When designing connected hardware products, there are many ways in which a project can go sideways and end up way over budget before the product is completed or in a manufacturable state. We’ve seen this happen countless times and it’s never one person’s fault, rather a lack of understanding of the full product design process and a lack of the experience necessary to avoid these cost surprises.
After years of seeing companies head down the wrong path, we’ve come up with a short list of considerations for product engineers and managers to take into account when designing, or hiring a development team to create, a connected hardware product. This list includes:
Define your product specifications early. Include unit pricing at either retail or wholesale level with a breakout for the enclosure.
Make your environmental and safety definitions accurate but restrained. For example, if you don’t need a fully submersible waterproof solution and a showerproof solution will work, use the latter.
If a limited budget will drive choices, the order of steps taken is very important. Do early PCB architecture, thermal assessments and volume sizing before industrial design begins.
Find a team with comparable production quantity experience. Having experience with production methods equal to the quantities you expect to develop and sell will help to keep your product development focused on solutions that meet your goals and won’t trigger inappropriate manufacturing or tooling costs. If you want an app along with the product, find a team that can do both physical and digital design in order to make it more cohesive.
Specifications – Define Early, Comprehensively yet Keep the Product Focused
There are many reasons to define your product specifications early. These include things like product components, price targets, user demographics, product volumes, operating life. By defining these early in the development process, a product designer can advise you on how to hit your targets, as well as point out things to avoid. For example, if you want to hit a cost target of $10 for plastic parts in your product, you may have to give up some of the aesthetic options like over-molding or an unnecessarily complex part count. To get a part that is manufacturable for your price, you may need to accept a simpler design that reduces the number of injection molds or slides (moving parts in the tool) or minimizes additional steps in the production process.
Another thing to keep in mind is precision in your specifications. By including all components, such as the number of PCBs, displays, LEDs, or wifi chips, preliminary costing ranges will be more accurate. The requirements of these parts should also coincide with a review of any compliance or regulatory requirements that your product will need to meet for its market, whether this is driven by certifications, safety standards or your competition.
The countries you plan to sell your product into will also drive development costs. If you choose only the Americas, there are different safety submissions and requirements than if you sell into Europe. Making sure you’re hitting all desired country certifications at from the start will be a lot less costly than addressing an additional set of certification requirements later in the development process. If you keep in mind all the countries you want to sell into, now and in the future, it can help to determine which components you should use at the outset. For example, using a power supply that is already UL rated will prevent any power-supply compliance issues in the future.
The Lily Drone: Famous for an ambitious design that was too costly to produce
Many companies forget that it is advisable to keep product specifications restrained. Water repellency, for example, requires much development work and has high manufacturing costs. Poured water, blown rain and full repellency all have significantly different technical requirements, so it might make more sense to just make a product rainproof and maybe waterproof it for the next generation. Or it may turn out that the market never really needed the product to be waterproof and customers didn’t want to pay for it. One quote we live by is Reid Hoffman’s “If your product is perfect in the first iteration, you took too long to get to market.” For multinational companies with millions of dollars to spend on development this may not be the case, but for a small company it’s sometimes more important just to get the product out and show customers the value you provide, rather than perfect everything about it. The first iteration can have some compromises that you can upgrade or change in the second generation, but in our experience companies often impede their success by setting their sights too high on their first product launch.
The Order of Steps can Make or Break Development Costs
The order of development is also a factor in risk mitigation and ultimately product development cost. If there’s a functional requirement which defines a set of components and these components have both a physical size as well as a set of performance drivers – for example they don’t work well at high temperatures – and there are power, longevity or servicing limitations which mandate no active cooling (i.e. a fan), then these design restrictions must be defined and assessed before industrial design work or formal enclosure engineering can begin.
For example, we once worked on an outdoor wireless system which needed to operate from sea level to 18,000 feet at temperatures of 55C without a fan. The mechanical engineering component layouts and initial thermal assessment work were done at the start of the program. This back-and-forth optimization between component architecture and the thermal engineering work led to sizing, heat sink plate location and a layout that allowed industrial design to begin accurately. Early concepts reviewed by our client didn’t need to be reworked or changed because the thermal and radio frequency performance engineering required a new size or architectural layout. It saved a lot of time and costly reworking.
Early architecture studies can also point to other issues. Just by having a mockup you can see things, for example a piece of sheet metal that might block an antenna and create a dead zone in connectivity. In one of our connected product development programs, we found exactly this and had to optimize the antenna location to prevent connectivity interference by the sheet metal within the device. It was these early architecture studies that warned us of this before we got too far into the project.
Map Design Experience to Your Product Goals
Finding a team with the relevant work experience can help “keep you out of the weeds” in product development. If your product includes an app, find a team that has designed apps before and knows how to design them in a way that engages trial users and turns them into buying customers. Having the app and hardware product design team together under one roof can bridge the gap between the physical and digital worlds and leave your users with a more memorable product experience.
A product development firm’s experience in production methods should also be in line with your product quantity goal. For example, a manufacturing goal of 200 products annually is very different from 100,000 monthly – and should drive very different design team choices. Experience in high-volume plastic housings does not make a firm an expert in brake-formed sheet metal or vice versa. If the right designers are involved early on, they can help to guide natural choices which lead to great design without the pitfalls of difficult or inappropriately expensive manufacturing methods.
Bringing the manufacturer (sometimes called the contract manufacturer or CM) in early in the development will also save money and often months of production startup time. Coordinating with the CM early means they can begin planning for tooling, training and material/component preparation while the development is still happening, thus gaining months of preparation time. Once, during an early design review, a contract manufacturer suggested that if we could flip a PCB over within the enclosure, they could do in-process testing (test the function of the PCB while building the product) thus reducing assembly time, which meant cost savings for the client. For us that was ‘free for the asking’ in the process at the time but would have been a substantial and expensive change later.
The Right Stuff
There is no perfect product development formula where you can input the product you want to make and get a list of steps you should take. Every project varies greatly, and finding partners or a team with the right experience can save you a lot of trouble in the long run. Many times, people have come to us after using a design firm that only does creative work with the complaint that none of the parts were manufacturable. This happens because many firms don’t have the engineering and manufacturing experience required to bring a product to production. Changes late in the development process can be very expensive, as they may mean that parts need to be re-engineered and Design for Manufacturing (DFM) needs to be reevaluated.
Being aware of such things can help save time and money in the long run, and save a project from painful patches and fixes down the line. Giving clear specifications is one of the first things we require when starting a new product design, because over the past 34 years we’ve seen it go wrong too many times without them.
