Hardware. Isn’t there just an app for that? Such is the life of the hardware entrepreneur. When most people ask me how a project is going they say, “How’s your app coming along?” There are so many software developers; app developers, web developers and anything-developers making digital products that people are forgetting the electronics that make it all possible! We constantly underestimate how much effort goes into that tiny piece of cell phone magic you’re probably reading this on.
Now, I’m not belittling software developers at all, their work is astounding and worthy of a mountain of praise. They are, however, just one piece of the puzzle that encompasses our digital lives. So why are so few people trying to create hardware products? I believe it’s entirely due to the high barriers of entry. Here are some of the challenges we faced when developing our product DrinkMate which we find especially unique to hardware development
1. Microcontrollers
Everyone has something with a screen on it – just bust out that laptop, phone, or tablet and start coding away. There are countless tutorials online to help you through any and all programming questions. Hardware products, however, and specifically the microcontrollers that control most of them, are a bit trickier. There are still plentiful resources online, but many of the tips and tricks for creating a manufacturable and mass-produceable product are kept as trade secrets.
Tip: Use a brand of microcontroller from a company/manufacturer that has invested in tutorials and example firmware for its products. These take most of the hard work out of the firmware coding so you can concentrate on, say, circuit design. Atmel is especially good at this.
What we learned: Forums are your best friend! Ask questions as you go along to see if anyone has already solved your problem before you invest too much time. This way you can focus on the truly difficult obstacles.
2. Product Quality and Consistency
When you attempt to create a new digital product like a website or app, the final product is exactly the same every time. Yes, there is always individual and platform customization, but the underlying code is guaranteed to remain unchanged. With hardware, quality control becomes a much more complicated issue. All of a sudden you have a large number of different real-world components that all are at risk of either failure or improper installation. If just one of these “cogs in the machine” fails, the entire product will be a hunk of useless junk, which is often more costly to repair than to replace.
Tip: Samples, samples, samples! PCB (printed circuit board) samples from manufactures are easy, but don’t get overconfident! There’s a difference between hand-made samples and machine-made production line final products, so you need to make sure you have a final testing procedure in place. Make sure to get enclosure samples as well. It is much easier to fix problems prior to a product launch than try to recover and repair post-launch. Take your sample and drop it, kick it, touch it, see what makes it fail or not fail. If there is a particularly weak design characteristic, talk to your manufacturer. Many times the manufacturer will offer to make the design change or potentially have ideas for how to fix the problem.
What we learned: Our samples had certain seemingly minor components switched with inferior brands. It wouldn’t make a difference for most products, but for ours it caused significant quality problems.
3. Costs
Digital products have bugs and glitches all the time, but they are always solved by software updates. The fix takes time and money, but the implementation is free. With hardware, any error with, say, a plastic injection mold not only requires time and money to redesign, but a massive fixed cost of mold reproduction. We’ve found a good rule of thumb is that every small plastic piece in your product will cost around $5,000 in mold production costs alone (excludes design costs).
Tip: Crowdfunding! Hardware products don’t have the luxury of a digital soft launch, it’s just too expensive to build anything in small quantities (less than 1,000). Put together a well-planned campaign and launch on a crowdfunding website such as Kickstarter or Indiegogo. This allows you to see if there’s a demand for the product you’re trying to produce. Additionally, there are now companies founded by hardware product experts who offer pre-crowdfunding manufacturing plan reviewing, advice and certifications such as Dragon Innovation.
What we learned: Make sure you have your product 100% designed before crowdfunding. Seriously – every single detail no matter how small. You will not have a spare second to do anything other than logistics once you launch!
4. Packaging
Create packaging for your product that’s based on your target market. If you want your product to be sold in convenience stores next to lottery tickets, package it in a plastic anti-theft blister pack. If you’re trying to get luxury points, package it in a gift box like you’d see in an Apple store.
Tip: Nobody saves blister pack plastic, but many people hold onto nice gift boxes out of sentimental value (i.e., “too nice to throw away”) or for reuse. You know you still have the box your cell phone came in lying around somewhere…
What we learned: Have your manufacturer prepare quotes for both “luxury” and “retail” packaging options. Your distributors and retailers will have different needs.
5. Shipping, Handling, and Storage
Ecommerce order fulfillment services do cost a pretty penny, but they can take all the pain out of actually delivering your product. Amazon is the leader in this, but they are very expensive. There are other options such as Fulfillrite. Make sure they handle product returns!
Tip: Collaborating with a smaller company like Fulfillrite was an invaluable experience. They helped us through every step of the fulfillment process that led to a fully successful product launch with no shipping problems at all. We use them to this day and they’ll be scaling with us as we grow our company.
What we learned: Direct API interfaces into your product’s website will save an incredible amount of time and make sure your product gets shipped ASAP.
6. Lead Times
Anything hardware-related has a lead-time, either to manufacture or ship. If it’s being custom-made, this time usually quadruples at a bare minimum.
Tip: Adopt the policy of under-promising and over-delivering. Most people attempt to meet a certain ship date, but never consider the option of shipping (or simply being ready) early. That means their options are either ship on time or ship late. Most of the time it ends up being late. It’s important to consider that delays happen and extra time needs to be factored in to any promised delivery dates.
What we learned: Paperwork delays add up. This can range from order processing to bank transfer holds. Always add a “buffer” to your anticipated lead times to account for the logistics of getting the order started in the first place.
Last Words of Advice
Always be ready to learn. There is no perfect way to deliver a hardware product. There is no “compile” or “upload” button. Keep an open mind and always seek advice from others who have gone through this process before. Prepare backup and contingency plans based on the experiences of you and others. This way, should the unexpected happen, you’ll at least feel like you already have a path towards a solution or have the resources to reach out and ask for help.
Good luck and have fun along the way!
Shaun Masavage is an engineer, entrepreneur and the inventor of DrinkMate, the world’s smallest portable breathalyzer which plugs directly into your Android phone.
Use a brand of microcontroller from a company/manufacturer that has invested in tutorials and example firmware for its products. These take most of the hard work out of the firmware coding so you can concentrate on, say, circuit design.
