YesGraph

YesGraph identified a critical gap in how apps managed user invitations. ​​‌‌‌‌‌‌‌​‌‌​​‌​​​​​​‌‌​‌‌‌​​​‌‌Most applications presented users with their entire contact list, forcing them to manually scroll through hundreds of names to find who might actually want the service. This created friction that killed referral programs before they could gain momentum. Mobile app developers and consumer platforms experienced this most acutely—their growth depended on viral adoption, yet their invite flows were fundamentally broken. The problem was measurable: companies tracked abysmal invitation acceptance rates and could see users abandoning the invite screen within seconds. Existing alternatives were crude: some apps sorted contacts alphabetically or by recency, while others simply randomized suggestions. YesGraph's early validation came from observing that LinkedIn's "People You May Know" feature drove significant engagement despite being a secondary product feature. When YesGraph applied similar machine learning to invitation contexts, they saw invitation acceptance rates jump dramatically. Beta customers reported 3-5x improvements in referral conversion, proving that intelligent ranking of contacts—not just presenting them—fundamentally changed user behavior and app growth trajectories.

YesGraph built their invitation recommendation engine for mobile app developers and growth teams seeking to optimize referral mechanics. Their initial assumption was that app makers would pay for algorithmic recommendations on which contacts users should invite, similar to LinkedIn's connection suggestions. The founding team targeted product managers and growth leads at consumer apps, believing these users faced a genuine friction point in their invite flows. Early validation came through direct outreach to app developers who immediately recognized the problem—many referral programs suffered from low conversion because users didn't know whom to invite. YesGraph's machine learning approach to analyzing social graphs resonated with technically sophisticated teams. However, the available source material doesn't detail whether they encountered significant audience pivots or discovered unexpected customer segments during their go-to-market efforts. The core targeting assumption—that growth teams would value algorithmic invite recommendations—appeared sound based on initial traction, but specific data about customer acquisition channels or whether they shifted their buyer personas remains limited in the provided information.