Balancing speed, safety, and compliance

Bringing most hardware products into the market will eventually require some sort of safety or performance certification from a third party. But while you’re still in the development phase, there may be a gray area you can operate in without proof of compliance with safety standards. Rapidly iterating your product while pushing certification down the line is a balancing act, but one that will allow you to enter the market as soon as possible and hit commercial scale.

Edward Chiang is co-founder and CEO at Moment Energy. In the fourth installment of our series in partnership with Unreasonable Impact, we sat down with him to discuss the right time to invest in certification and how to stay one step ahead of the costs and timelines involved.

A roadmap towards safety certification

1. Proving that your product works
Before you even start thinking about certification, your most urgent goal is proving your technology is actually feasible. Not just to the market, but to yourself, because you’re about to bank the next 10 years of your life on it. 

If it doesn’t work, this isn’t as bad an outcome as it seems: you might have spent four to six months, but the time wasn’t wasted – you’ve learned how to evaluate technologies and when to walk away. 

2. Get a clear picture of what certifications you need
Identify the certifications and safety requirements your product will need in its final state. What does perfect look like? And is there a midway point that might be acceptable in certain cases? 

This matters almost as much as whether your technology works, because the certification journey could also be an insurmountable barrier to entering the market. It's expensive and can take years, so you need to decide early on whether the juice is worth the squeeze. Your competitors can teach you a lot here, especially the ones who gave up.

3. Validate that the market wants the product
Next, you need to be sure the market for your product actually exists. Let customers define the problem and build the solution around it, not the reverse. 

Often, your earliest adopters will be the most desperate for a solution – they’ll be willing to pay a lot, and to take technological risk, if you promise to solve their problem. In hardware, iterations happen over months, not days, and these real-world deployments will give you the customer data you need to iterate as quickly and effectively as possible.

You’ll of course need to be uncompromising on safety when putting your product in the hands of these initial customers. But certification won’t be necessary or possible at this stage due to the money and time the process demands. 

4. Build with certification in mind
Throughout your development process, you should be doing your best to build the safest product possible. By this point, it should be certifiable, or close to certifiable, even if it doesn’t have that explicit stamp of approval. 

This approach is common in regenerative agriculture, for example, where farmers often operate organically without certification. They might lose out on the price premium that comes with being certified organic, but they also avoid the cost of going through that process.

5. Going mass market
Once you have the resources and a clear plan in place, go full steam ahead on certification. This is an absolute necessity to unlock big customers who would never deploy uncertified equipment, because it would make their sites or end product uninsurable. Fighting your way through certification might be hard, but it's also the most scalable path to real sales, rather than a string of one-off projects.

Avoiding costly certification mistakes

1. Get to the final product as soon as possible
Only go for certification once you’re sure your product is in its mature state. If you go for certification too early, when your product is still half-baked, you'll have to re-certify later, which could cost another $3–4 million and 2-4 years of your time.

2. Hire for the phase you're in
Early on, you’ll need to balance your limited resources with the necessity of putting together a nimble, fast-moving team so you can get your prototype in front of customers as quickly as possible.

But down the line, when you’re entering the process of certification, you may need a different team and a different mentality. Here, mistakes start to really matter because each one can add six months to the certification process.

Hiring people with the right experience is critical in hardware, and there will likely only be a handful of engineers with the right expertise for your sector. These hires will be very expensive, so you’ll need to budget for them. If you're an early-stage startup, plan for your next raise to happen right before you need to go through certification, and build that headcount into it.

3. Build a sufficient buffer
If this is your first time going through certification, double the time you think it’ll take. If you're experienced, add 25%.

4. Speak to someone who’s been there
Early on, talk to someone who's been through the same process. Ask how to avoid the traps and whether there are any shortcuts. 

This makes the most sense if, for instance, you’re trying to get approval for a new fertilizer in agriculture – you might have a new compound, but there’s still an established process for certification, so you can hire someone to help you navigate it. 

But if the concept is new and you’re creating a new market rather than disrupting an existing one, the rules will be brand new – perhaps no one has ever gotten this certification before. You’ll be figuring it out together with the testing labs, which will add a lot of lead time. 

However, even if your technology is novel, elements of it will likely overlap with existing technologies, and the methodology for testing it will remain the same. Other companies will have been through this type of certification, so it’s worth speaking to them.

5. Be prepared for testing costs
The larger your hardware device, the more expensive the certification process becomes. If you’re testing a small chip, and it breaks, you just build another one, which might cost $20 or $100. For battery systems, the lab has to light the entire system on fire. Every time you have to run that test – and you’ll run it multiple times – that means tens or hundreds of thousands of dollars leaving your bank account. This can be a hugely prohibitive cost for a startup and will need to be factored into your fundraising. 

6. Align your supply chains 
Your supplier network needs to move on the same timeline – you can't have one component on a six-month timeline and another on 12 days. Reliability is also key, because if a critical component is delayed, this can push back certification, extend your time to market, and force you to raise more capital than planned.

To get preferential pricing and timing, you might take a million-dollar risk and buy 20 units when you only actually need one for certification. Of course, if the design turns out to be wrong or the component isn't correct, you'll be stuck with 20 useless units. Weighing up this type of risk is part of the reality of being a founder in hardware. 

Edward Chiang is the CEO and co-founder of Moment Energy, a company dedicated to providing universal access to clean and reliable energy by upcycling retired electric vehicle batteries.

Beyond his entrepreneurial pursuits, Edward is a member of the Royal Canadian Air Force as a flight instructor, having trained over 15 pilots in active service and civilian commercial aviation. This experience is a key driver behind Moment Energy's focus on developing energy storage for military applications, as demonstrated by their deployment with the Canadian Department of Defense. Edward envisions Moment Energy playing a crucial role in protecting critical national infrastructure, such as hospitals and airports, by developing reliable and resilient battery energy storage solutions in America.

Prior to co-founding Moment Energy, Edward gained valuable experience at the Canadian Nuclear Laboratories. There, he contributed to the development of advanced nuclear fuel technologies and worked on critical anti-terrorism technologies aimed at preventing the theft of nuclear material.