Pick a metal card thickness and you’ve basically picked the personality of the card: how it flexes, how it clacks on a table, whether it feels “premium” or just “metal-ish,” and, less romantically, whether it survives readers, wallets, and years of fingerprints.
I’ve watched projects stall because the team treated thickness like a minor spec. It isn’t. Thickness drives tooling, finishing, postage weight, and even how forgiving the edges are when the card gets dropped on concrete.
One line that holds up: you’re not choosing a number, you’re choosing a failure mode.
The thicknesses people actually buy (and why)
Metal cards tend to cluster into a few common ranges because suppliers keep sheet stock, cutters, and finishing workflows tuned for them. Once you wander off the beaten path, lead time and scrap go up fast.
Here are the typical “buckets” you’ll see:
– 20, 25 mil (0.51, 0.64 mm): thin, springy, easiest on cost and mailing weight
– 30, 32 mil (0.76, 0.81 mm): the practical sweet spot (and close to payment-card norms)
– 40 mil (1.02 mm): noticeably heavy, stiff, “luxury object” vibes
– Beyond that (1.2 mm+): possible, but you’re in specialty-tag territory and you’ll pay for it
If you’re curious to compare or understand these differences, you can read more about metal card thickness and how it affects both looks and function.
And yes, those conversions matter. Vendors will quote in mils, mm, or “gauge” depending on where they sit in the supply chain, and “gauge” is where misunderstandings go to breed.
Hot take: 40 mil is overrated for most card programs
It feels expensive because it is. But “thicker” doesn’t automatically mean “better” once wallets, reader slots, and edge damage enter the chat.
40 mil can be fantastic for a showpiece membership card that rarely goes into a magstripe reader (or doesn’t need to). For everyday access credentials and anything that must behave like a standard ID-1 card, 30, 32 mil is usually the smarter kind of boring.
I’ve seen 40 mil cards come back not because they broke, but because they annoyed users.
Thickness changes the feel more than the finish (sometimes)
This part surprises people. They’ll spend weeks on coating options, then choose a thickness that makes the card feel wrong in-hand.
– 20, 25 mil: light and snappy. You can flex it a little. Some people interpret that as “cheap,” even if the metal is legit.
– 30, 32 mil: stable in the hand. You get that confidence without turning the card into a pocket brick.
– 40 mil: substantial. The card announces itself when it hits the table.
One-line truth:
A premium finish can’t fully hide a flimsy core.
The technical bit: rigidity, denting, and the “bend story”
If you want the engineering view, thickness is doing three things for you:
- Raising stiffness (stiffness scales strongly with thickness; small increases can feel big)
- Improving dent resistance (not the same as scratch resistance, finish matters there)
- Reducing warp sensitivity during finishing (but thick stock can also “remember” stress from forming)
Now, this won’t apply to everyone, but if you’re doing deep engraving, heavy coining, or aggressive edge machining, going too thin often forces compromises: shallower features, more reject risk, or edges that feel sharp because you can’t break them enough without distorting the card.
Also: thicker doesn’t mean “won’t bend.” It means when it bends, it may kink or stay bent rather than spring back, depending on alloy and temper.
“Gauge” standards: messy, regional, and occasionally misleading
Here’s the thing: sheet metal “gauge” is not one universal system. Stainless gauge charts aren’t aluminum gauge charts. Suppliers will quote a gauge that sounds precise while the actual delivered thickness lands somewhere in a tolerance band.
So: don’t spec “24 gauge stainless” and assume the number is the number. Spec the thickness in mm or mils, then add tolerance.
A useful anchor point from the card world:
ISO/IEC 7810 ID-1 (the credit-card form factor) defines 0.76 mm nominal thickness for standard cards. Source: ISO/IEC 7810:2019.
That doesn’t mean your metal card must be 0.76 mm, but if you need broad compatibility with wallets and a lot of existing card-handling equipment, it’s a powerful reference.
Measuring thickness: don’t trust a single reading
If you’re validating incoming stock or checking a production run, grab a method that matches the surface.
Micrometers are great for spot accuracy, but coatings and textures can trick you. Calipers are fine for trends, less so for final sign-off. Non-contact optical methods are excellent when the finish is delicate, but not every shop has them.
If you do one thing consistently, do this:
Measure at multiple points (center + corners), on a flat reference surface, with consistent pressure.
Warp is sneaky. And warp becomes a “reader problem” long before it becomes a “customer complaint.”
Cost and availability: what the quotes won’t say out loud
Thicker stock costs more in material, sure. The hidden costs are more interesting:
– Handling and throughput: thicker pieces move slower through certain finishing lines
– Tool wear: edge finishing and deburring can get more expensive with thicker cards
– Scrap risk: higher on unusual thicknesses because setups aren’t dialed in
– Supply stability: 30, 32 mil tends to be stocked; oddball thicknesses can trigger long lead times
The procurement pattern I see most: teams pick 40 mil for “premium,” then quietly drop to ~30 mil after the first production schedule gets real.
Finishes that fight (or favor) your thickness choice
Finishing isn’t just decoration; it changes how thickness behaves in the real world.
Thin cards can look amazing with crisp laser marking, but if you add coatings that build up at the edge, you may get a visible “reveal line” that makes the edge feel cheaper than it is. Thick cards let you do more dramatic edge profiles, bevels, chamfers, even deeper perimeter details, without the whole piece feeling knife-like.
A few shop-floor truths I’ve learned the hard way:
– Rounded edges are more forgiving on thinner cards and reduce chip-prone corners.
– Beveled edges look incredible on thick stock, but they can telegraph any finishing inconsistency.
– Plating + heavy thickness can amplify flatness problems if stress relief isn’t managed.
(And yes, different base metals behave wildly differently under the same finish stack.)
Use-case matching, minus the marketing fluff
Access badges (daily swipes, daily abuse)
Go for ~0.8, 1.0 mm if you can, with attention to edge durability. Repeated reader contact punishes corners and perimeter finishes. If you’re integrating RFID/NFC, thickness may also interact with inlay placement and reader distance expectations, prototype it, don’t theorize.
Short version: durable edges beat brag-worthy thickness.
Business cards (hand-to-hand, pocket-to-wallet)
30, 32 mil is the dependable choice. Thin can feel “promo,” thick can feel like you’re forcing someone to carry a token. If you’re doing a lot of fine detail (microtext, intricate marks), don’t sabotage it with a thickness that warps during finishing.
Membership / VIP cards (kept, shown, talked about)
This is where 40 mil makes sense, sometimes. If the card mostly lives in a presentation sleeve or a wallet slot and doesn’t need to run through legacy readers, thickness becomes part of the theatre.
But if users actually carry it daily? Mid-thickness with a great edge treatment often wins.
A decision framework that doesn’t pretend life is simple
Look, there’s no perfect thickness. There’s a thickness that fits your constraints and fails in the least painful way.
Ask yourself (in this order, not the other way around):
- Does it need to match ID-1 handling norms (0.76 mm-ish)? If yes, stay near 30 mil.
- How will it be used? Swiped, tapped, handled, mailed, stored in wallets?
- What’s the finish stack? Coatings, plating, engraving depth, edge profile.
- What’s your tolerance reality? Your supplier’s, not your wish list.
- Can you prototype and abuse-test it? Bend, drop, abrasion, reader interaction.
Then pick the thinnest option that still hits your durability and feel targets. Overbuilding is real, and it’s expensive.
Real-world thickness picks I actually see working
– Access control credentials: 0.8, 1.0 mm, depending on reader + durability needs
– Payment-grade compatibility targets: 0.76, 0.88 mm (close to ISO ID-1 expectations)
– Premium membership “statement” cards: 1.2, 1.6 mm when reader constraints are minimal
– Industrial tags / harsh environments: 1.5, 2.0 mm with finishes selected for corrosion and abrasion
Notice the pattern: thickness rises when compatibility constraints drop.
That’s not romance. That’s physics and infrastructure.
If you want a recommendation with teeth, I’d start most programs at 30, 32 mil and only move thinner or thicker after you’ve decided how the card is finished, how it’s handled, and what “failure” you can tolerate: cosmetic wear, bending, reader issues, or replacement rate. Each thickness buys you something, and charges you for it somewhere else.