Business Card Photo: Should Your Face Be on It, and Will It Survive Print?

A business card photo is standard in professions where the client is choosing a person rather than a firm — real estate, insurance, financial advice, recruiting — and unusual almost everywhere else. If you use one, build it for print: around 300 pixels per inch at final size, converted to CMYK, cropped tight enough to read at one inch.

Those two halves of the question get answered in completely different rooms. The first is a judgement call about your industry's conventions, and the honest answer is that reasonable people in the same profession disagree. The second is a technical problem with correct and incorrect answers, and it is where most people get burned — because a file that looks flawless on a phone can come back from the printer soft, dark and slightly muddy, and by then you have a thousand of them in a box.

Which professions actually put a face on the card

The convention splits along a single line: whether the buyer is choosing you or choosing your employer. Where the transaction is personal, infrequent and high-trust, a face on the card is expected and its absence goes unnoticed. Where you are one interchangeable representative of a larger brand, the same photo reads as a small act of self-promotion.

Where a face is the norm

• Residential real estate. The most photo-saturated category there is. Agents are sole traders competing for a decision the typical seller makes about once a decade — the National Association of Realtors' 2025 Profile of Home Buyers and Sellers puts median seller tenure at 11 years, an all-time high — so the photo has to do its recognition work over a very long gap. We cover the profession-specific case, across cards, yard signs and portal thumbnails, in our guide to real estate headshots for signs, portals and cards.
• Insurance and financial advice. The product is functionally similar across providers, so the differentiator is the person. A face on the card is a claim that you are the one who will pick up the phone.
• Recruiting and staffing. Recruiters meet dozens of people in a day at fairs and on-site visits, and hand out cards in volume. The photo is a sorting aid for the candidate holding six cards afterwards.
• Solo and small-practice legal and medical. Personal injury, family law, dentistry, chiropractic, cosmetic practice. Common in the US, though it varies sharply by market and by how the practice positions itself.
• Speakers, coaches, consultants and anyone whose name is the business. If the brand and the person are the same entity, the photo isn't self-promotion — it's the logo.

Where it reads as unusual

• Corporate roles inside a recognised brand. A card from a bank, a manufacturer or a professional services firm carries the institution's mark, and a face competing with it looks like a personal side project.
• Engineering, technical and scientific roles. Cards here are functional data carriers. A photo signals that you thought the card was about you rather than about the contact details.
• Design, architecture and creative agencies. The card is itself a portfolio piece; the stock, the typography and the restraint are doing the talking. A headshot on it is the visual equivalent of over-explaining.
• Luxury and premium categories. Attention is deliberately routed to the brand. Material quality and spacing carry more weight than any image of a person.
• Anyone in a large team who will be replaced on the account. If a colleague may take over the relationship, a card built around your face makes the hand-off awkward.

The messy middle is real. B2B sales, mortgage brokering, franchise owners, freelancers in service categories — these sit on the line, and the answer is usually decided by what the rest of your local market does. Collect ten competitor cards before deciding. Convention is a signal, and being the only person in the room whose card has a face on it communicates something whether you intended it or not.

The recognition argument, and what the research actually says

The standard defence of the business card photo is that people remember faces better than names. It is repeated everywhere, and it turns out to be shakier than it sounds. When psychologists tested both under the same conditions, names came out ahead — which changes what you should reasonably expect a card photo to do for you.

Mike Burton, Rob Jenkins and David Robertson at the University of York compared memory for faces and names using the same recognition framework for each, publishing in the Quarterly Journal of Experimental Psychology in 2019. Summarising the results, Jenkins said participants scored consistently higher at remembering names than faces, “recognising as little as 64% of faces and up to 83% of names.” The everyday intuition is misleading, the authors argue, because we only notice a forgotten name after we have recognised a face — we walk past the faces we failed to recognise entirely and never learn about it.

So a photo on your card is not buying you superior memory encoding. What it is genuinely buying is narrower, and still worth having:

• Identification at re-contact. Someone pulls your card out of a drawer four months later and can attach it to a specific meeting rather than a name they have half-forgotten.
• Disambiguation in a stack. After an open house, a conference or a job fair, the recipient has eight cards and no way to tell which conversation was which. The photo is the index.
• Continuity across channels. If the same portrait appears on your card, your LinkedIn profile and your email signature, the person who searches for you afterwards lands on an obvious match rather than a maybe.
• A commitment signal. In categories where anonymity is a red flag — anyone handling your money, your house or your health — showing your face is a low-grade accountability gesture.

None of those require the photo to be large. That matters when we get to layout, because the most common failure is giving a photo a third of the card when a sixth would have done the same job with better typography around it.

The three costs nobody prices in

Cards are typically ordered a few hundred at a time and then live in a drawer for two or three years. That long shelf life turns two of the following into slow-burn problems, and the third is a design constraint you feel on day one.

1. 1The photo ages, and the card doesn't. A card handed over in year three shows a version of you from year zero. That is a mismatch the recipient notices in the first two seconds of a meeting, and it undercuts exactly the recognition benefit the photo was there to provide. Our piece on how often to update a profile photo lays out the honest refresh interval; a card is less forgiving than a profile, because you cannot swap it silently.
2. 2It locks the relationship to you personally. That is the point if you are a sole practitioner, and a liability if you are building something with partners or associates. Clients who chose the face expect the face.
3. 3It eats a third of a seven-square-inch canvas. A standard US card is 3.5 × 2 inches. Every square inch the photo takes is space the name, the role, the phone number and the whitespace no longer have. Cards that look cramped are usually not badly typeset — they are photo-first layouts with the same amount of text squeezed into two thirds of the room.

A five-question test

If you are still undecided, this settles it faster than another hour of looking at templates. Answer honestly and count the yeses — three or more and the photo earns its place; two or fewer and you are better served by better paper and better typography.

1. 1Does the client choose me rather than my company? (Sole trader, named practice, commission-based, personal referral pipeline.)
2. 2Do I hand cards to people who will meet several near-identical competitors in the same week?
3. 3Would a peer in my industry find a photo on my card unremarkable? (Check ten of their cards, not your assumption.)
4. 4Will the same portrait be live on my website, LinkedIn and email signature, so the card isn't an orphan image?
5. 5Am I willing to reprint when the photo stops looking like me — realistically, inside three years?

Print resolution: why a file that looks sharp on a phone prints soft

This is where most business card photos fail, and it fails silently — nothing on your screen warns you. Print packs far more detail into the same physical inch than a display does, so an image that fills a phone screen beautifully can carry only a fraction of the data a card needs at the size you are placing it.

300 DPI (dots per inch) at final print size is the working standard, and it is what commercial printers ask for in their own specifications. MOO's business card design guidelines state that photographs should be “300dpi or higher.” PrintingForLess recommends 300 DPI for images and 400 DPI for any image containing text, notes that “computer monitors generally have a display setting of 72 DPI or 96 DPI,” and warns outright: “Do not save images or graphics from a website to use in your print project!”

The arithmetic is simple and you should do it before you place anything. Multiply the printed size in inches by 300 to get the pixels you need:

• The whole card at trim size: 3.5 × 300 = 1,050 px wide; 2 × 300 = 600 px tall.
• The whole card including bleed: MOO gives 1,098 × 648 px at 300 dpi as the ideal image size for a standard card, which is its 3.66 × 2.16 in bleed area.
• A small corner inset, 1 × 1.25 in: 300 × 375 px minimum.
• A vertical panel down one third, 1.2 × 2.16 in (bleeding top and bottom): roughly 360 × 648 px.
• A full-bleed photo covering half the card, 1.83 × 2.16 in: roughly 549 × 648 px.

Why a LinkedIn download often isn't enough

LinkedIn's recommended profile photo size is 400 × 400 px — the bottom of an accepted range that runs all the way up to 7680 × 4320 px. Run the low end through the same maths: 400 ÷ 300 = 1.33 inches. A 400 × 400 file prints acceptably at up to about 1.3 inches square and not one millimetre more. That is fine for a corner inset and nowhere near enough for a panel or a full-bleed half-card. Screen specs and print specs are different problems — we keep the on-platform numbers in our reference on LinkedIn photo sizes and specs, and none of them are print specs.

Two related traps. First, what you download back from a platform is often a re-compressed derivative smaller than what you originally uploaded — always go back to the original file, not the version the site served you. Your own upload may have been 3,000 px; the copy you right-click and save may not be. Second, upscaling doesn't work. Enlarging a 400 px image to 1,000 px invents detail that was never captured; you get softness with sharpening artefacts on top, which prints worse than plain softness.

Whatever the source, check the file you actually have

This applies equally to a photographer's gallery proof, a phone selfie and an AI-generated portrait. Open the file, read the pixel dimensions, divide by 300, and confirm the result is bigger than the space you are placing it in. Web previews and thumbnails are not print files regardless of how they were produced — if a tool distinguishes between a preview and a full-resolution download, the download is the one that goes to the printer.

CMYK, RGB and what happens to skin tones

Screens make colour by adding light; presses make it by layering ink that absorbs light. The two gamuts do not overlap perfectly, so some colours you can see on a monitor simply cannot be mixed from cyan, magenta, yellow and black. The conversion happens whether you manage it or not — the only question is whether you see the result before the run.

Vistaprint's own guidance is to convert to CMYK for print, warning that with an unconverted RGB file “bright and neon colors lose their pop” and your print “might look less vibrant.” Left to an automatic conversion at the printer's end, out-of-gamut colours get approximated, typically landing duller and flatter than what you approved on screen.

On a portrait specifically, here is what to expect and what to do about it:

• Skin itself usually survives. Neutrals, greys, pastels and most skin tones sit inside the CMYK gamut and convert without drama when the file is prepared properly. This is the reassuring half.
• Shadows are the risk. Dark tones plug up — detail in the shadow side of a face, in dark hair, in a navy jacket can collapse into a flat mass that looked fine backlit on a screen. Lift shadows slightly before converting rather than after.
• Saturated brand colour is where you will actually see the shift. The vivid teal or electric blue behind the head is far more likely to disappoint than the face in front of it. Test that colour, not the skin.
• Heavy grading travels badly. A portrait with a strong warm or cinematic colour treatment has already pushed tones toward the gamut edges. A neutral, accurately white-balanced file converts more predictably.
• Convert deliberately and look. Do the conversion yourself, view it, and adjust — rather than uploading RGB and discovering the printer's interpretation on 500 cards. For a large or colour-critical run, ask for a hard proof on the actual stock; a monitor proof does not account for how the paper takes ink.

One more paper variable: uncoated stock has no clay or varnish layer, so ink soaks into the fibres and the halftone dots spread — dot gain. The practical result is lower contrast and duller, softer reproduction than the same file on coated stock. If you have chosen a beautiful uncoated card, expect the portrait to lose a little snap, and pick a brighter, cleaner original to compensate.

Bleed, trim and the safe margin around a face

Cards are printed on large sheets and cut in stacks, and the guillotine drifts by a small amount card to card. Two conventions exist to absorb that drift: bleed, which extends artwork past the cut line, and the safe area, which keeps important content away from it. A face interacts with both.

The exact numbers vary by printer, so take them from your printer's own template rather than from a blog — including this one. Two real examples show the spread:

• MOO, standard card: trim 3.50 × 2.00 in, bleed area 3.66 × 2.16 in, safe area 3.34 × 1.84 in — 0.08 in of bleed and 0.08 in of safe margin per side.
• Jukebox Print: 0.125 in (3 mm) added per side, giving a 3.75 × 2.25 in document with a 3.25 × 1.75 in safe area.

What that means when the artwork contains a head:

1. 1If the photo bleeds off an edge, extend the image itself into the bleed — not a background colour approximating it. A photo that stops at the trim line leaves a white sliver on any card the blade cuts slightly wide.
2. 2Keep the eyes and the top of the head inside the safe area. A crop that puts the crown of the head exactly at the trim line will lose it on some cards in the stack and not others, which reads as an error rather than a style.
3. 3Never place text over the face near an edge. Your name overlapping a portrait is already a risky move; near the trim it becomes a lottery.
4. 4Mind the minimum type sizes. MOO specifies keeping all text at 8pt or higher and sharp lines at 0.5pt or thicker. Cards fail these limits precisely because a photo took the space and the designer shrank the type to compensate.

Why a tight crop on a light background reproduces better

At its printed size a card photo is somewhere between one and two inches tall. That is smaller than a postage stamp's worth of face, and it eliminates most of what makes a portrait look good at full size. Two choices decide whether it survives the reduction: how tight the crop is, and how much separation there is between subject and background.

Go tighter than feels comfortable. At one inch, a head-and-shoulders crop is the only framing that resolves — anything wider turns the face into a few dozen pixels of feature detail surrounded by information nobody can see. If you are unsure which crop you are even looking at, our explainer on what counts as a headshot and what each crop is for sorts the terminology out.

On backgrounds, the light-versus-dark question isn't aesthetic here, it is mechanical:

• Dark hair against a dark background merges. At full size your eye reads the subtle tonal edge; at one inch on absorbent stock, that edge disappears and the head becomes a silhouette with facial features floating in it.
• Heavy ink coverage shows its flaws. Large dark areas need a lot of ink, and they are where banding, scuffing and fingerprints become visible — particularly on coated stock that people will handle.
• Light backgrounds carry the card's own colour. A pale grey or off-white behind the head lets the card stock and brand palette do the work instead of fighting a dark rectangle.
• Even, soft light beats dramatic light. Moody side lighting throws half the face into shadow, and shadow is exactly what plugs up in CMYK. Our plain-English guide to headshot lighting covers how to get flat, even coverage without it looking sterile.

If you are choosing a background from scratch rather than working with what you have, the background guide walks through the options by purpose — it is the companion piece to this one, covering the aesthetic decision where this article covers the print constraint.

Where the photo goes, and how much card it should take

Layout is where the judgement call and the technical constraint meet. The photo needs enough size to be legible and little enough that the card still functions as a contact card. Three arrangements cover almost everything you will see in the wild, and they differ mainly in how much they commit.

The corner inset

A small square or circle, roughly 0.8 to 1.2 inches, sitting in a corner or beside the name block. The most conservative option and the one that ages best, because it is small enough that a slightly dated photo doesn't dominate. It also has the lowest resolution requirement — a 400 × 400 px file is genuinely adequate here. Circular crops need care: a circle mask cuts more off the corners than people expect, so leave headroom.

The vertical panel

The photo occupies the left or right third of the card, bleeding top and bottom. This reads as deliberate design rather than an add-on, and it gives the face real presence without overwhelming the layout. It needs a portrait-orientation source image with room to crop — a square export usually doesn't have the vertical extent, and stretching it is not an option.

The photo back

Contact details on the front, portrait on the reverse. Underrated. It gives the photo a full 3.5 × 2 inches, keeps the front clean and typographically generous, and lets you refresh one side of the design later without touching the other. It also removes the whole problem of a face competing with a logo. The trade-off is that a good share of the people you hand it to never turn it over.

Whichever you choose, apply one proportion rule: your name should remain the largest single element on the card. If the photo out-competes the name, the card has stopped being a contact card. As a rough ceiling, keep the photo under a third of the front face's area, and check the result at arm's length rather than at 400% zoom on a monitor — that is the distance at which it will actually be read.

Where the photo should come from

A card photo is a low-demand application in one sense — it is small — and a high-demand one in another, because print is unforgiving about resolution and colour in ways a screen never is. That combination changes the calculation slightly compared with sourcing a photo purely for online use.

• A photographer's file is the safest input for print. You get a large, properly exposed, colour-managed original, and a good photographer will hand over a file suitable for CMYK conversion if you ask — which is exactly the part an amateur file most often lacks. Our honest benchmark of what headshots cost covers the realistic range and what changes it.
• A careful DIY shot can work at card sizes. A modern phone camera has far more resolution than a card needs, so the constraint is lighting and background control rather than pixels. Shoot in soft, even daylight against a plain wall, and keep the full-size original rather than a copy that has been through a messaging app.
• AI-generated portraits are a reasonable option with two caveats. Check the output resolution against the maths above rather than assuming, and be aware that generative models still handle some hair textures, headwear and facial hair less reliably than others — we document those failure modes candidly in where AI headshot models still fail. Print magnifies rendering artefacts that a small screen forgives, so inspect at 100% before committing.
• Whatever the source, keep the master file. You will need it again for the reprint, the website, the conference programme and the badge. Archive the largest original you have, not the version you cropped for one purpose.

Pre-flight checklist

Run this before you upload. It takes four minutes and it is the difference between a card you hand over confidently and a box of 500 you quietly stop using. Every item here corresponds to a specific failure that shows up only after the run is printed.

1. 1Pixel maths done. Printed width in inches × 300 ≤ the file's pixel width. If not, shrink the photo area or find a bigger file. Do not upscale.
2. 2Downloaded from the original source, not re-saved from a website, a messaging app or a social platform's served copy.
3. 3Converted to CMYK and visually checked, with shadow detail intact and any saturated background colour verified rather than assumed.
4. 4Bleed extended if the photo touches an edge — the image itself, into the printer's stated bleed area.
5. 5Eyes and crown inside the safe area, using the numbers from your printer's own template.
6. 6Crop tightened to head-and-shoulders, with background separation you can still see when you zoom the layout out to actual size.
7. 7Type still legal — 8pt minimum, 0.5pt minimum line weight — after the photo took its space.
8. 8Photo consistent with the one on your website and LinkedIn, so the card and the search result agree.
9. 9Proof ordered for any run over a few hundred, on the actual stock you have chosen.

The bottom line

A photo on a business card is a convention, not a rule, and conventions are local. In a profession where clients choose a person, leaving it off costs you nothing but gains you nothing either; in a profession where they don't, putting it on is a small, visible misread of the room. Look at ten competitor cards and the answer will be obvious.

The production side is not a matter of taste. Roughly 300 pixels per inch at final size, a CMYK conversion you have actually looked at, artwork extended into bleed, face inside the safe area, tight crop, light background. Get those six right and the card comes back looking like the file you approved. Get any of them wrong and you will spend three years handing out something you apologise for.