You’re not broken. The test is doing something very specific — and once you see it, the flipping letters start to make sense.
If you’ve taken a 16-type test more than once, there’s a good chance you’ve watched a letter or two change. INFP one year, INFJ the next. ENTP on Monday, INTP after a rough week. People usually take that personally — maybe I don’t know myself — or they conclude the whole thing is nonsense. Neither is quite right.
Every one of the four MBTI dimensions is a spectrum. Nobody is 100% introverted or 100% a “thinker.” Your real score on each dimension is a point somewhere along a line, and for most people, at least one of those points sits close to the middle.
The test, though, doesn’t give you the point. It gives you a side. It draws a line down the center and sorts you left or right. If your true position is 52% Feeling, you’re an F. If it’s 48%, you’re a T — and you get a completely different four-letter label, a different profile, a different set of career suggestions.
Now imagine retaking it after a stressful week, or reading one question a little differently. A few answers shift, your score drifts three points, and you’ve crossed the line. Your personality didn’t change. The cutoff caught you on the other side.
Not all four dimensions are equally shaky. In the reliability data, Thinking–Feeling is consistently the least stable scale, with Judging–Perceiving close behind. That’s why the most common “wait, which am I?” pairs are INFJ/INFP, INTJ/INTP, ENFP/ENTP — types that differ on exactly those dimensions.
Here’s the reframe that helps people most: the letter that never changes is your clearest signal. If your N has held across five tests while everything else wobbled, that’s not a flaw in the test — that’s a strong, real preference. And the letters that flip aren’t random noise; they’re telling you that you live near the middle on that dimension, which is its own kind of information. Someone who is 51% F genuinely does think and feel their way through decisions in nearly equal measure.
I’m a therapist, and I’ve watched people carry a four-letter type around for years like a diagnosis — then feel destabilized when a retest hands them a different one. My honest take, after a lot of these conversations: you are more than a label, and no questionnaire captures how you show up across every context of your life. You may lead with Thinking at work and Feeling with your kids. Both are you.
The healthiest use of any typing system is as a mirror for self-reflection and a shared vocabulary for talking about differences with the people you love — not as a verdict. If a result helps you understand a recurring pattern, keep it. If it stops fitting, that’s allowed.
I’ll be straight about what this is and isn’t. The Insight Game isn’t claiming to be a more “accurate” questionnaire — no test escapes the problem above. It’s a quick, card-based way to explore type that takes about five minutes and is honestly kind of fun to play.
What it does differently is refuse to hide the coin flip. When your result is close on a dimension, the game tells you so and points you to both neighboring profiles instead of pretending a 51/49 split is a confident answer. We also publish our live community accuracy data — how often results agree with players’ known types — on the research page, so you can judge for yourself rather than take our word for it.
The descriptions behind it come from the clinical Type and Temperament work of Dr. Martine J. RoBards, Ph.D. (1946–2017), whose framework The Insight System is built on.
More honest answers: browse all the type guides →
Already know your type? Browse the 16 type profiles — and if you’re between two, read both.