TYPETAP
MENU

The question

What Is My Type?

What is my type? The question sounds like it wants a label. Usually it wants a pattern: which faces make you stop, and which ones you keep scrolling past.

What people usually mean by “my type”

Ask ten people to describe their type and you will hear the same eight words in different orders — kind, funny, confident, tall, dark hair, good smile. The descriptions are not lies. They are summaries of something the person has never had to inspect closely, produced for a question that rarely matters in real life.

A type is not a checklist you carry around. It is a bias in attention. You notice certain faces in a crowd before you notice anything else about them, and that happens faster than language. Which is why asking for words is the wrong instrument: it measures how well someone can narrate their attention, not where the attention actually goes.

Where the phrase comes from

“What is my type” has meant at least four different things over the years, and each meaning asked for a different kind of answer. In astrology it was a prediction: your sign decided who you were compatible with. In personality tests it was a self-report: answer forty questions and get four letters. In dating apps it became a filter — a list of acceptable attributes, typed into a search box. In conversation it is usually none of those. It is a way of asking why the same kind of face keeps catching your eye.

Those four meanings get mixed together, which is why the question is hard to answer honestly. A prediction cannot be checked, a self-report can be flattered, and a filter describes what you would accept rather than what you actually notice.

Why word-based quizzes keep missing

A quiz that asks “do you prefer sharp features or soft features?” assumes you know. Most people do not. The preference is real, but it is comparative and situational: it shows up when two faces are side by side, not when one adjective is held up alone.

There is a second problem. Words carry status. “Sharp” sounds sophisticated, “soft” sounds safe, and the answer someone gives often reflects the person they would like to be rather than the faces they actually tap. A test built on self-description measures self-image twice: once in the question and once in the answer.

Is a type even a stable thing?

Partly, and the honest answer is worth more than a confident one. What stays stable is coarse: the shape of a face, the weight of an eye, the overall register — clean or rugged, soft or sharp. What moves is finer than that. Mood, novelty, the pool you happen to be shown, and whether you are comparing two faces or judging one alone all shift the answer.

So a type is better thought of as a distribution than as a point. Two runs of any honest test will agree more on the headline than on the three specific faces it picks, because the headline is built from tags that repeat across your taps while the faces are a sample. A test that gives you one word and no sense of how firmly it holds is overselling what it measured.

What tapping faces measures that words do not

A face test removes the vocabulary step. You are shown twelve faces and asked for one binary action: tap or don't. There is no scale to game, no adjective to misinterpret, and no time to construct a flattering answer. What remains is the same bias in attention that operates in real life.

The useful signal is not only what you tap. It is also what you skip. In a twelve-face grid, every face has been looked at; a skip is a real negative judgement, not missing data. After six rounds there are enough of both to fit a small model and see which tags — hair colour, length, brows, lips, eyes — keep showing up on the positive side.

That is the whole trick, and it is why the face test asks for taps rather than adjectives.

The words a result actually uses

Once you look at what people tap, the vocabulary stops being abstract. A result names three things, and they are the three a tap can settle:

A type name in this vocabulary looks like “Blonde · Long straight · 20s”. Not poetry, but it is checkable: you can point at the faces that produced it, and so can someone else. That is the difference between a label and a description of what you actually did.

Only those three appear in the headline for a reason. They are the features the test controls when it builds a face, so a preference in them is a preference and not an accident of which faces happened to be in the pool. Skin tone, lips and eyes are measured from the images rather than assigned, so they stay in the supporting detail instead of becoming the headline.

What a result can and cannot tell you

It can tell you which visual traits you consistently favour, how selective you are relative to the faces you were shown, and which mainstream faces you walked past without noticing. It can also tell you how much your taste overlaps with someone else's, which is a more interesting number than either of you expected.

It cannot tell you who to date, whether a relationship will work, or what kind of person you deserve. A preference for a jawline is not a compatibility test. The result describes attention, not judgement, and it is deliberately built to avoid ranking people — including the faces in the pool, which are AI-generated and carry no real identities.

One more thing it cannot do: it cannot tell you what your type is in the sense a horoscope would. There is no hidden category you belong to. There is a pattern in what you looked at, and the pattern is the answer.

How to work it out without a quiz

If you would rather work out what is my type on your own, the same signal is available directly — just slower. Three habits are enough.

Compare in pairs rather than in the abstract. “Do I like tall people” has no answer; “of these two, which would I notice first” does. Keep a note of the faces you actually stop on for a week, rather than the ones you describe when asked. And when you do describe someone, write down the feature you saw before you knew anything about them — that first detail is usually the honest one.

What you are looking for is repetition, not intensity. One face you found striking is an event. The same jaw, the same eye, the same register in five different people is a type.

So what is my type, in one sentence

A type is the set of features you notice before you know anything else: a face shape, an eye area, a register. It is stable enough to describe and loose enough to surprise you.

If someone asks “what is my type” and expects a category — a sign, four letters, an archetype name — the honest reply is that categories are other people's abbreviations for this pattern. Useful for talking, useless for deciding.

Which is also why the answer is worth checking rather than accepting. You can read a description of a type and nod along. You cannot nod along to a face you did not tap.

If you want your own answer

Reading about what is my type will only get you so far, because the interesting part is comparative. Six rounds, about ninety seconds, no account: take the what's my type quiz and see what you actually tap.

This page explains the idea. The quiz itself runs entirely in your browser; the test you may have heard about is on the front page. It is called TypeTap, and it is not affiliated with mytype.fyi.