When I started Neya, there were a few things I was pretty sure of. First, I wanted to maintain 100% control of the company. Contrary to lots of really good business advice, I didn’t bring on any co-founders. This was a risk, but one I was comfortable taking. Second, and related, my desire to keep 100% control of the company meant I couldn’t really go after any form of dilutive funding (i.e., venture capital). Finally, I had developed a broad base of experience and customer relationships at my previous companies that I wanted to be able to leverage.
This all led me to decide to stay in the government space and focus on defense applications of autonomy. Keep in mind that in 2009 (when I started Neya), commercial funding for robotics wasn’t anywhere close to what it is today. Between 1999 and 2009, there were approximately 100 companies started world-wide with some play in robotics according to index.co, a business information aggregator. This sounds like a lot, but many of these were motion control and component companies focused on industrial applications. The commercial explosion in autonomous and mobile robots hadn’t really begun. Compare that to today, where index.co lists over 700 companies in the robotics space founded since 2009.
The US Government, however, has actively funded research in robotics, autonomy, perception, and machine learning for over 30 years. Not all of it was far-reaching DARPA research either. All the Services had and continue have active programs in robotics R&D. They were also starting to see the value of robotics to the warfighter. The Foster-Miller Talon™ and iRobot (now Endeavor Robotics) Packbot™ robots, which were originally developed under the DARPA Tactical Mobile Robots program, had been in use in Afghanistan and Iraq since 2002.
So, when I started Neya, the choice was pretty easy for me. I’d continue what I know, and build a company focused on robotics R&D, funded through government contracts.
I’d note, though, that while using government R&D funding makes sense for companies that develop products and services that require upfront R&D, it’s not where you’d go if you have a new idea for a commodity product (i.e., what you might see on Shark Tank or Home Shopping Network). Think high tech and science. If your business idea fits in that really large mold, then it’s worthwhile considering federal funding.
Why This Was a Good Idea
There are a lot of advantages to starting your company using funding from the federal government.
The Government invests a lot in R&D. In 2018, total federal R&D outlay was about $140B. Total Venture funding in 2018 was on the order of $80-100B. Now, not all the federal R&D dollars are sent to new companies. Some of it is spent on internal R&D (R&D done at government agencies by government employees), and a substantial fraction is “Basic Research” which typically goes to universities (roughly $37B in 2018). “Applied research” and “development” are typically equivalent to seed rounds and A, B, C rounds – and the bulk of this goes to US companies. The upshot is that the government spends a lot of money on funding innovation at small and large companies. Furthermore, between all the major federal agencies (DoD, DoE, NASA, NSF, USDA, NIH, NIST, NOAA, and USGS), there is almost no area of research that isn’t covered by potential federal dollars.
The Government could buy your product. The government can be a pretty good market to be in. I won’t downplay the difficulty – it’s relatively easy to get the government to pay for R&D, but really hard to get them to buy products. This is the exact opposite of most large companies. If you have a product a company needs, they will buy it. If, however, you ask them for R&D dollars to help develop a concept, then that triggers all kinds of issues and it’s quite likely the other company will want substantial rights in what you develop using their money. Having said that, if you develop a product, on government funding, and the government decides that they want to buy it, then you’ve got a long-term high-value customer. Although we focused on R&D at Neya, there are multiple avenues for selling products to the government. The General Services Administration has some useful introductory material on this. There will be lots of consulting firms who are happy to help you for a fee – most of them are not worth engaging.
This funding is non-dilutive. This was, to me, the most important factor in deciding to use government funding to bootstrap Neya. When you receive funding to investigate an idea or develop a prototype, the government does not get equity in your company. You still own exactly what you owned before accepting the funding. There’s no notion of pre-money or post-money valuation. There’s no cap table. There’s no “can we have 2 board seats, preferred shares, and a ratchet clause”. But the funding isn’t completely without strings…
You (mostly) own your IP. Government contracting IP rules are complex and if you go down this route, you’ll do yourself a favor by getting expert advice. However, broadly, government contracting IP terms fall into three major categories:
Government Purpose Rights: The government has a royalty free non-exclusive perpetual license to use the IP they paid you to develop for government project / program purposes only. However, you have to understand that your IP could be released to your competitors, for them to use on specific government programs. In most cases, the recipient will be required to execute an NDA with the government or with you in order to get access.
Government Unrestricted Rights: The government has a royalty free non-exclusive perpetual license to use the IP they paid you to develop for any purpose whatsoever. They can decide it’s in the national interest for your competitor to have access to your IP for purely commercial uses. This is rare – but it’s something to be aware of
SBIR Data Rights: This is unique to small businesses that participate in the Small Business Innovation in Research (SBIR) program. SBIR rights place substantial restrictions on what the government can do with your IP, even for government purposes, for a period of 4-5 years from the completion of your contracted development work. This is the friendliest terms you’ll see from the government, but is restricted to SBIR contracts.
At Neya, most of the work we did ended up having GPR or SBIR data rights. This gave us a substantial advantage in that the government funded a huge amount of our R&D, but we had near exclusive commercial rights to it, which allowed us to leverage a lot of our past work for new commercial efforts.
Why This Could Have Been a Bad Idea
While there are a lot of positives to starting out as a government contractor, there are some definite risks you need to be aware of.
It's not a fast process. Every stereotype about the government being slow to work with is absolutely true when it comes to government contracting. The individuals you’ll work with are, in general, all competent, hardworking, and deeply caring folks. Many of them could be making a lot more money in the commercial sector but care deeply about the country and are patriots. Having said that, the system itself can be extremely slow. It’s not unusual for 6, 9, or 12 months to pass from the time you submit a proposal to the time a contract is issued. This is even true for SBIRs, where the people involved are much more attuned and sensitive to the cash flow issues of a new small business. There were more than a few times in the very early days of Neya where I knew that if a contract didn’t get issued by a particular date, we’d have trouble with payroll. So – if you go this route, consider the cash flow impact. Try to line up consulting or other commercial work to be able to pay bills.
It’s hard to talk to the end customer. This is pretty different than the commercial world. When you’re developing a commercial product, especially an app or other piece of software, it’s not unusual to iterate quickly, while getting continuous feedback through A/B testing or other methods to understand and improve product / market fit. That does not happen with government work. If you are doing R&D, in many cases, you’ll work with and through one of the major government labs that each service has. These labs are the ones who set research priorities – while taking into account the needs of the real end user (the service personnel you’re trying to build technology to help). They act as the voice of the customer, and can have a great deal of impact on the direction you take. However, some folks are better at being the voice of the soldier than others. So that’s something to be aware of – there will be times in your development cycle where you’ll have the opportunity to engage directly with potential end users of your technology. Grab those opportunities and learn as much as you can. An alternative approach, if you have the funding (or even better, can find the right co-founder) is to directly bring on retired or inactive service personnel / veterans who may have the direct experience you’ll need to make sure your product can really help the soldier.
Priorities change. Government R&D and product priorities change, usually with each administration. At the very lowest level, it may not affect you much. But as you start to work towards larger contracts – they typically need to be aligned with one or more major service priorities. As those service priorities change, programs change as well. You can be caught totally off-guard by this. At Neya, there was only one case where we had a program completely cancelled because the larger program that would ultimately require that technology was cancelled. It’s rare – but it happens. Luckily, we were big enough by then to be able to easily absorb that.
It impacts your company culture. This one is much more subtle. Government contracting rules can flow down into all aspects of your operations. This includes everything from accounting to employee timekeeping rules. Things like Total Time Accounting become important. Employees have to be trained to think about billable hours and making sure that they are charging their time to the correct contracts or overhead accounts. This can affect innovation. The idea that a motivated employee could just work extra hours “for free” to pull off a project, or create a new prototype ends up being difficult to execute – you have to make sure to have the appropriate charge number or IRAD account set up. It takes work to maintain a culture of innovation while balancing the (mandatory) rules you’ll have to follow. Be prepared for that.
Why We Were Successful
Given the potential substantial downsides of starting a government contracting R&D company, I went for it anyways. Partly because it’s what I knew how to do, so my risk was lower. Partly because, as I mentioned in the beginning, I didn’t want to give up any equity that comes with an investor, or control that comes with a co-founder.
We were able to be successful because we attracted extremely talented employees, many who had prior experience working with large and small government contracting companies and understood what went with that. We were successful because we took real advantage of programs like SBIR to develop technologies that we then leveraged into other commercial (higher margin) work. We were successful because we entered the robotics and unmanned systems space early, and established ourselves as leaders in some very difficult technical areas related to off-road autonomy.
Ultimately, we pulled it off because we created a good business. We pay attention to our customers; we develop technology and products that are needed; we take care of our employees. While being a government contractor is certainly different than running a venture-backed (or bootstrapped) company that’s B2B or B2C, the key tenets of building a solid business aren’t any different.
Where to Start
So, if you want to do this yourself, one of the best resources for small businesses interested in government contracting is the Small Business in Innovation Research (SBIR) program. Basically, SBIRs are government contracts specifically tailored to small businesses, to allow them to develop innovative solutions to a huge range of technical problems faced by nearly all branches of the federal government. This includes the Department of Defense, the National Science Foundation, DHS, DoT, and others. Each agency with an annual R&D budget of over $100M is required to allocate 3.2% of that budget to the SBIR program.
There are a number of consultants who will gladly charge you more than you’ll make on an SBIR contract to help you win one. Sometimes they can be valuable (at Neya, we used one when we were going after an NSF SBIR, since we had no experience with that). Often, they’ll tell you things that I’m happy to share here for free. If you have any questions, please feel free to reach out!