Nothing is ever designed from scratch
Carry-over engineering, and why it dominates real product development.
Nothing is ever really designed from scratch because almost every new product is built around parts, tools and processes that already exist and already work, and reusing what is already proven is nearly always cheaper and faster than redesigning it, so real product development spends most of its effort deciding what can be carried over and saves fresh invention for the rest.
Renovating a kitchen around the pipes
Renovating a kitchen around existing plumbing and wiring, instead of ripping the house back to bare walls, shows the logic in a setting most people know. The pipes and cables already running through the walls work, and moving them costs far more, in demolition, in disruption, and in the risk of damaging something nearby, than designing the new layout to work with where they already are. A renovation that insisted on freshly routed plumbing for every fitting would take longer, cost more, and very likely still end up compromising on something else to make the budget work. One that accepts the existing services as a fixed starting point and designs intelligently around them usually reaches a better kitchen, sooner, for less money.
What an existing part already carries
A clean-sheet design, with every part chosen purely for what would be theoretically best in isolation, sounds appealing but rarely survives a real budget or a real schedule. Every part designed fresh also has to be tooled fresh, tested fresh and proven reliable fresh, and all three steps cost time and money however good the new part turns out to be. An existing part arrives already carrying the value of every hour spent validating it, every failure mode already found and fixed, and every supplier relationship and tooling investment that exists to keep making it.
A designer choosing between a new part built to a slightly better specification and an existing part that is merely good enough is therefore often choosing between a large, certain cost paid now and a small, uncertain cost paid later. The existing part wins that choice more often than a purely theoretical view of engineering would predict, in the same way that the old pipe run wins against the ideal one in the kitchen.
Where the engineering judgement goes
An experienced product engineer starts a new design by asking which existing parts, tools and processes could plausibly do the job, and reserves fresh design effort for the places where nothing existing comes close. That moves the real skill somewhere less obvious. Deciding whether an existing part is good enough for a new application, understanding where its limits are and whether the new use pushes past them, is a different and often harder judgement than specifying an ideal part from a blank page, and it takes up most of real product development.
A new product's bill of materials, the full list of every part number that goes into it, shows this balance plainly once it is laid out line by line. Most of the lines usually trace back to something the company was already making or buying before the project began, and only a minority represent anything designed fresh for this product. Going through that list and marking which lines are new is a quick, honest way to check whether the design has spent its effort where it was needed. A list with an unusually high share of brand-new part numbers often belongs to a design that redesigned more than its schedule or budget could justify.
Spending the budget on what is new
A team that treats every new product as an opportunity to redesign everything tends to spend its engineering budget re-solving problems that were already solved, while the parts that needed fresh thinking get less attention because the money ran out before reaching them. Recognising which parts of a design are novel and which are variations on something proven is one of the first judgements a new engineer has to learn to make well. Getting it wrong in either direction, over-trusting a carried-over part that cannot do the new job, or redesigning a part that never needed to change, wastes the same limited pool of time and money the novel parts need.
When the proof does not travel with the part
Carry-over stops being the right instinct once an existing part is being stretched to cover a load, an environment or a duty cycle it was never validated against. The value of reusing a proven part comes from the fact that it has already been proven, and that proof covers only the conditions it was tested under. A bracket validated in a dry cabin may corrode in a wheel arch, and a motor proven at one duty cycle may overheat when run continuously. A part carried over into a new context needs its own validation for that context, even if the part itself is unchanged, and skipping that step in the name of reuse turns a sound economic decision into an untested assumption dressed up as a proven component. Going back to the kitchen, the old pipes are a sensible thing to keep until the new layout asks them to carry a boiler's worth of hot water they were never sized for.