The handover is the risk, not the design. A working prototype proves the product performs; it says nothing about shrinkage, ejection, cycle time or repeatability, and those four decide what unit 1,000 looks like. What you need is a partner who revises the design for manufacture before any steel is cut, builds tooling against a written specification, releases it on a dimensional report rather than an opinion, and is still answerable during the first production run. So do not look for someone who will copy your prototype in plastic. Look for someone who tells you, before quoting the tool, what has to change about the part.
What actually changes between a working prototype and unit 1,000?
Four things, and all four move your dimensions.
The material shrinks. A moulded part cools in the tool and shrinks as it does. PP shrinks roughly 1.2 to 2.0%, ABS around 0.4 to 0.7%. The tool is cut oversize to compensate, but shrinkage is not uniform: a thick section pulls more than a thin one. A printed or machined prototype has none of this, which is why a prototype that fits perfectly tells you very little about the first moulded part.
The part has to leave the tool. Draft, parting line, ejector positions and undercuts are irrelevant on a prototype and decisive in steel. A 15 millimetre vertical wall with no draft is nothing on a printer and an ejection problem in a mould.
Wall thickness becomes a process variable. Prototypes often carry thick sections simply because they could. In moulding that produces [sink marks](/blog/engineering-decisions-that-impact-looks), long cooling and locked in stress. Keep wall thickness within about 10% of nominal (up to 25% on low-shrink materials such as ABS), and ribs at roughly 50 to 75% of the adjoining wall depending on the material's shrinkage, towards the lower end for high-shrink materials such as PP.
Repeatability becomes the requirement. With one prototype the question is whether it fits. At 50,000 units the question is whether every part fits, including those from cavity four at the end of a shift. That is a different question, and it needs tolerance limits agreed before the tool is ordered rather than argued about afterwards.
What does the handover to production tooling actually consist of?
A fixed sequence of gates, each with something written at the end of it. If a supplier cannot name these stages, they are planning to improvise.
1. Design for manufacture review on the 3D model. Draft, wall thickness, parting line, gate position, and which tolerances genuinely matter. Changes here cost hours in CAD instead of weeks in steel. 2. Tool design, signed off separately. Cavity count, side actions, cooling layout, steel grade. This is also where you fix the volume the tool is designed for. 3. Tool build. Six to twelve weeks for a production tool, driven mostly by side actions and cavity count. 4. T0: first shots. Does it fill, does it eject, are there visible defects? Not yet judged on dimensions. 5. T1: samples on realistic settings with a full dimensional report against the drawing. This is where you find out what has to change. 6. Correction round and T2. One round is normal, two is unremarkable, three means step 1 was skimped. 7. First article inspection and release. Every dimension on the drawing measured on parts from each cavity, establishing that the tool can make a conforming part. 8. First production run at production cycle time, where you establish that the process does it repeatably. Under PPAP this is a production run of at least 300 consecutive parts, with capability calculated from at least 100 measurements and an index of 1.67 the usual acceptance level for a new process under PPAP, and 1.33 a common minimum for ongoing production.
Steps 7 and 8 are where projects come apart, because they get treated as one thing. First article inspection says the tool can. The first run says the process does. Tie the final tooling payment to step 7, and hold back a small retention until step 8 is complete, so everyone still has a reason to care about the first run.
Is bridge tooling worth it, or do we go straight to a production tool?
Only when a launch date is worth more to you than the tooling spend. Bridge tooling, usually an aluminium or soft steel tool with fewer cavities, delivers parts weeks earlier at a lower investment, and it is genuinely useful for a trade launch, a pilot customer or a first 2,000 to 10,000 units of real market feedback.
Be clear about what it does not do. A bridge tool runs a different cycle, often a different gate and different cooling, so it does not validate the process you will eventually run. You will repeat sampling on the production tool regardless. Treat bridge tooling as the cost of being in the market sooner, not as a technical step on the way to production, and the decision becomes straightforward: if nothing commercial depends on those weeks, go straight to the production tool.
How long does it take, and where does the time actually go?
Four to eight months from approved design to a first production run. The distribution surprises most people:
| Stage | Duration | Where it slips | | --- | --- | --- | | Design for manufacture review and changes | 2 to 4 weeks | Your own decision making, not the engineering | | Tool design and sign off | 1 to 2 weeks | Unresolved questions about volume and cavity count | | Tool build | 6 to 12 weeks | Side actions and cavity count | | T0, T1 and one correction round | 4 to 8 weeks | Each extra round costs 2 to 4 weeks | | First article inspection and release | 1 to 3 weeks | Waiting on a report nobody specified in advance | | First production run | 2 to 6 weeks | Material procurement and press scheduling |
Tool build is the longest stage and the one you can influence least. The time you can genuinely win sits in stages one and four: a thorough design for manufacture review often saves a correction round, and that saving comes straight off the critical path.
An anonymised example. A consumer goods client running 25,000 units a year held the tool design for three weeks waiting on an internal decision about colour. That single delay cost exactly the three weeks later reported as toolmaker overrun.
Who stays accountable during the first production run?
Name the person before the run, not during it. The common failure is that engineering support quietly ends at tool release, so the first run is handed to a planner and a press operator who were not part of any of the earlier decisions. Then a short shot or a dimensional drift becomes a three week email exchange about whose problem it is.
Four things to settle in writing before the first run starts:
- Who signs off the dimensional report, and against which drawing revision. One named person on each side, one revision number.
- What happens on a deviation. Who stops the run, who assesses, and within how many hours. A rule written in advance costs nothing; the same rule written during a stoppage costs a week.
- Scrap reporting per cavity, not as one total. One weak cavity in a four cavity tool hides easily inside an average, and it will be the one producing your field returns.
- How many correction rounds are included and who pays for the next one. This is the single most common gap in a tooling contract and the most expensive one to discover late.
One honest trade off to plan for: the cheapest tooling quotation frequently carries the longest correction loop, because distance adds a shipping cycle to every sampling round. That is not an argument for or against any region, it is an argument for pricing the rounds rather than only the tool.
[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: shrinkage figures and design rules from publicly available material and moulding documentation; the capability run size and capability index thresholds from PPAP (AIAG), the widely used industry standard for production part approval. Durations are practical ranges rather than commitments, and the anonymised example contains no identifying detail.
Frequently asked questions
- How long does it take from approved design to first moulded production parts?
- Four to eight months is realistic. Design for manufacture review two to four weeks, tool design one to two, tool build six to twelve, T0 and T1 sampling with one correction round four to eight, then release and the first run. Every additional correction round adds two to four weeks.
- What does a production mould cost in Europe compared with Asia?
- Asian tooling is typically the lower invoice, European tooling the shorter correction loop. The gap narrows once you price in sampling rounds, shipping, your own travel and the cost of a delayed launch. Ask for the tooling price and the piece price together, because either one alone is meaningless.
- What is a realistic first order quantity?
- Well beyond the capability run, so the process can settle. Under PPAP that run is at least 300 consecutive parts at production rate, with capability calculated from at least 100 measurements. A first commercial run commonly lands between 1,000 and 5,000 units, depending on call off pattern and shelf life.
- How much is paid up front for tooling?
- Staged payment is standard: part on order, part at first samples, the balance on release. Tie that final stage to an approved dimensional report and hold back a small retention until the first run is complete. Paying in full up front gives away your only leverage during the correction rounds.
- Is bridge tooling worth it before the production tool?
- Only when a launch date or early market feedback is worth more than the spend, for example a first 2,000 to 10,000 units for a trade launch. An aluminium bridge tool is quicker and cheaper, but it does not validate the production process. Budget it as marketing, not engineering.
- Can the mould still be changed after the first samples?
- Removing steel is nearly always possible; adding it means welding, an insert or a new block. Keep any dimension still under discussion steel safe, so the toolmaker reaches the final size by removing steel and a correction stays a machining operation. It is the cheapest insurance in the whole project.