Couple compatibility

The “Couple compatibility” read is the paid part of a pair test: you and your partner answer the same 62 statements about the same relationship SEPARATELY, code computes two independent axes and thirteen facets for each of you, and then the map of divergences between the two pictures plus the blind spots where a prediction about the partner missed what they actually reported. AI writes three documents from those numbers: your read, your partner’s read and the shared one. There is no compatibility percentage here and never will be: research shows that how well partners match on traits is almost unrelated to relationship satisfaction. One payment covers both, and one side’s answers are never shown to the other.

Go to the test
  • paid decodings completed
  • 5–9 minaverage decoding time

Services

What else we decode

Not just this area — upload any data and get a clear breakdown in minutes.

Scope

What we analyse

  • Two people answer — otherwise there is no product

    Almost every “couple” test surveys one person and issues a result “for both”. What is measured here cannot be measured alone: the divergence between two pictures of the same relationship. The first one takes it and pays; the second gets a link, already paid for.

  • A map of divergences instead of one figure

    For each of thirteen facets you see your position, your partner’s, and the gap multiplied by how much that facet weighs. That arithmetic sets the order of the read, not what sounds more interesting. Agreement counts as a finding in both directions: if both say something is thin, there is nothing to argue about — and that is usually the easiest thing to move.

  • Blind spots — what nobody else has

    Each of you fills in five facets twice: for yourself and as a prediction of your partner. A missed prediction reveals a belief about the other person that their own answers do not support — in both directions: assuming things are fine matters as much as assuming they are not.

  • Your partner’s answers stay closed

    Neither the read nor the chat quotes or paraphrases a single answer from the other side. That is not politeness but a condition of the thing working: if answers were visible, the second person would answer to match, and there would be nothing left to measure.

  • Two axes, not a good–bad scale

    Connection and strain are computed separately: a couple can be very warm and very strained at once, and that is not the same as warm and calm. The point lands in one of four areas — the name of a region, not a grade for the couple.

  • Conversations with a ready opening line

    The output is not a list of recommendations but specific conversations: what to raise, why it is worth raising given your map, and an opening line that does not sound like an accusation. There are no instructions to hand to your partner.

  • Safety is free and always

    If the answers carry fear, control, coercion or a circle of people that has narrowed, you see it before paying anything. That is not solved by learning to communicate better, and the read says so plainly, without diagnosing anyone.

How it works

Three steps to a clear result

  1. Upload your data

    Take a photo of a test or scan, upload a PDF — or just describe your symptoms. Several files at once, no sign-up.

  2. AI analyses everything

    It reads test values and findings on images and checks them against reference ranges. For tests from different dates, it tracks the trend.

  3. A clear result

    What’s normal, what’s off and what it means — in plain language, with advice on which doctor to see.

Useful materials

Related articles and parameters — normal ranges, interpretation and causes of abnormalities.

Ask AI about health

FAQ

Common questions

  • No, never. They receive only the shared read, which compares pictures rather than individual answers. Neither the read nor the chat quotes or paraphrases the other side’s answers.
  • Because it would mean nothing. A large study across 43 datasets and 11,196 couples found that how well partners match on traits barely predicts their satisfaction — a partner’s traits explain about 5%. Such a number would look convincing and be an invention. Instead: a map of where the pictures agree, where they part and how much that weighs.
  • No. One payment covers both and is made by whoever goes first. The second opens a link and their test is already paid for.
  • Your full read stays with you — you get it right after payment and do not depend on your partner. The link never expires and can be sent whenever.
  • Free, you get your half of the map: both axes, the point with the name of the area, and a safety warning if one is needed. The paid read is your own read of thirteen facets, a separate read for your partner, the shared map of divergences, your blind spots and concrete conversations. Plus a PDF for both and an AI chat about your own read.
  • No. It is an educational tool. It makes no diagnosis, does not grade you as a couple, does not say whether you suit each other and does not predict whether the relationship lasts. The test has no normative sample: a position of 0–100 is how pronounced something is in your answers, not a comparison with other couples.
  • Yes. We ask nothing that does not affect the score, the questions read the same way for any two people, and the composition of the couple is never discussed in the read: research shows the predictors of relationship quality are the same.

Online decoding

Start your decoding now

Upload a photo or PDF — no sign-up, results in minutes.

Go to the test

The service is informational and not intended to diagnose emergency, oncological or psychiatric conditions. For acute symptoms, call emergency services (112).