Personalised Raksha Bandhan Card With Name and Photo

Personalised Raksha Bandhan Card With Name and Photo

For a personalised Raksha Bandhan card, pick a photo with one clear subject whose face fills at least a third of the frame, check you have the person's permission if the photo is of them rather than of you, and divide the photo's pixel width by the card's width in inches before trusting it in print. Those three decisions matter more than which design you pick. Across 2,739 pages created on SubhSandesh since 2026-03-12, 51.8% of the 33,007 recorded views were opened on a phone, so most cards are judged at about the size of a playing card. The festival falls on Friday 28 August 2026, so there is time to choose properly rather than upload whatever sits at the top of your gallery. This page is about which photo, not where to put it.

What a personalised card with a name and a photo actually is

It is a template with two variable fields: a text field for the recipient's name, and an image frame for a photo you supply. The frame has a fixed shape and size, and your photo is fitted into it — cropped, scaled, and on some tools compressed.

Whether it is printed or sent as a link is a separate decision, covered elsewhere on this blog. Either way the photo is the part you cannot quietly fix later — the only element carrying information the template did not supply.

Four checks to run before the photo goes near the card

Run them in order. Checks two to four each get a fuller section below.

1. One subject, and a face that is large on the card

Check how much of the frame the main face occupies, holding your thumb over it at the size the card will be seen. Recognition is not the problem — familiar faces survive severe degradation — but a face 12mm wide carries no expression. If it fails, crop tighter or pick another photo. The group shot everyone loves usually fails here.

2. Permission, if the photo is of an adult who is not you

Check whether the person has seen the photo and is fine with others seeing it on a card. Send it and ask. They have a reasonable claim over their own image, and a family group chat is an audience of dozens. If it fails, swap the file.

3. Pixels divided by inches, not "300 DPI"

Check the pixel dimensions against the card's physical size: 2,000 pixels across a 5-inch card is 400 pixels per inch. A resolution figure means nothing until a physical size is attached to it. If it fails, print smaller.

4. Your own crop, before the tool crops for you

Check what the frame will cut off by cropping a copy yourself to its proportions. Automatic cropping takes the centre, which on a standing portrait is a torso and a wall. If it fails, choose a photo shot in the right orientation.

Consent, and the childhood photo problem

The photo of your brother aged seven with a bowl haircut is funny to you. Whether it is funny to him is his call, and the answer changes when the card lands in a family group where forty people see it.

The closest evidence concerns parents posting photos of children, and it transfers. A 2025 systematic review of 21 sharenting studies across 13 countries found the same split repeatedly: people are pleased by images that show them well, and resent images shared without consent, especially personal or embarrassing ones, because it cuts across an identity they manage themselves. An adult sibling has been managing that identity for years longer than a child has.

There is a twist in the other direction. Across two studies with 610 participants, people chose more flattering images of an unfamiliar face than of their own; self-selection produced the weaker impression. Your judgement of which photo of your brother looks good is probably better than his — which argues for choosing on his behalf, not for skipping the question.

The working rule: photos of the two of you together need no conversation, photos where he is the subject do, and childhood photos of an adult always do. One message settles it.

Resolution: what your photo can actually print

Resolution has no meaning as a single number. The US Federal Agencies Digital Guidelines Initiative says so plainly in its 2023 guidelines for digitising cultural heritage materials: an object captured at 400 ppi renders at its correct size only when printed at 400 ppi. Pixels are a count. Sharpness is that count divided by a physical dimension.

The same document is the reason to stop repeating "300 DPI" as if it were the standard. For prints and photographs FADGI sets 200 ppi at its lowest star level, 250 at the second, 400 at the third and 600 at the fourth, measured at the original's size, and expects photographic masters at three stars or better. The figure most people quote from memory sits between two of its rungs.

A modern phone clears this easily. A 12-megapixel photo is about 4,032 by 3,024 pixels, which is 576 ppi across a 7-inch card, with pixels to spare for a crop. What fails is a copy of a copy — a screenshot, or an image forwarded through a chat app and re-saved smaller each time.

Pixel count is rarely the binding constraint; provenance is. Upload the original file, never the copy that came back to you on WhatsApp.

Why the group photo is usually the wrong choice

Not because groups perform badly. In an analysis of 1.1 million Instagram photos, images containing a face drew 38% more likes and 32% more comments than images without one — but the number of faces made no significant difference, nor did their age or gender. There is no engagement case against a group shot.

The case against it is arithmetic. A 5-by-7-inch card at 400 ppi is 2,000 pixels across. One subject filling a third of the frame gives a face about 660 pixels wide, roughly 42mm on the card. Six people shoulder to shoulder gives each face around 200 pixels, about 12mm.

Both are recognisable. The IEEE review Face Recognition by Humans reports that people identified more than half of a set of familiar faces at an equivalent resolution of just 7 by 10 pixels, reaching ceiling by 19 by 27 pixels. Recognition is cheap; expression is not. At 12mm you cannot tell whether anyone is pleased, and a card showing six half-legible people is not about the person whose name is on it.

Cropping a vertical phone photo into a landscape card

A phone photo is 3:4 tall; a landscape card frame is nearer 7:5 wide. Fitting one into the other costs about half the height, and the automatic crop takes the middle.

Do it yourself: crop a copy to the frame's proportions, decide what leaves the frame, and upload that so the tool has nothing to guess. If the composition cannot survive that, use a design built for the orientation you have — the animated Raksha Bandhan page for a brother is built around a phone-shaped photo, though it is written brother-first, so a card for a sister needs different wording.

Expect more than one pass. The median gap between a page's first save and its last edit is 6.9 hours, sampled across 914 pages, and the crop is one of the things people return to.

When a photo makes the card worse

Three cases, plus one where we are the wrong tool.

A formal card to an elder — a chacha, a teacher, a mother-in-law you address as aap — usually reads better with only a name. A photo makes it familiar, and familiarity is not always the register you want here.

A year with a death in the family. A smiling photo from last Rakhi, in a house that has lost someone, can land as an accusation rather than a memory.

A photo that dates the relationship. If the newest picture of you both is from a 2019 wedding, the card says out loud how long it has been. Occasionally that is the message; usually not.

And if the recipient has no smartphone, a link is the wrong artefact — a printed card wins. Even our own defaults work against them: 44.3% of creators password-protect the page before sharing (1,213 of 2,739), one more thing to walk an 80-year-old through on the phone.

Old prints and low-resolution photos: what is salvageable

Scan, do not photograph. A flatbed at 400 ppi across a 4-by-6-inch print gives 2,400 by 1,600 pixels, enough to reprint at original size. A phone held over the same print adds keystoning, reflections and your own shadow.

Then accept what cannot be recovered. FADGI is blunt that every generation of photographic copying involves some quality loss, and no upscaling tool reverses it — inventing pixels is not recovering detail.

That is often fine, because recognition survives degradation: a soft old photo of two children still reads as those two children. What does not survive is enlargement. Keep the old scan small and the sharp modern photo large, or send the old one on its own with a line saying where it came from. The SubhSandesh template gallery shows how much frame each design gives a photo.

Sources

• White, Sutherland & Burton, "Choosing face", Cognitive Research 2:23 (2017) — OA copy, German National Library.
• Motevalli et al., "Parents' Sharenting Behaviours: A Systematic Review", F1000Research 14:448 (16 July 2025).
• FADGI, Technical Guidelines for Digitizing Cultural Heritage Materials, 3rd ed. (9 May 2023).
• Sinha, Balas, Ostrovsky & Russell, "Face Recognition by Humans", Proceedings of the IEEE 94(11) (November 2006).
• Bakhshi, Shamma & Gilbert, "Faces Engage Us", CHI '14 (April 2014).