Injection Mold Surface Finish Guide: SPI Grades, Texture and Cost (2026)

Surface finish is one of the few decisions on a molded part that is set entirely by the mold — and frozen the moment the cavity is polished. Polish the tool to a mirror and every part comes out glossy; etch a texture into it and every part repeats that grain for the life of the tool. Because the finish lives in the steel, it has to be specified before the mold is cut, not negotiated after the first samples. Get it right and the part looks exactly as intended at the lowest sensible tooling cost. Get it wrong and you pay twice: once for polishing the wrong grade, and again to re-polish or re-texture hardened steel.

This guide explains the SPI finish system (grades A-1 through D-3), how finish drives tool steel, draft angle and cost, and exactly how to call finish out on a drawing so your quote prices what you meant.

Why Surface Finish Is a Tooling Decision, Not a Cosmetic Afterthought

The surface of the part is a copy of the surface of the cavity. That single fact connects finish to three things buyers care about: cost, release, and appearance.

• Cost — polishing to a fine SPI A grade is hand work under magnification. It adds hours to the tool build and demands a steel that can hold the polish.
• Release (draft) — a textured cavity grips the part far harder than a polished one, so it needs more draft per side or the part scuffs and drags on ejection.
• Appearance — gloss amplifies flow marks and weld lines; a matte or textured surface hides them. Choosing gloss purely for looks can expose defects a textured finish would conceal.

Treat finish as a spec you lock in DFM, right alongside wall thickness and draft, rather than a note added to the drawing at the end.

The SPI Finish System: Four Families, Twelve Grades

The SPI (formerly the Society of the Plastics Industry) classification is the shared language for mold finish. It grades the cavity by how the surface was worked — diamond buffing, paper, stone, or blast — across four families A, B, C and D, each with three grades running fine to coarse. The roughness figures below are typical reference values in microns (Ra); use them to compare suppliers, not as a guaranteed number on your own resin.

Family	Grade	Method	Typical Ra (µm)	Appearance / typical use
A (gloss)	A-1 / A-2 / A-3	Diamond compound buff	≈ 0.012 – 0.10	Mirror to high gloss — optics, light guides, glossy cosmetic parts
B (semi-gloss)	B-1 / B-2 / B-3	Sandpaper / emery, 600 → 320 grit	≈ 0.05 – 0.32	Semi-gloss to satin — consumer housings, general cosmetics
C (matte)	C-1 / C-2 / C-3	Stoning, 600 → 320 grit	≈ 0.35 – 0.70	Matte, low gloss — functional and hidden parts; hides sink marks and flow marks
D (textured)	D-1 / D-2 / D-3	Dry blast, glass bead → coarse oxide	≈ 0.8 – 18	Stippled to coarse grain — anti-slip, fingerprint/scratch hiding, grip

Note the span: the finest grade (A-1) and the coarsest (D-3) differ by more than a thousandfold in roughness. That is the range a single line on a drawing — "finish: SPI B-1" — has to pin down.

Matching Finish to the Surface

Most parts do not need one finish across the whole tool. The cost-efficient move is to grade the tool surface by function:

• Visible / class A surfaces (what the customer sees and touches): a gloss or semi-gloss — SPI A or B — chosen for how the product should present.
• Hidden / functional surfaces (inside a housing, brackets, mating faces never seen): SPI C. It costs less to produce and hides minor tool marks a gloss would reveal.
• Grip / anti-fingerprint / wear-hiding surfaces: SPI D texture. The texture does the job a coating or soft-touch material might otherwise be asked to do.
• Small features (snap fits, threads, ejector-pin zones): usually left at standard machine finish regardless of the rest of the part — polishing them rarely shows and only adds cost.

Masking and specifying finish by zone is normal practice and is one of the fastest ways to cut tooling cost without touching how the part presents.

Finish Is a Steel Decision Before It Is a Polish Decision

Not every tool steel can hold every grade. Soft or free-cutting steels polish quickly but dull early; a durable mirror or high-gloss finish belongs in a harder, cleaner steel such as P20 pre-hardened, H13 hardened, or a stainless grade like S136 where the part also runs a corrosive resin. Specify an A-grade gloss into tooling that cannot hold it and the finish fades within the first production runs — the parts drift from "gloss" to "satin" while the tool is still new.

This is why a finish callout can change which steel is quoted. Our comparison of P20 vs 718H vs H13 vs S136 walks through how hardness, corrosion resistance and polishability trade off — the same decision your finish grade forces.

Texture, Draft and Cycle Time

A texture is not free geometry. It sits proud of the cavity wall, so the part has to be pulled across it on every shot. More texture depth means more draft and more ejector force:

Texture depth	Rough draft per side	Effect if under-drafted
Fine grain (≈ 10–20 µm)	≈ 1.5–2°	Light scuffing, gloss streaks on the pull face
Medium grain (≈ 20–40 µm)	≈ 3°	Visible drag marks, "chatter" texture, whitening
Coarse grain (≈ 50 µm+)	≈ 5°+	Part hangs, tears or sticks; scrap climbs

Because draft and texture are linked, the texture must be chosen before the cavity is drafted. Adding a deep texture to a mold that was drafted for a glossy surface is one of the most expensive late changes there is — it can mean re-cutting and re-polishing the whole pull face. This is exactly the kind of coupling a proper defect-prevention review is meant to catch before steel is cut.

The Resin Interacts With the Finish

Two effects trip up buyers who treat finish and material as independent choices:

• Abrasive, filled resins dull a fine polish faster. Glass- or mineral-filled grades wear a glossy cavity quicker than an unfilled resin does, so a mirror finish intended for a long production run may need a harder steel or a slightly less demanding grade.
• Filled grades change how a texture reads. A texture proven on unfilled nylon is not guaranteed to look identical once the resin becomes a glass-filled version — flow behaviour and fibre orientation shift the grain's appearance.

Glossy surfaces also make weld lines and flow marks far more visible than a matte or textured surface does. If your part has a complex fill pattern, a matte grade is often the smarter call — it delivers a clean look at lower tooling cost than fighting flow marks in a mirror cavity.

How to Call Finish Out on a Drawing or RFQ

Vague finish requests ("smooth", "matte", "nice surface") are the main reason finish cost comes back wrong. Make the callout unambiguous and per-zone:

• Name the grade per surface zone — e.g. "class A face: SPI B-1; internal: SPI C-1; grip band: SPI D-2".
• State the texture standard where a pattern is needed (Mold-Tech or an equivalent grain reference), and give the texture depth in microns.
• Quote a roughness (Ra) only as a reference, not the governing figure, unless you truly measure it — SPI grades are defined by process, not by a single Ra number.
• Flag where gloss is functional (optical, sealing, cleanliness) so the supplier prices the right steel and polishing time.

What Drives Finish Cost

Three levers move the finish line on a quote:

Change	Typical impact on tooling cost / lead time
C-grade → B/A gloss on a visible face	More hand polishing; +days on the tool build
A-1 mirror across a large face	Highest polishing cost; may force a higher-grade steel
Adding a deep texture after drafting	Re-work of the pull face — expensive; specify texture up front
Grading finish by zone instead of blanket gloss	Lower tooling cost, same appearance where it matters

The pattern is consistent: precision finish is expensive only when it is applied where nobody sees it, or added after the draft is already set.

Pre-Quote Checklist

• ☐ Which surfaces are class A (visible/touch) and which are hidden or functional?
• ☐ Is a specific SPI grade named for each zone (A/B/C/D and the level within it)?
• ☐ If a texture is required, is the grain standard and depth in microns specified?
• ☐ Has the needed draft per side been derived from the texture depth?
• ☐ Can the chosen tool steel hold the specified gloss for the intended production volume?
• ☐ Is the resin abrasive or filled in a way that will dull a fine polish?
• ☐ Are gloss and matte zones masked on the drawing, not left to the shop's default?
• ☐ Has the finish been fixed before the cavity is polished (not requested after first samples)?

Need a Finish That Matches the Part?

Precimold designs and builds injection molds for silicone and plastic parts and works the finish into the DFM stage — so your tool steel, draft and texture are chosen together instead of corrected later. Send us your 3D model and a note on which surfaces are class A; we will return a finish recommendation per zone with the tooling cost and lead-time impact. We reply within 12 hours.