Getting more out of your R&D investment by lowering BOM costs.
It is a wonderful time to be a maker of products. Find an underserved niche and give the world a tangible solution. Famous movie quote, “If you build it, they will come.” The hard part becomes finding out who “they” are and how many of “it” they need. Good products will spread their wings organically and even find markets that were not considered in the planning stages.
Image credit Fast Bom
I can recall two different test fixtures that were built for internal consumption but found external customers who would use them in unintended ways out in the real world. In a case like this, having one single-source component on a board that is going viral can be a supply chain nightmare. Just like in school, success follows doing your homework.
Going from an FPGA to an ASIC can take numerous iterations of chips and/or boards that eat into the profit margin. It may be the only way to scale up, but moving the break-even point out by one quarter can be the difference between success and a participation ribbon. Plan ahead as best you can. You only get better at it by doing it.
We Keep Score With Money
Top line revenue in the billions is meaningless if the bottom line profit isn’t there. If your company is selling at a loss, then the design win is ultimately a loss. Reigning in these costs is next to impossible after the fact, so a thorough up-front BOM analysis is a must-have step in the design cycle. The old saying is true, it is not what you make but what you keep that matters. A big part of the fixed cost of goods sold is the materials.
For example, let’s look at the exotic and luxury car brands that are branching into the burgeoning SUV market. They are inspired by the benchmark Ford Explorer or perhaps the up-market Lexus RX, but the high-end manufacturers will sell a tiny fraction of the mainstream auto companies. Is the steel in a Range Rover three times stronger than the steel used in an everyday SUV? No, but it is priced as though it was. The brand’s rugged reputation from being on safari in those BBC documentaries gives it the cache to command that price.
It becomes a numbers game. Selling one-tenth the number of units means that each unit has to cover 10x the portion of the R&D expenses compared to the mass-produced vehicles. Couple this with higher overall R&D expenses to integrate the luxury/performance equipment, and soon it becomes clear they’re not simply profiting off of the snoot-factor. Crunching the numbers on the cost of goods sold (COGS) with the expected sales numbers informs the price point. Executing on the COGS and sales figures is where the rubber meets the road.
Upstart automakers do not usually implement model years because it takes longer than that to get their money back. Instead, they’ll build the same thing for as long as it sells with ongoing refinements regardless of the calendar. The early products are nothing but a money pit, and part of that is the higher cost of the materials along with the green-field engineering. A partner with adaptive tooling and a bigger list of vendors might have given a more lucrative outcome. A strong brand will open the door. A strong bottom line on the balance sheet will allow one to pass through that door to the next with minimal drama.
You can bet that everyone wants the lowest price on their materials, but not everyone has access to those deals. Volume speaks volumes! By that, I mean that components in short supply are often allocated based on how many were ordered by each customer. Once upon a time, I ran a stockroom for a brand new product segment called a Local Area Network. LANs are commonplace now, but back then compute power was quite precious. Our product was based on the (don’t laugh) insanely fast Intel 386 processor. We ordered 100s of them while IBM and others ordered tens of 1,000s.
I remember hand-carrying two tubes of processors from receiving inspection to my stockroom. That was about $21,000 (1988 dollars!) worth of world dominating power in my mitts. We needed–and ordered–about a million dollars in chips, but those two tubes were our allocation that first quarter. There is little you can do when riding the leading edge along with the major players. Taking a step back in technology is often a more sane approach.
The every-day components can also fit this criterion. You can order reels of one-micro-Farad capacitors just about anywhere. Getting them in the 0201 or smaller package will cost more and devour additional lead time. Clicking the filter for 0402 or larger package sizes will get your cost down and your velocity up. When everyone else wants that one thing, it makes sense to want something else. Do you really need the low ESR, high-temperature rating and 1% tolerance? Economize where you can and pony up when you must.
Too good to be true?
Sometimes, a bargain is to be looked at with a skeptical eye. If the majority of the vendors have the same price and lead time with one or two outliers that offer an extreme discount or immediate availability, you may be looking at salvaged or counterfeit parts. Even if it is a legit deal, it could signal that the part in question is being closed out and will go off of the market soon. There is usually a reason for a bargain. Cost reduction begins with the technical design stage and carries into the tactical build stage through the entire life of a product. It takes vigilance to stay on top of the supply chain. The Bill of Materials is the blueprint for that effort.
Two vendors are better than one and three make an insurance policy against the other two falling down. Too many approved vendors can create a headache as well. It is easier to maintain a relationship with a few close partners than to spread the orders out over many. If you can maintain three suppliers, you can keep them all honest. If I am being honest, we are lucky to find two vendors for every line item.
Let the internet do it.
The joy of B2B communications in the modern age is one of the reasons it is great to be a maker right now. Bill of Material analysis as a service can free up a lean company or solo inventor to do what they do best. I would rather tinker with the technical than tackle the tactical. (Say that out loud!) If you are like that, then services such as FindChips Pro can be a game changer. A good friend, Sophi Kravitz, advised me on this shortcut to success. In her words:
“FindChips Pro is a tool that provides complete visibility on your BOM or database of parts. You upload a file and FCP searches its distributor data to see how many of your parts are sourceable, which parts may be about to be end-of-life'd, and to assess fit-form-function options. The response is instantaneous and you'll know if your product is likely to make it through the supply chain without too much incident.”
Bringing it together
Material cost containment never really sleeps. The supply chain is too dynamic for sleep, so Product Lifecycle Management (PLM) is a thing you need to do.
Negotiating on the most significant cost drivers by agreeing to a set volume and delivery schedule is the best you can do. Only lock into an agreement if you are certain that you will need those parts at that price.
Make/buy decisions are some of the most important ones made on a project.
One missing item will prevent you from shipping so have a second source if possible.
Keep the Bill of Materials as short as possible by experimenting with depopulating a capacitor here and there to see if the product performs without the entire complement that the chip vendor lists on the data sheet. Count on them to be conservative.
You will want a good estimate of these costs as part of the initial market research.
The Bill of Materials serves as the nucleus for this data. It is a natural place to record the number and revision of relevant drawings. Links to the process docs and Source Control Drawings (SCD) will turn the mere spreadsheet into a gold mine of information.
Image source: ManEx - note the right column gives standard cost which is a key data point on the way to success.
Whether you work with distributors, individual vendors or us an Original Design Manufacturer, (ODM) getting a volume discount is how the big players stay big. Using parts that are already in your pipeline is advisable. Custom connectors can be part of the plan if your expected sales targets are high enough to amortize the upfront cost and lead time.
With the vast number of parts available off the shelf, an hour of surfing can probably get you into something off the shelf that meets your needs. This holds true with most things, but you need to make an informed decision based on the run rate. Is it a Ferrari for the few or a Mazda for the masses?
3D printed prototypes can ensure proper fit even if they would not pass the longevity or functional requirements of the final product. Knowing that everything fits together before committing to hard tooling and big component orders will help you sleep at night after a long day of what-if scenarios.
Having a competent team of designers in your stable will also help contain costs of the custom parts. These so-called “A” parts make up a big chunk of your B.O.M. costs. B and C are the labels for the bulk of the line items with C being the lowest cost. The make/buy decision is a crucial inflection point on the cost curve. The parts mix and the product launch go hand in hand in determining the final outcome. Take control of the COGS and let the market decide if you were right with your product’s features and release date. Good luck. We’re counting on you.
About the Author
John Burkhert Jr is a career PCB Designer experienced in Military, Telecom, Consumer Hardware and lately, the Automotive industry. Originally, an RF specialist -- compelled to flip the bit now and then to fill the need for high-speed digital design. John enjoys playing bass and racing bikes when he's not writing about or performing PCB layout. You can find John on LinkedIn.More Content by John Burkhert