By the time the cost-reduction programme starts, the cost has already been decided.

A manufacturing business under margin pressure does something predictable and reasonable. It launches a cost-down programme. Procurement renegotiates. Operations attacks scrap and changeover time. Engineering is asked for a value-engineering pass. Everyone works hard, and the result is a few per cent — real, worth having, and far short of what was needed.

Meanwhile, three engineers in a design office are making decisions about a new product that will determine the great majority of its manufacturing cost for the next decade. They are doing so under schedule pressure, with limited manufacturing input, against requirements written by someone else, and their decisions will not be reviewed by anyone with commercial authority until the product is in production and the cost is fixed.

This is not an engineering problem. It is a capital allocation problem that has been delegated, unexamined, to the point in the organisation with the least commercial visibility and the greatest leverage.

The Strategic Context

Two curves move in opposite directions across a product's life.

The first is committed cost. Every design decision — material selection, tolerance, number of parts, choice of joining method, whether a feature can be formed or must be machined — commits future expenditure. This curve rises steeply and early, because the decisions that determine cost are made at the front.

The second is remaining influence. As commitments accumulate, the scope to change cost without re-engineering falls. Once tooling is cut, the design is effectively fixed; once production is stable and customers are qualified, changing it invites re-validation cost and risk that usually exceed the saving.

The consequence is structural. At the point where the organisation has spent almost nothing, it has decided almost everything. At the point where it is spending heavily, it can change almost nothing.

The widely quoted figure is that around eighty per cent of manufacturing cost is committed by the end of the design phase. The precise proportion varies by industry, product complexity and how design is defined, and it should be treated as an order-of-magnitude heuristic rather than a measured constant. [FACT CHECK REQUIRED — the 80% figure is a widely repeated design-management heuristic; the primary source and its measurement basis should be confirmed before citation.] The direction of the relationship, however, is not in serious dispute, and it is the direction that carries the management implication.

What Leaders Commonly Misread

That cost reduction is an operations activity. Operations can influence the cost the design permits. It cannot influence the cost the design requires. A part specified with a tolerance tighter than the function needs will cost more to make in every unit, forever, and no amount of operational excellence recovers it.

That value engineering after launch is equivalent to designing for cost. It is not. Post-launch changes carry re-validation, re-qualification, tooling modification and sometimes customer approval — costs that consume much of the saving and consume engineering capacity that could have been applied to the next product.

That the design phase is the expensive phase to extend. Design is usually the cheapest phase in absolute spend and the most expensive in commitment. Compressing it to protect a launch date is one of the most reliably value-destroying decisions available to a manufacturing business, and it is almost never framed as an investment decision.

That engineering knows the cost consequences. Frequently it does not, not from lack of capability but from lack of feedback. If designers never see the manufacturing cost of their choices in a form they can act on, they will optimise for what they are measured on — function, weight, schedule — which is exactly what a rational person does.

Reframing the Issue

The useful reframe is to stop treating design decisions as technical and start treating them as capital allocation decisions with a long tail.

A tolerance decision is a commitment to a machining operation, an inspection regime, a scrap rate and a supplier capability, repeated across every unit for the life of the product. A decision to join two components by welding rather than fastening commits a process, a skill requirement, a quality risk and a repair path. These are not drawing details. They are recurring financial commitments made at a moment when their reversibility is highest and their visibility is lowest. [Related article: Tolerance Is a Commercial Choice]

Once framed this way, the governance question becomes obvious and uncomfortable: which of these commitments currently receives executive review, and at what threshold? In most organisations, a twenty-thousand-dollar discretionary spend requires approval, while a design decision committing several million dollars of lifetime manufacturing cost requires only a drawing sign-off by a peer.

Where the Leverage Actually Sits

Three decisions carry disproportionate cost consequence and are worth naming explicitly, because they are the ones that reward early attention.

Part count. Every part carries a procurement cost, an inventory cost, a handling cost, an assembly operation, a quality risk and a potential failure mode. Reductions in part count compound across all of these simultaneously, which is why consolidation usually outperforms unit-cost negotiation on the same components.

Tolerance and surface specification. Cost rises non-linearly as tolerance tightens, because at some point the requirement moves the part to a different process, a different machine class or a different supplier tier. The step changes matter more than the gradient. A designer who knows where the steps are can often hold function at materially lower cost.

Material and process pairing. Material selection determines available processes, and process determines tooling, cycle time and achievable geometry. Choosing a material without considering the process it implies is common and expensive.

None of these require sophisticated analysis. They require that someone who understands manufacturing economics is present and has standing when the decision is made.

Decision Framework

Four questions, applied at design review rather than at cost review.

1. What does this decision commit, per unit, over the expected volume? Not the design effort — the recurring cost. An order-of-magnitude estimate is sufficient; precision is not the objective, visibility is.

2. What is the function that requires it? For each tight tolerance, difficult feature or costly material, state the functional requirement it satisfies. Requirements that cannot be traced to function are candidates for relaxation, and there are usually several.

3. When does this become irreversible, and what would it cost to change it after that point? Establishing the point of no return changes the urgency of the review. [Related article: Strategy Is a System of Choices, Not a Document]

4. Who holds commercial accountability for this decision? If the answer is no one, the decision is being made without a commercial owner, which is the underlying condition this article describes.

From Strategy to Execution

Immediately. Require that manufacturing or cost engineering is present — with standing to object, not merely to comment — at design reviews for any product expected to exceed a defined lifetime volume or value. Presence without standing produces politeness, not challenge.

Over one to two quarters. Give designers cost feedback they can act on: a costed comparison for common features, tolerance bands and process choices, maintained by manufacturing and updated as capability changes. This is a modest piece of work with an unusually long payback, and it changes behaviour without changing anyone's objectives.

Over one to three years. Move the design gate into the investment governance framework. A product design review that commits significant lifetime cost should meet an evidence standard comparable to a capital request of the same magnitude. This will be resisted as bureaucracy, and the resistance is worth taking seriously — the objective is a proportionate gate on the few decisions with large commitment, not a review of every drawing. [Related article: Design Decisions Are Capital Allocation Decisions]

Signals to Monitor

  • Cost-down programmes with declining yield year on year. The accessible cost has been taken; what remains is committed in the design.
  • Engineering change requests concentrated in the first year after launch. Manufacturing reality was discovered after commitment rather than before.
  • Quotes from suppliers clustering above estimate on specific features. A signal that a specification has crossed a process step change.
  • Design schedules compressed while launch dates hold. Commitment quality is being traded for time, with the consequence deferred to production.
  • Scrap and rework concentrated on a small number of features. Those features are telling you where the specification exceeded process capability.
  • Manufacturing input arriving as a review comment rather than a design input. Too late to influence, early enough to be blamed.

Questions for the Leadership Team

  1. What proportion of our product cost is committed before the first tooling order, and have we ever measured it?
  2. When did a design decision last receive commercial review before it became irreversible?
  3. Do our designers see the manufacturing cost consequences of their choices in time to act on them?
  4. Which of our current specifications exist because of function, and which because of habit or an inherited standard?
  5. If we extended design by six weeks to reduce lifetime cost by five per cent, would our governance allow it — and who would have to agree?
  6. What is the commercial accountability for design decisions in this organisation, and is it held by anyone with authority over price and margin?

Closing Perspective

The difficulty here is one of sequence, not competence. The organisation applies its sharpest commercial attention at the moment the cost becomes visible — which is also the moment the cost becomes fixed. Everything before that is treated as technical work, governed technically, and reviewed by people who are not asked about money.

Moving commercial judgement earlier is neither expensive nor complicated. It requires putting manufacturing economics into the room where design decisions are made, giving it standing, and accepting that a slower front end buys a cheaper decade. The organisations that do this do not become better at cutting cost. They stop committing to it in the first place.