A lot of dating frustration is not a chemistry problem. It is a timing problem in the data.
Two people talk for weeks, get along, and then discover they were in different products the whole time: one looking for a committed relationship, the other for casual company. Neither lied. The intention was simply never stored anywhere, so it could not be checked until checking it felt like a loss.
The original feature is at the canonical link above. This post covers three design choices that came out of treating intention as data, and what each one taught.
1. Make intention a structured field, not a bio line
Free text feels kinder, but it fails as a filter. "Here for something real", "seeing what happens", and "open to anything" all sound warm and mean almost nothing that two readers will interpret the same way.
A structured field — a small, plain set of options such as a serious relationship, marriage in time, or a slower start with a real direction — is less poetic and far more useful. It can be displayed consistently, compared at a glance, and used to sort discovery without guessing.
What worked: a short list with honest labels beat a long list with clever ones. People do not need twenty shades of intention; they need the few distinctions that actually change behaviour.
What broke: a list that is too short forces people into the nearest box rather than the right one, and then the data looks clean while the conversations still mismatch. The fix was wording, not more options: labels written the way people describe their own aim, tested against how readers understood them.
2. Show the intention before the first message
An intention that only appears deep in a profile, or only comes up in conversation, arrives too late. By then both sides have already spent the attention the mismatch was going to waste.
Putting the intention on the profile and in discovery — visible before anyone writes a word — changes the job of the first conversation. Instead of detective work ("what is this person actually here for?"), the first exchange can test the only thing software cannot: whether these two particular people suit each other.
What worked: early visibility filtered kindly. People seeking something else moved on without a difficult conversation, and people seeking the same thing relaxed sooner because the opening was honest.
What broke: visibility alone does not resolve timing inside a match. How often to message, when to meet, whether to keep seeing other people — none of those has a universal answer, and showing an intention does not answer them. It just makes those conversations cheaper to have in week one than in month three.
3. Keep verification and intention as separate signals
It is tempting to merge trust signals into one badge. That is a mistake here, because verification and intention answer different questions.
Verification says: this is a real person, who they claim to be. Intention says: this is what that person is here for. A verified profile with an unstated intention still leaves the most common early failure untouched, and a stated intention from an unverified profile still leaves the other one.
What worked: treating them as two independent, visible fields. Together they remove the two most common reasons early dating goes wrong, and neither one has to overclaim what it proves.
What broke: both signals can look more definitive than they are. A stated intention is a statement at a point in time, not a guarantee, and the interface has to present it that way — as an honest opening, not a contract.
The through-line
Clarity early is not unromantic, and it is not just copywriting. It is a data-modelling decision: which facts about a person are structured, visible, and shown before effort is spent, and which are left to be discovered expensively later.
The live example is 6on7, built for dating aimed at actual relationships, with verified profiles and intentions visible up front.
Questions for other builders: in your product, what is the equivalent of intention — the fact everyone discovers too late that could have been a field on day one?
Top comments (0)