Blog

How can I reduce the manufacturing costs of my existing product without sacrificing quality?

Work from landed cost per unit, not from the quoted unit price. Where the money actually sits in a product that is already in production, and what it costs to move each lever.

October 6, 2026

Work from landed cost per unit, not from the quoted unit price. Ask your manufacturer to break the price down into material, cycle time, scrap, assembly, packaging and freight, then model what a design change, an extra cavity or a different call-off pattern does to each line. On a product that is already in production, the largest savings almost always sit in wall thickness, part count and cycle time, and only after that in a different supplier. Negotiating a price whose build-up you cannot see rarely moves more than a few percent, and the few percent usually come out of something you did not intend to give up.

Where do the costs in my product actually sit?

In six lines, and they do not respond at the same speed. Ask for this breakdown from whoever makes your product today. The response itself is informative, because a manufacturer who cannot produce it is not steering on cost either.

| Cost line | What drives it | The lever | How fast the effect lands | | --- | --- | --- | --- | | Material | Shot weight, polymer grade, recycled content | Wall thickness, ribbing instead of solid sections, grade choice | First production run after the change | | Cycle time | Cooling time, cavity count, machine size | Wall thickness, cooling layout, hot runner, more cavities | 4 to 10 weeks, after tooling work | | Scrap | Process window, material behaviour, inspection criteria | Realistic tolerances on non critical features | Within a few runs | | Assembly | Part count and number of operations | Combining parts, snap fits instead of screws | After the design change | | Packaging | Volume per unit, protection requirements | Nesting, bulk packing instead of retail packing | Immediately | | Freight and duty | Volume, weight, route, incoterm, tariff classification | Shorter chain, smaller pack, verified classification | From the next shipment |

The first two lines are usually the biggest, and both hang on the same variable: wall thickness. Thinner means less material and a shorter cooling time at once. That is also where the trade off sits, because thinner means less stiffness and a narrower process window.

What does a design change deliver, and when is it worth it?

Usually more than procurement can find, because a design decision travels with the product to every supplier and every future order. Design engineering literature has cited the same rule of thumb since the 1990s: 70 to 80% of a product's cost is committed once the design is fixed (Kamal et al., 1995, among others). On an existing product that does not mean you are too late. It means the room you still have is mostly in the drawing, not in the quote.

In rough order of return per euro invested:

1. Even out the wall thickness and reduce it where the function allows. Hits material, cooling time and warpage at once. 2. Combine parts. Every part you remove takes a purchase line, an assembly operation and an inspection step with it. 3. Revisit tolerances. A dimension that does not mate with anything does not need to be held to a hundredth. 4. Add cavities or a hot runner. No change to the product, only to how the series is set up. 5. Material grade and recycled content. Several polymers accept a meaningful share of recyclate, with a trade off in colour consistency and sometimes impact strength. On a hidden component that is a cheap trade.

One example from a consumer goods client running roughly 120,000 units a year: the retail pack was designed around a product that had since become slightly shorter. Re nesting the pack put 18 more units in the same carton, which changed nothing about the product and reduced freight cost per unit on every shipment from then on. Not every saving needs an engineering change order.

Should I renegotiate or move to another supplier?

Renegotiate first, unless the current manufacturer structurally cannot deliver what you need. A move costs three to six months of sampling, first article inspection and running two sources in parallel, and during that window you pay twice in attention. Renegotiating costs a meeting and an argument.

The argument is the condition. With a cost build up in front of you the conversation is about cycle time and scrap rate. Without it the conversation is about a discount, and a discount gets earned back somewhere you cannot see.

Moving does pay when the region no longer fits the product. Eurostat put hourly labour costs across the EU in 2025 between roughly EUR 12 and EUR 57 for the economy as a whole, the Netherlands at EUR 47.9 and Romania at EUR 13.6. On a labour intensive product that gap decides the case. On a fully automated moulded part with one manual operation it barely registers, and the freight line matters more: Drewry's World Container Index for Shanghai to Rotterdam moved between roughly USD 2,800 and USD 4,900 per 40ft container during 2026. On a bulky product a lower unit price can disappear inside that swing.

What does it cost to lower the cost?

Anywhere from a mould insert to a full new tool, and the price is set by cavity count, side actions, steel grade and the volume you design for. That is why a tooling price on its own tells you little: ask for it alongside the piece price and work out the payback. Modifying an existing mould usually sits far below a new one, because only the insert is remade.

Model it with two numbers. Take a series of 60,000 units a year at EUR 1.80 per unit. A wall thickness change shortens cooling time and takes EUR 0.22 off the unit price. The inserts cost EUR 7,500. Break even lands at just over 34,000 units, so within seven months, and after that it is EUR 13,200 a year for as long as the product sells.

Run the same sum against a new EUR 45,000 tool and you are at just over three years, which turns it into a question about the product's remaining life rather than about manufacturing.

How do I protect quality while the cost comes down?

By defining, before anything changes, what quality means for this product in measurable terms. Tolerances on the dimensions that genuinely matter, cosmetic requirements per surface, a functional test with a pass limit, and a scrap level you accept. That is your baseline.

After that, every change follows the same routine: samples from the modified tool, a first article inspection with a dimensional report, and a comparison against the baseline. Production restarts only when that is signed off, including for a tenth of a millimetre of wall, because those are exactly the changes that shift the process window.

Be straight about what you are giving up. Recycled content costs colour consistency. A thinner wall costs stiffness. A faster cycle costs process margin. None of those are arguments against reducing cost. They are the items you weigh deliberately instead of discovering through a field complaint.

Production Heroes takes products from design to series production, from feasibility and cost price to sourcing and supply chain, and supplies moulds and tooling to manufacturers across Europe, from the Netherlands and its own presence in China.

Sources: Eurostat (March 2026); Drewry World Container Index, Shanghai to Rotterdam, 2026; design engineering literature on cost committed at the design stage (Kamal et al., 1995, among others); tooling ranges from our own project history.

Frequently asked questions

How much can I realistically take out of a product that is already in production?
It depends on whether the product has ever been costed properly. If nobody has reviewed wall thickness, part count and cavity layout since launch, there is usually room. If it has been through a cost programme twice, assume the remaining gains sit in freight, packaging and order size.
What does new tooling cost, and when does it pay back?
It varies widely, driven by cavity count, side actions, steel grade and the volume you design for. Ask for the tooling price and the piece price together, because either one on its own tells you little. Payback is the tooling cost divided by the saving per unit, and a mould insert is a fraction of a full tool.
Will a lower unit price show up as a lower landed cost?
Not automatically. Freight, duty, packaging volume, payment terms and scrap all sit between the two. Drewry put Shanghai to Rotterdam between roughly USD 2,800 and USD 4,900 per 40ft container during 2026, which can erase a unit price gain on a bulky product.
Should I change the design or change the supplier first?
Design first. A design change follows the product to every supplier and every future order, while a supplier change lowers the price once and costs three to six months of sampling and qualification. You also get a cleaner specification to quote against.
How long does a cost reduction project take end to end?
Count on four to ten weeks for a mould modification plus sampling and first article inspection, and three to six months if a new supplier is involved and you run both sources in parallel. Packaging and freight changes can land within one shipment.