you don’t have multiple weeks into it than to invest what is going to be thousands of dollars worth of your time into something that has zero chance of winning. So, to summarize — fail fast by talking with your customers and really getting to know them before trying to write a proposal.
- Sell the Combined Phase I/Phase II Story: This was a piece of advice I got from Greg Mungas, who used to run a propulsion R&D shop5 out in Mojave, and before that used to work at NASA JPL, where he was involved in reviewing SBIR proposals. Basically Greg’s point is that nobody funds SBIR Phase Is just for the Phase I. They fund Phase Is that they think will lead to good Phase IIs. So, when you tell your story in your SBIR, you always want to talk as though you’re proposing both phases. Here’s where we get to in Phase I, which sets us up to do X, Y, and Z in Phase II, getting to this really awesome demo or this point where the technology is ready for flight demonstration, or something like that.
- Hook Them With the First Page: Greg also pointed out that you get a lot of different sorts of people reviewing SBIRs, but one thing they have in common is that they almost never have time to thoroughly review everything they’re given. So, it’s really important to quickly hook the reviewer with an exciting first page or two, so they’ll be interested and engaged for the rest of the proposal. My typical formula for the first page includes a) a pretty piece of faux CAD6, b) a paragraph setting up a compelling-sounding problem you’re trying to solve, c) a paragraph introducing your concept in 1-3 sentences as a solution to said problem, d) a short bulleted list with some key highlights of the features/capabilities you think you can achieve with your concept, usually with a 1-3 word summary bolded at the start of the bullet followed by an unbolded one sentence description7, and e) a one paragraph summary at the end talking about what you’ll do in Phase I, what technology readiness level that gets you to, and how that sets you up to do awesome things X, Y, and Z in Phase II that will get you to an ever cooler end-state. Basically it’s almost a mini-proposal on one page that gets them excited to read the rest of your concept, and helps them understand where you’re going from there.
- Cite Chapter and Verse: Most SBIR solicitations ask you in your first section, where you’re describing the innovation, to explain why your innovation is relevant to the topic. In these cases, I’ve found it handy to quote the solicitation verbatim, and highlight or bold or italicize the parts of the topic that you specifically hit. Here is also where you can mention insights you gleaned about their needs from your customer discovery interviews.
- Quantity Has a Quality All of Its Own: Even if you’ve done your homework well, and are writing an exciting intro, and properly quoting chapter and verse, SBIR solicitations still can be fairly competitive. Your raw odds are typically around 1 in 6 to win something. If you’ve done your homework well, you can get up into the 50/50 range. But that still means you’re best off proposing more topics than you expect to win. We try to shoot for 4-6 proposals each cycle (if we have time), in the hopes of winning around 2 Phase Is, with the hope of converting at least one of those two to a Phase II.
- Sometimes Its Them, Not You: Sometimes you’ve done everything right and you still get shot down on a topic. Never fear–you should get a debrief a few months later with reviewer comments. Sometimes you caught someone on a bad day, or they misunderstood something. A lot of times we’ll rebid things a few times in a row, with tweaks and improvements based on the feedback you got, and any customer discovery you can fit in between cycles. Rebids typically don’t take as much time to write, and you have more data going in on what they liked and didn’t like. So don’t be afraid to rebid something if you think you had a good idea that got ignored last time around.
- Don’t Waste Time on an Overly Detailed Materials Budget: Most SBIR solicitations require you to provide a fairly detailed cost budget, including labor rates and hours, indirect rates, and “other direct costs” including materials. The annoying thing is that they want detailed quotes for everything you’re going to order in this research project that you haven’t even started yet. You don’t have final designs yet, and don’t really know what things are going to cost, but they want not just numbers, but quotes or at least some other form of cost justification. A lesson we’ve learned is that proposals never get rejected because your cost justification isn’t detailed enough. If you win the proposal, during contract negotiations they’re going to demand you back things up in more detail, but at that point its a winner’s problem. You still need to get in the right ballpark or you could get yourself in trouble8, but a lot of times our cost justification in the original proposal is a handwavy “estimated based on engineering judgement from past projects of similar complexity.” You do have to get them more justified numbers during contract negotiation, but now you’re dealing with a much smaller number of contracts.
- Contract Negotiators Hate Travel in Phase I: The people getting you under contract are paid to try and find wasteful proposed expenditures that they can disallow, to cut down a little on how much the government has to pay you for your research. I’m pretty sure they almost always cost the government more than they save, but they’ll go over things with a fine toothed comb. And one of their favorite things to nitpick is travel in Phase I. So it might be worth trying to do as much as possible via video calls and such to avoid giving them a juicy target to go after. I’ve only had my travel request rejected once, but they nitpick it and second guess it every time. In Phase II they’re more understanding, but not in Phase I.
- Baking In Commercialization Help in Phase I Might Be a Good Idea: In the Phase II proposals, at least for NASA, the Part 7 on your commercialization plan is a bear. In Phase I, they just want to know what applications you think there are, and your rough development plan for Phase II. But in the Phase II proposal, they want what amounts to a 5-10pg business plan for taking this product to market. It’s a lot of work, especially for technologies that are often at least a few degrees off of your main business plan. So, getting some help on that section could be worthwhile. One thing we’re going to try this time is that there are companies that specialize in “Commercialization Technical Assistance”. In Phase II, you can request $5k of CTA funding that doesn’t come out of your $750k cost cap. The CTA companies then help you with identifying potential Phase III opportunities, commercial customers, and government programs you could infuse your technology into. They won’t give you that bonus $5k for CTA help in Phase I, but there’s no reason you couldn’t include a subcontract to a CTA firm to provide that help during Phase I–so long as you can fit it into the allowed ~33% subcontracting limit. That’s what we’re going to try doing on a few of our Phase Is this time around.
- Pay Yourself a Competitive Wage: As a startup founder, it’s often tempting to underpay yourself. The logic goes that doing so means more money going back into the business, which increases your effective runway for the company. And after all, you own a chunk of it, so why not take a lower salary to improve your odds of being successful. Unfortunately, I’ve gotten bit by this several times, and only finally learned my lesson. Basically, since you’re proposing more contracts than you expect to win, you almost always win some combination of contracts that leaves someone important double-booked, where you’re now going to have to hire someone new to fill in the gap. But if you were paying yourself below your replacement cost, that makes it really hard. Either you end up trying to recruit people paying them below market wages, or you end up losing money on the contract. But if you pay yourself at least what it would take to replace yourself as an engineer, then if you have to replace yourself, it’s a lot easier to do so without losing money because you can offer them a competitive salary.
- Always Have a Fun Proposal Just in Case: When we do our “gut gate” meeting, where we go over topics we could propose on, to triage things down to a reasonable number of proposals, I always try to leave at least one proposal that’s my “in case we get done with the others early enough” proposal. This is often the fun one that you can’t quite justify prioritizing over the better aligned topics that you feel compelled to do for business alignment reasons. You’ll need to spend a little time up front on long-lead time items like faux CAD, figuring out your teaming strategy, and coming up with a high-level plan for your tech objectives and key tasks. But other than that, you don’t spend a lot of proposal time on that one until you’ve knocked out your high priority ones. But by leaving the fun one for last, a) you’re incentivized to take care of your responsible ones first so you can get to the fun one, and b) once you’ve gotten into the groove of writing proposals, that last one goes a lot faster than you think. Our one Phase II we won, for our Cryo Coupler was just such a “fun” proposal. I had done just enough to keep it alive while focusing on the others, and I freed up the morning the solicitation closed with 8hrs left until proposals were in. After you’ve written a few in the past few weeks, I’ve found it’s often possible to crank out a full Phase I proposal in a single day at the end–so long as you’ve taken care of the long lead time items up front. Even if that last one is skimpier, and not as detailed, it’s still worth sending it in. Even if your odds are only 10% instead of the normal 15%, that’s still one extra shot at a contract award that you wouldn’t have had if you didn’t at least try.
- Take Advantage of I-Corps Training If Possible: The NSF SBIR program found that they were getting a lot of PhD researchers who couldn’t commercialize their way out of a wet paper bag. So they created a program to train researcher sorts on how to do customers discovery and commercialization. NASA recently started offering it as an option, and at least when we did it, it both forced you to talk with customers, but also gave you training on how to do it, and at least a small additional travel budget that they don’t nitpick as much to go meet with customers and/or attend conferences or trade shows. It’s a great program, and if you’re at all the typical engineer founder, you probably don’t know the first thing about customer discovery. I-corps won’t make you a master, but they can at least get you from the totally suck level to the solidly mediocre level, if you take it seriously. I really wish this had been available during my first year or two of running Altius. Knowing how to actually talk with customers, and realizing that engineers are really good at creating business hypotheses, but often horrible at actually testing those hypotheses, has been one of the biggest insights I’ve made in the last few years. The sooner you get good at this, the sooner you can get to something scalable and worth investing in.
I could probably go on, but those are the best pieces of advice I could think of in one sitting. I may update this post later if I can think of any more items, but I wanted to at least get these out there, in the hopes of helping at least some startups avoid making all the same mistakes we’ve made. Good luck, and happy proposaling!
- Though by their definition, “small” means less than 500 employees
- STTRs or Small Business Technology Transfer Research grants are similar to SBIRs but you have to have both a small business and a research institution like a university on the proposal.
- Once a solicitation opens, you enter a “blackout period” where government employees typically can’t answer questions anymore.
- NASA tends to recycle topics a lot. One of the ones we’re wrapping up a Phase II effort for has been around in every SBIR solicitation for over 12 years now. I remember bidding on it back during my time at Masten.
- Firestar Engineering
- “Fake CAD” or high level CAD concept artwork that is meant to give an artist’s impression of the technology at hand, and to show some level of engineering thought has gone into it.
- For example you could have a bullet that says something like “Lightweight: Our widget is made of gee-whizium, so it can potentially weigh as little as 100g.” or “Mechanically Simple: Our snazzy magnet technology means we can solve this problem with no moving parts.”
- You can’t ask for more money if you screwed up your budget estimate.