Headcanon Examples for Appearance: One Trait, One Source

Short answer: an appearance detail only earns its place when you can name where it came from. Apply one rule — one visible trait, one stated source — and let the source be one of exactly three kinds: Use (what the trait is for), Origin (what made it), or Cost (what it takes to keep). Write the trait as something a camera could film in one shot, in one sentence. If none of the three fits, delete the trait; there is no fourth option, and decoration is not one either. The useful headcanon examples for a character's appearance are never a description of a face — they are one visible thing plus the reason it is still there.

The headcanon examples below are working examples, not showcase lines. Each one is paired with the bank line it was built from and the source that line already carried, so you can inspect the reasoning instead of grading the prose. Nothing here depends on taste. It is a filter you can run over a face you have already written, and it tells you in about four seconds which of your sentences were doing work.

One visible trait, one stated source Three sources, one sentence, and a delete test at the end. 1 · Use the mug only they may use 2 · Origin meals that came with a limit 3 · Cost a photo that stays face down One visible trait — the only column a reader can actually see Filmed in one shot. Written in one sentence. Never a mood. Delete test: if none of the three fits, the trait goes. An eye colour with no history behind it is a paint chip, not a character.

Three ways a visible trait can earn its place, and the test that removes it when none of them applies.

Why most appearance description reads as filler

A face delivered as a list of features is a stock-photo caption. Fair skin, dark hair, grey eyes, a scar above the left eyebrow. Each item is true, each item is visible, and not one of them came from anywhere. The reader has nowhere to put them. Nothing in the sentence can be contradicted, developed, or paid off later, so the description slides straight back out of memory.

Writers usually diagnose this as a length problem and respond by writing more. That makes it worse. The failure is not that there are too few details; it is that no detail is answerable to anything. A trait with no source is unfalsifiable, and the mind discards what it cannot test. Give the same eyebrow scar a source — a doorframe in a house they no longer live in, a fight they lost on purpose, a surgery they still owe money for — and it stops being scenery. It becomes a claim about the person that the reader can hold, agree with, or doubt. Doubt is fine. Doubt is engagement.

There is a second reason to run this filter early. Most of what you know about a character is invisible: how they decide, what they avoid, what they would apologise for. Appearance is the only part of a character that always arrives at the same time as they do. Spent well, it carries the invisible half into the room with them. Spent badly, it delays that half by a paragraph and teaches the reader to skim.

The rule: one visible trait, one stated source

Write the trait as a thing, not an adjective. Nervous is not a trait; a coin they turn over in their pocket is a trait. Then attach exactly one source, and make that source do a job the trait cannot do on its own.

One trait, one source. Not three traits and a mood. Not a paragraph of costume jewellery. If you can write the visible trait and the source in a single sentence of twenty words or fewer, you have a finished unit; if you need a second subordinate clause to make it land, you have probably picked a trait with no source and are patching it with adjectives.

What each source sounds like

The three sources are not interchangeable, and each one produces a different kind of sentence.

A trait can honestly carry more than one, but write them one at a time. Two sources in one sentence produces a paragraph that reads like a character file rather than a person standing in a doorway.

Six headcanon examples, built line by line

These six come straight out of the bank used by the generator on this site: 180 hand-written lines arranged as 5 categories by 6 tones, six lines in every cell. The lines below are quoted at the category and tone they live in, with the character's name filled in as Nadia so the sentences read properly; the bank swaps the name and changes nothing else. In each case the first job is to notice that the bank line already contains a source — and the second is to build the visible half that the bank does not supply.

Read the six together and a pattern appears. In every case the bank supplied a behaviour with a reason attached, and the appearance work was one noun, one gesture, or one position in a room. None of the six needed a new adjective. That is not a coincidence; it is what happens when a description is built from a source instead of a palette.

What a 180-line bank actually contains

The reason the method above works from bank lines rather than from features is measurable, and I counted it rather than assumed it. The full 180 lines of this site's bank were scanned line by line against a vocabulary list; here is what came back.

The practical reading of that table is blunt. A generator of this kind cannot hand you an appearance, because it does not contain one. What it hands you is a source: a reason, a history, a habit being paid for. Your job is the half a camera can see. This is the opposite of how appearance usually gets written, and it is why the outputs in the section above are short — the bank already did the expensive part, and the visible half is one noun.

It also explains a failure you have probably seen. If you ask a generator for a character and then reach for a mirror, you will end up describing whatever your own defaults are, because the tool gave you nothing to describe and you filled the gap from habit. The numbers above say the gap is 78.3% wide. That is not a flaw to complain about; it is the division of labour.

Five ways appearance prose dies

The four-slot sheet

Four slots, filled in order, and a sourceless trait never gets past slot two. Copy this into your notes file and run it once per character.

One trait per scene, and never two traits built from the same source type in a row. A character whose appearance is explained entirely by Origin reads as a history exhibit; one explained entirely by Cost reads as a diagnosis. Alternating is what makes a person look lived-in rather than documented. If you want a second opinion on how much personality you are actually carrying per line, the same fixed pool is broken down category by category in the 30-cell matrix.

Running it from the generator

The bank lines quoted on this page are real entries, and each one can be found at the category and tone given in brackets:

Leave the filters off and the pool widens to all 180, which is the right setting for hunting a line you cannot name and the wrong setting for filling a specific slot. If you have already used a line, do not reuse it in the same scene; the full set of slots works better when the trait you derived is not standing next to its own second draft. For the version of this workflow that starts from a name rather than a line, see reading a name before the first roll, and for turning a derived trait into something a character would actually say, see putting one line of dialogue back into the prose.

Quick answers

More headcanon guides