Most people treat a business card as a formality: a small rectangle handed over in a hurry and forgotten by dinner. In practice it is one of the few marketing assets that gets physically carried, kept, and re-read — often weeks after the first meeting. That durability is exactly why card design deserves more than a template and a rush job.
Two shifts made card production both faster and sloppier. Online printing removed the print shop from the middle of the process, and generative design tools removed the blank-canvas problem. Now anyone can produce something card-shaped in an afternoon, which also means anyone can produce something forgettable just as quickly. The difference between the two is a workflow: a defined brief, deliberate design decisions, correct file preparation, and a review pass that catches the errors printers cannot fix.
This guide covers that workflow end to end, including how to turn it into marketing and design tutorials that keep working for you after the cards arrive.
A card does three jobs at once, and it has to do all of them on 85 × 55 mm of paper. It identifies who you are, it states the one action you want the other person to take, and it buys the two seconds of attention needed for someone to decide whether to keep it.
Nothing digital replicates that sequence reliably. A professional network search works when someone already remembers your name. A card works when they do not.
Cards also behave differently from other collateral in ways that should shape your design choices:
The practical implication: treat the card as a compression problem, not a decoration problem. Every element either helps a stranger remember you or it takes space from something that would.
Generative tools changed the early phase of card design more than the late phase. Understanding which half they own prevents a lot of rework.
The most valuable use of a generative design tool is producing twenty rough directions in ten minutes. Prompt with intent — industry, brand personality, three adjectives, the color family you already use, and the tone of your audience. You are not looking for a finished card. You are looking for two or three compositional ideas worth developing: a type-led layout, an image-led layout, a grid that echoes your logo geometry, a strong horizontal band that separates contact details from identity.
Tools that analyze your existing brand assets can suggest palettes that harmonize with your primary color rather than fight it, and type pairings that respect the personality of your existing wordmark. Use these as constrained options, not verdicts. A shortlist of three palettes and three type pairings is far more useful than a hundred variations, because you can actually test each one at final size.
Three failure modes appear again and again:
Treat generated comps as sketches pinned to a wall. Final artwork gets rebuilt with real type, real vector marks, and verified contact data.
A one-page brief costs ten minutes and saves hours of revision. The brief is also the raw material for your tutorial content later, so write it in plain language you would not mind recording.
If your card tries to communicate a tagline, three services, two phone numbers, an address, four social handles, and a slogan, the reader keeps none of it. Choose the one line that a stranger should be able to repeat back an hour later. Everything else either supports that line or moves to the back of the card — or off the card entirely and onto the page the card points to.
Write the brief as six labeled lines: audience, next step, setting, keep-reason, must-include, must-avoid. That structure fits on an index card, survives a handoff to a designer, and converts directly into a script for a tutorial video. When a stakeholder asks for changes later, you can point at the line that anchors the decision instead of arguing about taste.
Gather your vector logo (SVG, AI, or EPS), exact contact strings, licensed font files or links, any photography you own, and your brand palette with hex or CMYK values. Keep a single folder. Half of all late-stage design chaos comes from hunting for a logo file at midnight.
Produce a broad set of rough comps. Discard anything that requires explanation. Keep the two or three that read clearly when viewed at thumbnail size — that is roughly how much attention a card actually receives.
Set up a real document at final trim size with the correct bleed. Place your vector logo. Type the contact information with your licensed font. If you are using a generated illustration or texture, place it at a resolution that supports 300 DPI at final size, and keep it out of the safe area.
Document size equals trim size, not paper size. Add bleed on all four sides. Define a safe margin for text. Place crop marks outside the bleed when you export. Name layers clearly, and keep a separate layer for anything a printer may need to remove, such as a version marker.
Standard trim sizes
| Region | Trim size | Common bleed |
|---|---|---|
| North America | 3.5 × 2 in | 0.125 in |
| Europe and most of Asia | 85 × 55 mm | 3 mm |
| Japan | 91 × 55 mm | 3 mm |
Export a press-ready PDF at 300 DPI with bleed and crop marks, and a separate low-resolution version for email and mockups. Preflight checks: fonts embedded, images at 300 DPI, nothing important inside the bleed zone, large black areas built as a rich black blend, and no RGB colors left in the file.
Print the file at 100 percent on plain paper, cut it to trim size with scissors, and hand it to someone who has never seen the design. Watch where their eyes land first. If they read the tagline before your name, the hierarchy is inverted. Paper exposes problems that a zoomed-in screen view hides.
Screens forgive; paper and ink do not. A few rules prevent the most common disappointments.
Type size and weight. Body-size contact details below roughly 7 pt (about 2.5 mm cap height) become a squint. Thin light weights vanish at small sizes, especially on uncoated stock where ink spreads slightly. Use a regular or medium weight for anything below the name line.
Contrast. Aim for a clear luminance difference between text and background, not just a hue difference. Mid-blue text on mid-gray is legible on a monitor and muddy in print.
Small reversed type. White text on a dark flood looks striking on screen and can fill in on press. If you reverse small type, choose a heavier weight and give the printer stock that handles it.
Rich black versus single black. Large black areas print better as a rich black blend. Small text should stay single-channel black so it does not register-blur.
Color mode. Convert to the printer's CMYK profile before export, and expect some drift: vivid RGB greens and oranges will dull. If exact color matters, request a proof.
Consistency across materials. Once the palette and type scale work on the card, they should scale to signage, packaging inserts, and slide templates without change. Set the type scale once and reuse it; a card is usually the smallest surface in a system and the best place to define it.
Bleed is the extra artwork beyond the trim edge that gets cut away. Without it, a background that reaches the edge can show a sliver of white after trimming, because cutting machines drift by a fraction of a millimeter. Safe area is the opposite idea: keep text and logos well inside the trim so small shifts never clip them. A common rule is to keep all live content at least 3 mm from the trim on every side, and more if the card uses a dark background where a white sliver is most visible.
Photographs and textures should be at 300 DPI at final placed size. Vector artwork stays vector. Export a single PDF with embedded fonts, or supply the source document plus linked assets if your printer asks for it. Avoid flattening a layered file until you have confirmed no further edits are needed.
Lightweight stock feels disposable; heavy stock feels considered. Soft-touch matte and uncoated papers take ink differently and change perceived color. Finishing options such as spot varnish, foil, or edge painting add cost and craft. Use one, and let it reinforce hierarchy rather than decorate randomly. A single spot-varnish highlight on the logo often communicates more than three finishes competing for attention.
If you include a QR code, keep it at least 2 × 2 cm (0.8 in) with a quiet margin around it, put it on a light background, and test it from three phones in low light before you approve the file. Also test the destination page on mobile. A code that leads to a slow, desktop-only page wastes the only interactive element on the card.
Plan how the cards will be transported and handed over. A cheap card in a nice holder reads better than a nice card that arrives bent. Consistency between the card and its container is a small, inexpensive upgrade most people skip.
The same work you do for your own card is a ready-made content series. The advantage of recording it is that the tutorial is specific, verifiable, and reusable.
Tutorials that perform best answer a narrow question. Good angles include:
Screen recordings of menu navigation make poor tutorials. What viewers need is the reasoning: why this measurement, why this weight, why the logo shrank, why the secondary line moved to the back. Narrate the trade-offs out loud. An eight-minute video that explains three decisions beats a twenty-minute video that shows forty clicks.
For each tutorial, capture:
Repurpose each recording into a long-form article, a carousel, a short vertical clip focused on the single most surprising step, and a checklist your audience can paste into their own notes.
Avoid promising that any tool will produce a print-ready file automatically. Teach the verification step, and your tutorial becomes trustworthy instead of promotional. If a tutorial shows a tool that made a mistake, that moment is often the most valuable part of the piece.
Too much information. If the card contains more than roughly seven pieces of discrete information, memory drops sharply. Cut the weakest two.
Tiny logos. A logo sized to look elegant on screen usually looks small in hand. Test with a physical printout at final size, at arm's length.
Insufficient contrast. Elegant low-contrast palettes fail under dim lighting. Push the difference between text and background further than feels necessary on a bright monitor.
No bleed or missing safe area. Text that sits close to the trim edge will eventually get clipped, and it will be the batch you already had printed.
RGB files sent to a CMYK press. Colors shift, and the shift is most visible in saturated brand colors. Convert early and proof if the brand color is the identity.
Oversized files with 1200 DPI scans. File size does not improve quality past the printer's effective resolution; it just slows production and can trip upload limits.
Unchecked generated details. An invented phone format or a misspelled name from a generated mockup reaching press is the most expensive error in this list, because it is invisible in the design and obvious to the recipient.
No proof. Order a single physical proof before a full run when color, stock, or finishing matters. It costs far less than reprinting an entire batch.
You do not need a large stack. You need one tool per job and a documented process.
Standardize the boring parts. Save a master document with bleed, safe area, and guides for each card size you use. Save a text file with your exact contact strings, in the format you will type them, so you never retype a number from memory. Save your export preset so a press-ready PDF takes one click.
Then write a short internal checklist for every new card or refresh:
That checklist is also your best tutorial outline. Publishing it turns an internal process into a resource, and resources attract exactly the audience you want holding your card.
Can AI design a business card for me?
Generative tools can produce strong starting directions and help with palette or pairing suggestions. The final file still needs real vector artwork, licensed fonts, verified contact details, and correct bleed and safe margins. Treat the tool as a concept partner, not a production department.
How much bleed should I add?
Three millimeters is standard in most metric markets, and 0.125 inches in North America. When in doubt, use the printer's own template — their cutting equipment defines the correct value, not a general rule.
What resolution do printed cards need?
300 DPI at final placed size for anything raster. Vector elements stay resolution-independent. Sending a 1200 DPI scan adds file weight without adding visible quality.
Should I use matte or glossy stock?
Matte and soft-touch stocks handle small text and deep colors more gracefully and feel more premium in hand. Gloss can make photographs pop but shows fingerprints and glare. For a text-heavy card, matte is usually the safer choice.
How many different fonts should a card use?
Two at most: one for the name and brand line, one for supporting details. A single well-chosen family with two weights often outperforms a pairing, and it reduces the chance of a licensing gap in the final file.
Is a QR code worth including?
It is worth it if it leads somewhere mobile-friendly and specific — a booking page, a portfolio, a location. It is not worth it if it simply points to a homepage the recipient can find by searching your name.
How often should I redesign a card?
Refresh when your positioning changes, not on a calendar. Practical triggers include a role change, a new primary offer, a service area shift, or a rebrand. Otherwise, keep the design and update the printing quality or stock instead.
What is the biggest avoidable error?
Unverified text. Names, numbers, domains, and postcodes copied from a generated mockup cause more reprints than any design decision. Read every character aloud against the source before export.
The short version: let generative tools widen your options at the start, then make the final thirty decisions yourself — measurements, contrast, type size, verified details. Your card is the cheapest long-lived marketing asset you will ever produce, and the workflow behind it is reusable content in its own right.

Alexander
Vidu Q4 alternatives compared: Kling 4.0, Seedance 2.5, Wan 3.0, and Runway Gen-4 — 4K support, input modes, durations, audio, and what each costs in credits on Domer.
The top 4K AI image generators on Domer compared: Flux 3 Klein, Nano Banana 2.1, GPT Image 2.5, GPT Image 2, and Nano Banana Pro — 4K credit costs, reference limits, and speed.
Nano Banana 2.1 vs GPT Image 2.5 Flare compared: credit costs per resolution, aspect ratios, reference images, text rendering, and which tasks fit each model.