Solve a hyperlocal balanced liquidity problem by igniting one dense social graph by hand rather than broadcasting thinly everywhere, then let each seeded network organically pull the next one into existence.
Tinder
Tinder is a location-based mobile dating app, built inside IAC's Hatch Labs and launched in September 2012 by a team including Sean Rad, Justin Mateen, Jonathan Badeen, and Whitney Wolfe, where a mutual swipe-right ("match") opens a chat. It is worthless to a lone user: value exists only when enough nearby, mutually interested people are present to swipe on. Its two stories trace one arc, from lighting a single dense campus by hand to letting each seeded network pull in the next.
Cold Start: Igniting one sorority at a time
The problem. Dating-app liquidity is hyper-local and balanced: you need matches near you, now, and both sides present in roughly the right ratio. A national launch would have scattered a thin user base across thousands of cities, leaving every individual with nobody nearby to swipe on and guaranteeing churn. Tinder faced the hardest possible version of the empty-room problem, where the first user to open the app sees no one.
The approach. Rather than broadcast to the whole internet, Tinder narrowed to a single dense, balanced social graph: US college campuses, and specifically the Greek system. Whitney Wolfe, a former Kappa Kappa Gamma member who understood Greek social dynamics, physically toured campuses starting at USC, presenting inside sorority houses and having the women install the app on the spot, then crossing to the corresponding fraternity, where brothers immediately saw profiles of women they recognized from campus. Tinder also threw exclusive campus parties where the price of admission was downloading the app.
How it solved it. Seeding the women's side first and pre-loading each network with recognizable faces made the very first session valuable, defeating the empty-room problem that kills most dating apps. Wolfe's grassroots college tour reportedly grew Tinder from roughly 5,000 to 15,000 users, and each campus was a self-contained launch that hit critical density before the next one began.
Network Effect: From campus density to a global two-sided loop
The problem. Hand-seeding one campus at a time does not scale to a global product on its own; Tinder needed the density it manufactured in each location to compound and spread outward rather than stay trapped in isolated pockets.
The approach. Tinder let the seeded liquidity run a classic two-sided loop: within any locale, more users on one side raised the match rate, which pulled in more of the other side. College networks are not isolated (students travel, transfer, and graduate into cities), so densely seeded campuses bled into adjacent geographies, and the swipe-and-match mechanic made every match a shareable, exciting, inherently word-of-mouth event.
How it solved it. The loop turned hand-seeded campuses into explosive growth: Tinder was processing about 350 million swipes per day by late 2013, and by October 2014 it had passed over one billion swipes per day producing roughly 12 million matches per day, reaching more than 50 million users within about two years. It became the highest-grossing dating app in the world and the anchor of Match Group's portfolio.