{"id":3262,"date":"2026-10-18T22:00:10","date_gmt":"2026-10-19T10:00:10","guid":{"rendered":"https:\/\/footworklab.net\/?p=3262"},"modified":"2026-10-18T22:00:10","modified_gmt":"2026-10-19T10:00:10","slug":"custom-insole-mold-supplier","status":"publish","type":"post","link":"https:\/\/footworklab.net\/ar\/news\/3262\/","title":{"rendered":"Custom Insole Molds: Sizes, Turnaround and What to Specify"},"content":{"rendered":"
\n

About the author<\/strong> \u2014 FOOTWORK LAB is a manufacturer of foot assessment equipment and custom orthotic insoles, based in Ruichang, Jiangxi Province, China. Founded in 2014, with its own production plant since 2018. Certified to European Union and ISO quality system standards. See the factory<\/a> or view certification<\/a>.<\/p>\n<\/div>\n

A mold is the least glamorous object in an insole business and the one that decides whether the business can repeat itself.<\/strong> It is a physical record of a decision: this shape, this size, this revision. When a customer returns for a second pair, the mold is what makes that pair a reorder rather than a new project.<\/p>\n

Buyers tend to regard molds as a detail of the manufacturing process, ordered when needed and forgotten afterwards. That works until the first reorder arrives, or the first specification change, or the first customer who wants to move production and discovers that the mold is not theirs.<\/p>\n

FOOTWORK LAB produces insoles in our own plant in Ruichang, Jiangxi Province, across molded, milled and thermoformed builds. Molds and models are part of our daily work, and the questions below are the ones that separate a mold library that works from a shelf of objects nobody can identify.<\/p>\n

What a mold is in an insole workflow<\/h2>\n

In a molding workflow, a mold or model is a positive representation of the foot or of the insole shape, used to form a heated sheet into the intended geometry. It can be produced from a physical cast, machined from captured data, or built up by hand and adjusted.<\/p>\n

Its function is repetition. The first pair can be made by hand; the second, third and hundredth need something that holds the shape exactly. That is what a mold provides, and it is why the quality of the mold matters more than the speed of the first pair.<\/p>\n

The mold is also a specification artefact. It embodies decisions about arch height, heel cup depth, medial extent and size. When those decisions change, the mold has to change, and the record of that change is what allows a lab to know which pair came from which revision.<\/p>\n

Sizes, grading and the size-run decision<\/h2>\n

A mold is size-specific. A lab serving a population needs either a size run of molds or a way to adapt a single mold to different feet. Both approaches exist, and they produce different products.<\/p>\n

A size run produces consistency within each size and a clear story for the customer: this is the shape for your size. It also multiplies the library, the storage and the cost, because every size needs its own object and its own revision control.<\/p>\n

A single mold adapted per patient produces a more individual shape and a smaller library, at the cost of more operator time per pair and more variability between pairs. For low volume with high individualisation, that trade is often worth making.<\/p>\n

The decision should be made before the first mold is ordered, because retrofitting a size run onto a single-mold workflow means rebuilding the library. Ask any supplier how they handle grading, and whether the size steps are proportional or adjusted by region.<\/p>\n

\"Foot
FOOTWORK LAB INSOLE. Measurement data only \u2014 clinical interpretation remains a professional judgement.<\/figcaption><\/figure>\n

What grading actually changes<\/h3>\n

Grading is not scaling. A foot does not grow proportionally in every dimension, and a mold that is simply enlarged produces an insole that is too wide at the heel or too narrow at the forefoot. Good grading adjusts regions rather than multiplying the whole.<\/p>\n

That matters for fit and for the brief. A size run built by proportional scaling will produce a pair that fits the middle sizes well and the extremes poorly, and the complaints will come from the customers at the ends of the range.<\/p>\n

Ask how the size steps were derived and whether they were checked against real feet. A supplier who can show the check has done the work; one who describes grading as a multiplier has not.<\/p>\n

Half sizes and the honest limits<\/h3>\n

Half sizes multiply a library again. Many labs avoid them by specifying the insole to the shoe’s internal length rather than to a nominal size, which handles the in-between cases without doubling the tooling.<\/p>\n

Where half sizes are offered, the reason should be a real difference in shape rather than a marketing convenience. A pair that is a scaled copy of the next size up is not a half size; it is the same shape with a different label.<\/p>\n

Be honest with customers about this. A lab that explains that the insole is built to the shoe and the foot rather than to a number on a box is more credible than one that offers a size for every occasion.<\/p>\n

Material and durability<\/h2>\n

Molds are made from several materials, each with a different life. Resin and polymer molds are accurate and comparatively quick to produce; denser materials last longer and cost more; wooden or plaster models are traditional and require more care.<\/p>\n

The right choice depends on how many pairs the mold will produce. A mold used for one pair can be light and cheap. A mold used for a production run has to hold its shape through heat and pressure, and the material has to be chosen for that.<\/p>\n

Ask how many cycles a mold is expected to survive and how the supplier knows. A mold that deforms after a few cycles produces a batch of pairs that are subtly wrong, and the failure is invisible until someone compares them to the reference.<\/p>\n

Revisions and version control<\/h2>\n

Every mold should carry an identifier and a revision number. The identifier links it to a customer or a specification; the revision links it to a decision. Without both, a library becomes a collection of similar objects.<\/p>\n

When a specification changes, the change should produce a new revision rather than a silent modification. A hand-adjusted mold with no new label is the most common cause of a reorder that does not match the original pair.<\/p>\n

The rule is simple and it costs a marker pen: one mold, one label, one revision, and no change without a new label. Labs that follow it can answer a question about a two-year-old pair in minutes.<\/p>\n

\"Foot
FOOTWORK LAB INSOLE. Measurement data only \u2014 clinical interpretation remains a professional judgement.<\/figcaption><\/figure>\n

Storage and retrieval<\/h2>\n

Storage is where libraries fail quietly. A mold stored under weight deforms; one stored in heat warps; one stored without a label becomes anonymous. The room matters as much as the rack.<\/p>\n

Retrieval is the test of the system. Ask a colleague who did not create the mold to find it using only the records. If they cannot, the system is a personal memory rather than a library, and it will fail when that person leaves.<\/p>\n

A simple index, whether on paper or in a spreadsheet, is enough: identifier, customer, date, revision, size and location. The index is the library; the rack is just where the library lives.<\/p>\n

Molds versus files: two ways to hold a shape<\/h3>\n

A physical mold holds a shape in an object. A digital file holds it as data. Both can reproduce a pair, and they fail differently. A mold can deform, warp, lose its label or go missing. A file can be lost, overwritten, exported in the wrong format or opened by software that no longer exists.<\/p>\n

Labs that run milled or printed builds are effectively running a file library, and the discipline required is the same: naming, versioning, backup and a rule that no change happens without a new version. The physical object is replaced by a filename, and the failure modes move from the shelf to the server.<\/p>\n

The practical recommendation is to hold both where the workflow allows. A file reproduces the shape; a physical reference pair proves what the shape produced. Together they answer the question every customer eventually asks: is this pair the same as my last one?<\/p>\n

How a mold order goes wrong<\/h3>\n

The most common failure is an order placed from a description rather than from a recording. A supplier who receives a sketch and a size has to interpret, and the interpretation is invisible until the pair arrives.<\/p>\n

The second is an unstated revision. A customer asks for the pair from last year with one small change, and the change is applied to a mold without a new label. The next reorder produces the modified shape, and nobody can explain why the pair feels different.<\/p>\n

The third is a size mismatch between the mold and the shoe. A mold that produces a faithful foot shape in a size that does not fit the intended footwear generates a pair that cannot be worn, and the error surfaces at the fitting appointment rather than at the factory.<\/p>\n

What a good supplier asks before quoting<\/h3>\n

A supplier who knows the work will ask for the recording, the intended footwear, the volume, the specification revision and the size range before quoting. Those five questions are the difference between a quotation and a guess.<\/p>\n

They will also ask what happens if the first mold is not right. A supplier who expects to iterate on tooling has priced that iteration into the relationship; one who regards the first mold as final has priced it into the risk.<\/p>\n

Our own process asks the same five questions and quotes tooling and production separately, so the cost of a new shape is visible rather than hidden in a unit price. That transparency is worth more to a buyer than a lower headline number.<\/p>\n

Turnaround and scheduling<\/h2>\n

Mold production has a lead time, and it sits in front of every pair that depends on it. A lab that quotes a delivery date without accounting for mold lead time will miss it on the first custom order.<\/p>\n

Where the mold already exists, turnaround is production time. Where it does not, the order has two stages: tooling and production. Ask suppliers to quote both separately, so the cost of a new shape is visible rather than buried in a unit price.<\/p>\n

Scheduling follows. A lab that batches molds can amortise setup; one that orders a mold per pair pays the setup every time. Neither is wrong, and the choice depends on whether the lab’s demand is for repeat pairs or for one-off shapes.<\/p>\n

\"Foot
FOOTWORK LAB INSOLE. Measurement data only \u2014 clinical interpretation remains a professional judgement.<\/figcaption><\/figure>\n

Ownership, transfer and exit<\/h2>\n

Ownership of tooling should be agreed in writing before the first order. If the supplier holds the molds, the customer may not be able to move production without rebuilding them; if the customer holds them, storage and transfer become the customer’s problem.<\/p>\n

Ask what happens at the end of the relationship. Will the molds be transferred, and at what cost? Will the data behind them be provided? A supplier who has not thought about this will improvise when asked, and improvisation at that moment is expensive.<\/p>\n

Our own arrangement is stated in the order: insoles are available under OEM, where the specification is the customer’s, and under ODM, where we carry more of the design. Tooling ownership follows the specification, and we put it in writing rather than leaving it to memory.<\/p>\n

Whatever the workflow, record the outcome. A pair produced from a new mold should be photographed beside its specification and kept as the reference for that revision. The photograph costs nothing and settles arguments that would otherwise become opinions.<\/p>\n

What to specify in a mold order<\/h2>\n

Specify the shape in the terms the lab actually uses: arch height and extent, heel cup depth and flare, forefoot width, and the size or size range. Attach the recording or the model it was derived from. State the material and the expected cycle life.<\/p>\n

Then specify the administrative side: identifier, revision, storage location, and who owns it. Those four items take a minute to write and prevent most of the disputes that arise two years later.<\/p>\n

Finally, specify the acceptance check. A mold should be checked against the model or the file before production begins, and the result recorded. That single habit turns a tooling purchase into a controlled process.<\/p>\n

Frequently asked questions<\/h2>\n
\n

Do I need a mold for every customer?<\/h3>\n

Only if the workflow is mold-based and the shape is individual. Many builds use a size run plus adaptation, and milled or thermoformed routes may need no physical mold at all.<\/p>\n

How long does a mold last?<\/h3>\n

It depends on the material and the process. A light mold for a single pair and a production mold for a batch are different objects. Ask for the expected cycle life and how it was established.<\/p>\n

Who owns the mold?<\/h3>\n

That should be in the purchase order. Ownership determines whether production can be moved later, and it should follow the specification rather than the convenience of the moment.<\/p>\n

Can a mold be modified instead of replaced?<\/h3>\n

Sometimes, and the modification should produce a new revision label rather than a silent change. An unlabelled modification is how a reorder stops matching the original pair.<\/p>\n

Do you produce insoles without a physical mold?<\/h3>\n

Yes. Milled and thermoformed builds can be produced from captured geometry without a physical mold, which suits low volume and frequent specification changes.<\/p>\n

What is the minimum order?<\/h3>\n

Two hundred pairs for insoles, mixed across models and sizes, with samples dispatched within three days of specification approval.<\/p>\n

How long does delivery take?<\/h3>\n

Six to seven working days by express, ten to twelve by air freight and thirty to thirty-five days by sea, depending on destination and clearance.<\/p>\n<\/div>\n

What we would do next<\/h2>\n

Before ordering tooling, decide whether the workflow needs it at all: a milled or thermoformed build may avoid the library entirely. Our article on how a production order moves from data to dispatch<\/a> shows where tooling sits in the sequence, and the piece on arch support construction<\/a> covers the shape decisions a mold has to hold.<\/p>\n

For minimum order, samples and OEM arrangements, see the FAQ page<\/a>. To discuss a build, contact our production team<\/a> or visit the factory page<\/a>.<\/p>\n

This article is intended for professional and educational reference. It is not medical advice. Foot assessment equipment produces measurements and images; clinical interpretation and any decision about care remain the responsibility of a qualified professional. Certification scope is limited to the documentation we hold: European Union certification and ISO quality system certification.<\/em><\/p>\n

\n

Related reading<\/h2>\n