Distribution · Electronics

Nuts and Bolts of assembling and selling electronic devices

smithsa Experienced software leader looking for new opportunities

August 29th, 2019

I am thinking about developing and selling a small electronic device that would be based on a standard single board computer (something like a Raspberry Pi). I'm trying to figure out how to actually go about the steps and here's my best guess:

1. Procure all of the pieces (probably from China) and have them shipped to me in Canada.

2. Write my software onto SD cards.

3. Install the SD card into the single board computer.

4. Put the computer into a case.

5. Put all of the pieces (i.e. cased computer, cables, power supply, etc) into a box.

6. Ship all of the product boxes to Amazon in the US.

7. Have Amazon manage the distribution.

I know this is not the best way to do this especially since it requires everything to cross 2 international borders which I'm sure is slow and expensive.

How do I streamline this process? Can I get these things completely built in China? What sort of services can I use to keep costs down and quality up?

Paul Garcia marketing exec & business advisor

August 29th, 2019

Oh my, it's not so simple.

While the logistics of manufacturing and distribution will certainly be important elements of your plan, it's kind of cart before the horse right now.

Just as much with hard goods as with soft goods, your marketing strategy needs to lead product development. Make a list of your assumptions, test each assumption individually, compose your marketing strategy, validate each part of your marketing strategy to find your product-market fit. Once you know for sure (through your research/validation steps) that you have conceived something that people will pay money for and you know how to sell it to them, your marketing strategy steps will have defined what the product will actually physically do and where you need to sell it. These are no-cost research steps. Well, it costs you time doing your work to get answers but rarely does it cost you cash money.

Because these are physical goods, you will be combining some research on manufacturing options, but without a technical specification for your final product, something that won't be solidified until you've got your product-market fit validated, you won't be able to eliminate manufacturing options from your list very easily. I don't know why you assumed China would be your source for hardware. I can think of several reasons why you would not want that. And you're likely to discover that the hassle of logistics in handling Customs is going to make it worth the (small) price difference for domestic production. And if your product quality and value is sufficient, the validated price people would be willing to pay for your goods may be sufficient to accommodate a slightly higher cost of production. I can't say for sure, because computer hardware often has extremely thin margins, like 2-3% profit. So unless your device is not a commodity item, you may discover through your research that the volume of business required to make any substantial amount of money (profit) is not worth your time, or the market won't support the required volume.

If you want to maintain quality, I don't recommend China. The attitude in China is that cheaper is always better. And although I have manufactured devices in China, I specified the components so distinctly, that substitutions were impossible. My factory would tell me "We can save you $0.20 with this other type of wire" (on a $75 product) and I would tell them, "Do it my way." The difference in the end is that my devices lasted a decade, and those made with the "other wire" lasted 8 months.

My recommendation in electronics is that you get to fully understand the effect of every single component that goes into your circuit boards, down to the solder, resistors, even the screws to put it together. Hiring an engineer who can give you an extremely detailed specification might take longer, but it can be the difference between disposable goods and durable goods.

Think about the market for shoes. There was a time when a pair of shoes was supposed to last, would be re-soled, and they were made with care. Now people buy dozens of pairs of shoes which each wear out quickly and are thrown away. That was a conscious decision by the footwear industry to generate more revenue, making shoes less durable. Do you want a disposable product or a durable product? It will make a difference in how your manufacturing is done.

Regardless, when you get to distribution, you'll have figured out whether your actual customers are domestic, international, and how far international. There are numerous fulfillment options, but shipping across borders is not straightforward no matter whether you do it in bulk or one at a time. Using a fulfillment house that is fully integrated with Canadian Customs will be essential. You will want to have your products pre-approved with Customs so they don't need inspection with every shipment. And you'll have to figure out how the taxes are handled. Your fulfillment house should be doing this with or for you.

When considering having your software installed in another country, make sure you are confident that your intellectual property will be protected. China, while happy to install your software, may copy it without any payment to you, and start selling your system elsewhere using your software, with little ability for you to enforce ownership protections.

I could probably keep writing on this subject for another hour, but this is enough to get you thinking.

David Pariseau

August 29th, 2019

The space you are considering is already pretty crowded with companies that have solid expertise in what you are attempting. These can develop and field products at very aggressive pricing. They way the meet those prices is through efficiencies but also with very LARGE volumes, so it's a very expensive commitment.

So, you have to make sure you have an idea that will pan out and be commercially viable before you invest a LOT of time and money in such a venture.

If you have something novel and that you think is commercially viable it might be useful to break the idea down into parts and identify just what it is that's the heart of the product. Perhaps that idea could be implemented with a small daughter-board that would be compatible with existing hardware (Raspberry Pi , Arduino, etc.) That would significantly increase your prospects will lowering your costs. Such add-ons command MUCH higher margins and will support MUCH lower volume builds. Or, perhaps the idea could be implemented entirely in software to run on these platforms in which case then you don't need hardware at all, though you'll have to deal with licensing and protecting your product in a hobby comunity...


smithsa Experienced software leader looking for new opportunities

August 29th, 2019

Thanks for such a thorough answers. Here are my thoughts:

The business idea is really not this device. It's a whole range of services that will be accessible through the device. Could people access these services from their already owned hardware? Well, no, but that's a longer discussion.

I'm really not interested in designing hardware down to the component level (I'm a software, not hardware guy), so will be using an off the shelf SBC. Not sure which one right now.

Do I have product-market fit? Nooooo, I have an almost-there MVP-prototype. Will the ultimate product still consist of a very specifically configured small computer talking to something in the cloud providing services? Probably, but I'm open to something different.

As for cost/price, I'm really not concerned. Things will be very low margin on this device because I'll make it up by selling services. That's the real business.

So here's my problem: I've got this big honkin' empty space in my business plan about how we get these things into people's hands. I'm wayyyy out of my comfort zone here. I'm really just trying to figure out if this can be done without completely taking over every working minute of this project.