In product development there’s an old adage:

“Fast, cheap, good: you can only pick two.”

The idea is that for any of these three attributes, increasing one has negative implications for one or both of the others. If we want fast development, we need to throw more money at it and/or cut quality (e.g., fewer features or less reliable). If we want cheap development, we must spend more time in development to whittle out costs and/or create a lower quality product. And so forth. The very best we can hope for is to do better on two of these at the expense of a third.

It turns out that the old saw is sometimes true, but often not, particularly if we think a bit out of the box. Let’s take a look at a real-world example…

Bluetooth: DIY vs. Off-The-Shelf

Suppose that our product needs to support Bluetooth. We can design a circuit using a Bluetooth chip, some discrete components and an antenna, lay out the PC board, have the boards fabricated and assembled, then test ‘em. If we find issues when we test — and we will — we’ll make some tweaks to the design and then another round of PC Board layout, fabrication, assembly, and testing. And there’s the detailed design documentation that that’s needed, particularly if it’s for a medical device. And then third-party testing for FCC (and possibly other) certification as an intentional radiator, that costs $10k – $20k — and there’s a distinct risk of failing and requiring a more tweaks. And we need to develop and install factory test systems, train the manufacturing folks, and there’s the hassle and cost of ironing out the inevitable hiccups and failures that happen when we first start to manufacture product, and additionally as systems start getting used in the field; perhaps these lead to product failures and/or redesign cycles.

A lot of work, and substantial expense. I’ve not done a detailed analysis, but my rough guess is that this Bluetooth development scenario will end up with fixed costs of no less than $100,000, and take at least three months until the design is ready for release to manufacturing.

To be optimistic, we’ll assume no additional costs from manufacturing issues, field failures, and so forth.

Now… what if, instead, we used off-the-shelf Bluetooth modules? We can buy perfectly good off-the-shelf modules (e.g., an ESP32-C3-WROOM) for two bucks. They just work, and have FCC certification to boot.

So comparing the two scenarios, make vs. buy…

Development cost (NRE)	Development Time	Unit Cost
$100,000	3 months	$1.50
$0	0	$2.00

If we DIY we spend $100,000 in fixed costs to save 50 cents per unit; we’d need to build 200,000 units to start seeing a cost advantage, and that’s before accounting for failures during and after assembly. And this is a pretty optimistic scenario — I’d wager that the DIY scenario would actually end up at twice the cost and twice the time I’ve guesstimated.

Did We Do It?

So compared to DIY, buying off the shelf is…

• Cheaper? Almost certainly!
• Faster? Definitely!
• Better? Almost certainly! Off-the-shelf products have had years for manufacturers to iron out bugs, improve reliability, and add features that customers find helpful.

For most applications (pretty much anything besides smart phones, watches and other consumer products), buying Bluetooth modules vs. designing from scratch is faster, cheaper, and better.

All three!

“Ha!” you might protest, “it’s obvious that buying is better than developing when it comes to Bluetooth! This is an unusual circumstance.”

Many Opportunities for Big Wins

But is it obvious or unusual?

In the past three years I’ve twice been engaged to help out on ongoing product development efforts for products that supported Bluetooth. Both were medical devices that would sell under 10,000 per year (at least for the first few years). In both instances, the developers were designing Bluetooth circuitry rather than using off-the-shelf modules. And in both instances, budgets and timelines were far from initial commitments to stakeholders.

So maybe using modules is obvious in retrospect; but not in prospect.

And supporting Bluetooth is just one instance of many where being a little clever can get us the unambiguous win of faster, cheaper, AND better. In my experience virtually every project can benefit from practices which result in faster, cheaper, AND better development. In total, these practices can substantially reduce project timelines and costs, typically with the added benefits of increasing predictability and reducing production costs. Huge wins!

In my book, Prototype to Product, I demonstrate many of these best practices; the general theme is that smart planning leads to development efforts that produce better products, faster, and at a lower total cost. (And with fewer bad surprises, frazzled engineers, and grumpy stakeholders).

As the US Marine’s say with appropriate pithiness,

“Proper Planning and Preparation Prevents Piss Poor Performance.”

(This is known as the 7 Ps)

Prototype to Product is a concise guide on how to effectively plan, prepare (and execute) in the less-lethal world of product development. It describes dozens of actionable best practices that save time and money, and increase quality. You can pick up your copy here.