Disclaimer

What this is not

Between Us does not provide, and is not a substitute for:

• therapy, couples counselling, coaching or mediation;
• medical, psychiatric or psychological care of any kind;
• a diagnosis, a screening instrument, or a clinical or psychometric assessment;
• crisis intervention, risk assessment or safety planning;
• legal advice, including anything you might want to rely on in a dispute.

Nobody at Between Us is acting as your clinician, counsellor or adviser. Using the product creates no professional relationship, no duty of care in that sense, and no confidential therapeutic privilege. Nothing on this site has been medically or scientifically validated, and we make no claim that it has been.

We are also not a crisis service. We do not read your answers as they come in, we do not monitor accounts for signs of risk, and nobody is alerted if you write something worrying. If you need someone now, the urgent support page lists organisations that can help.

What the results actually are

Here is the whole mechanism, stated plainly, so you can judge the output for yourself.

Two people answer the same questions separately. Our code compares the two sets of answers using fixed rules: it counts where the answers are close, flags where they diverge by more than a set amount, and — for the prediction questions — measures the distance between what one person guessed and what the other actually said. Then it describes those differences in sentences we wrote in advance, chosen by the pattern it found.

That is it. There is no model of your relationship, no database of couples to compare you against, and no hidden expertise. The results are only as good as the answers, and the answers are self-reported by two people who may be tired, generous, defensive or simply unsure what a question meant.

Which is also why the output can be genuinely useful. The value is not in the judgement — there is none — it is in having both sets of answers visible at the same moment, which is almost impossible to arrange in conversation.

Why there is no compatibility score

Every competitor will give you a number. We will not, and the reason is not modesty.

A single percentage requires deciding how much each area of a relationship is worth relative to the others. How many points is a disagreement about money against a disagreement about how often you see family? Nobody can answer that for your relationship except the two of you, and the weighting a product chooses is an arbitrary decision dressed up as a measurement.

The number also does harm in a specific way. Once it exists, it is the only thing anyone remembers. A couple who reads “68% compatible” stops reading the part that would have helped, and spends the evening arguing about the 32%. A high number quietly tells you to ignore a real problem; a low number tells you something about your relationship is settled when nothing has been settled at all.

So the results give you counts, differences and specific topics instead. How many rating items you answered within one point of each other is a fact. “You are 68% compatible” is a number with no units.

Why we will not label an absent partner

Some of our individual quizzes touch on difficult territory: manipulation, control, jealousy, narcissistic traits. They are answered by one person, about their own experience, with the other person not in the room.

A questionnaire filled in by one party cannot establish what another person is. It cannot tell you whether someone has a personality disorder, whether they are a narcissist, or whether they are deliberately manipulating you. Diagnosis requires a qualified clinician assessing the person directly, and even then it is careful, provisional work. A web form is not a shortcut to it, and any site that offers you one is selling certainty it does not have.

So those quizzes describe behaviour you reported and what it can feel like to live with, name the patterns worth taking seriously, and point you toward someone qualified. They do not hand you a label to carry into your next argument. If what you described sounds like a situation where you are not safe, they say so and send you to the urgent support page rather than scoring you.

What the AI sections can and cannot do

Scoring and comparison involve no AI. The counts, the perception gaps and the discussion topics in your free results are produced by rules in our code, which is why the same answers always give the same output.

AI is used only for longer interpretive writing in paid reports and a few tools. Every such section is labelled where it appears, so you always know which sentences came from a language model.

What it can reasonably do:

• write a readable summary of a pattern the deterministic comparison already found;
• suggest wording for a conversation you are struggling to open;
• offer possible underlying needs behind two different positions, as possibilities;
• restate both accounts of a disagreement without taking a side.

What it cannot do, and should not be read as doing:

• know anything about you beyond the answers you gave;
• assess either person's mental health or character;
• decide who was right about an argument — it is explicitly built to refuse this;
• predict whether a relationship will last;
• notice that you left something important out.

AI text can also simply be wrong. It can over-read a one-point difference, produce something bland, or state a guess with more confidence than the answers support. It is generated at the moment you ask for it and is not reviewed by a person before you see it. Read it as a prompt, and discard any part of it that does not match your experience — you are the authority on your relationship, not the paragraph.

If an AI section ever reads like a diagnosis or a verdict on someone, that is a product failure and we want to know. Write to hello@betweenus.app.

Attachment styles, love languages and other frameworks

Some of our quizzes and articles use popular frameworks — attachment styles, love languages, conflict styles. They are useful because they give people shared words for something they already half-noticed. A couple who can say “I think I withdraw when you want to resolve it immediately” is further along than a couple who cannot.

They are not settled science, and we will not present them as such. Attachment theory comes from a research tradition about infants and caregivers; the popular four-category adult version is a simplification of a more complicated and contested literature, and the questionnaires circulating online are not the instruments researchers use. Love languages are a framework from a popular book, not a validated typology.

Treat them as vocabulary, not as taxonomy. Nobody is only one style, styles change with circumstances and with the person you are with, and a result from a short quiz is a description of how you answered today rather than a fixed property of you. If a framework helps you say something true, use it. If it starts being used to explain away a problem — “that is just my attachment style” — it has stopped being useful.

When to see a professional

A reflection tool is the wrong instrument for some situations. Please consider speaking to a qualified therapist, counsellor or doctor if any of the following is true.

• The same conversation has been going in circles for a long time, and neither of you can change the pattern from inside it.
• There has been a betrayal — infidelity, a serious lie, a broken agreement — that you are not getting past.
• Either of you is dealing with depression, anxiety, trauma, an addiction or an eating disorder. Relationship strain is often downstream of something that needs its own care.
• You are making a decision about separating, children, or living arrangements.
• Conversations escalate in a way that frightens either of you, or you find yourself managing the other person's reactions rather than talking to them.
• You feel unsafe, controlled, or afraid of what your partner will do.

Taking something to a professional is not an escalation and not an admission of failure. It is what the trained option is for. Between Us at its best gets you to that conversation with a clearer idea of what you want to say.

If this is urgent

If you or someone else is in immediate danger, contact your local emergency services now rather than reading further. If you are having thoughts of suicide, or you are afraid of your partner, there are people available to talk to today, and the urgent support page lists them by country.