investors
recommendation is a solved problem for things you buy, and an unsolved one for how you spend your evening.
youconic is a personal curated recommendation engine: a small number of considered ideas for your morning, your day, your night and your next trip, with the reasoning attached. invite‑only, global from day one, and built so member data is never sold.
the problem
every discovery surface built in the last decade optimises for engagement, inventory or ad load. none of them optimises for whether you would actually have enjoyed the evening.
a five-star average is a measure of how few people were disappointed, not of whether a place is for you. the top result is the least objectionable one.
placement is inventory. the room you are shown is frequently the room that paid, and there is no way for you to tell which is which.
a feed that is good at holding you is not the same thing as a feed that is good at sending you somewhere. the incentives point in opposite directions.
the product
what a member actually receives, every day, in the order it is built.
a short set of questions establishes a starting position: how you like a room, how far you will travel, what you are here for. enough to be useful on day one, not so much that it feels like an application form.
members can attach the accounts that already describe their taste. the signal is read to sharpen the brief, and it is theirs to disconnect. nothing is sold, and nothing is brokered.
places are researched and written up rather than scraped from a listings feed, with opening hours checked and the reason a room is worth the trip recorded alongside it. coverage is built city by city.
a handful of picks with the reasoning shown, plus what is worth reading, watching and following underneath. one of them is a deliberate stretch beyond your usual taste, marked as such.
members who consistently respond to the same rooms sit near each other. that structure is what lets a new member get a good brief on their second day rather than their second month, and it is the part that compounds.
the model
the ranking is not for sale. that constraint is the product, so the revenue has to come from somewhere that does not touch it.
members pay for the brief. their interests and the engine’s interests point the same direction, which is the only arrangement under which a recommendation can be trusted.
a business can claim its listing, keep its hours and rooms accurate, and reach the members already being sent to it. claiming buys accuracy and reach. it does not buy rank.
member data is not sold, brokered or used to train anything outside the product. it is the promise the whole thing rests on, so it is not available at any price.
why now
a list of places has been possible for twenty years. what changed is that a machine can now read a person’s taste, hold a city’s worth of context, and write down why a particular room suits a particular evening in language a person will act on. the recommendation was never the hard part. the argument for it was.
who’s building this
a team of engineers, taste makers, sneaker heads, travellers, editors, philosophers, foodies and designers from around the world.
get in touch
tell us who you are and we’ll come back with the current materials and a time.