After
the Swipe
Testing what dating products leave unfinished after the match.
Dating apps optimize discovery and matching, then largely abandon the transition from a promising profile to a conversation worth continuing.
The problem
The match is treated like an ending. It’s barely a beginning.
After two people match, they still need context, disclosure, pacing, boundaries, and enough confidence to decide whether to meet.
Most products offer a blank chat box and call the transition complete.
Match→Blank chat→Good luck, tiny strangers
What changed my mind
The useful signal wasn’t whether people said they liked it.
People did things I didn’t explicitly ask them to do.
Unique participant count is still unknown. This is behavioral signal, not proof of product-market fit, retention, or conversion.
Three decisions
01
Design the transition, not another matching system.
The product moved beyond compatibility scoring and focused on the work people must do between matching and deciding whether to meet.
01Match↓
02Context↓
03Disclosure↓
04Conversation↓
05Decision↓
06Meet / Don’t
02
Treat compatibility as context, not authority.
A result can start a useful conversation. It should not pretend to finish one, rank two humans, or cosplay as fate with a percentage sign.
Not a verdictA prompt for disclosure.
Communication style · pace · boundaries · conflict · capacity
03
Instrument transitions before ornament.
The important events were dormant-chat recovery, deeper disclosure, contact sharing, meeting proposals, and confident exits.
BEHAVIORAL EVENTS
Result viewedobserved
Answer sharedobserved
Chat restartedobserved
Meeting proposedobserved
Meeting completednext
Confident exitnext
The product model
A transition layer, not a destination.
01Match
02Context
03Disclosure
04Conversation
05Decision
06Meet / Don’t
confidenceboundariescommunication stylepaceillness / capacityconflictbelonging
What became real
Early, but alive.
A lightly distributed prototype generated enough behavioral signal to justify another experiment.
It did not prove product-market fit, retention, or unique-user conversion. That uncertainty stays visible because flattering ambiguity is still ambiguity.
Experiment baselineSmall distribution.
Observable behavior.
Honest unknowns.
Observable behavior.
Honest unknowns.
The next experiment should measure transitions rather than vanity engagement.
Reflection
What I’d test next
Does a dormant chat restart? Does conversation deepen? Does a meeting get proposed?
A score can open a conversation. It should never close the question.
The best outcome may be helping two people confidently leave the product.
The most interesting version of After the Swipe may succeed by becoming unnecessary.