referrals-word-of-mouth

Check five readiness criteria: the majority of growth already comes from word-of-mouth, you have a large enough user base, you can offer a meaningful incentive within CAC limits, your users know other potential users, and your product requires trust to adopt. Referral programs accelerate existing WOM rather than create it.

Calculate K-factor as invites sent per user (i) multiplied by invite conversion rate (c), so k = i × c. A k-factor above 1.0 means exponential viral growth, while below 1.0 means growth decays without other channels. Cycle time between invites also significantly affects growth speed.

There are seven types across three categories: word-of-mouth (offline and online), invitation (social/collaboration, incentivized, utility), and experiential (passive exposure and active content sharing). Classifying your product's natural viral type determines which optimization strategies to prioritize.

Offer meaningful two-sided rewards such as cash or credits, and pay out only after the referred user hits a valuable milestone like a first transaction, not just signup. Watch for the incentive plateau where larger rewards stop improving conversion, and design fraud resistance into the program.

Programs fail when the underlying product lacks organic word-of-mouth, when rewards exclude power users (as in Duolingo's failed program), or when discoverability is poor. Burying the program in settings instead of surfacing it at moments of high user satisfaction severely limits volume.

Referral programs often provide more leverage on the supply side of marketplaces, with benchmarks around 10-15% of new supply coming from referrals at companies like Airbnb and DoorDash. Supply-side participants typically know other qualified suppliers, making them high-quality referrers.