What is a viral loop?
A viral loop is a cycle where using the product causes existing users to bring new ones, who then repeat the cycle. It's measured by the viral coefficient (K) — new users generated per existing user per cycle. K above 1 compounds; below 1, it usefully lowers acquisition cost but doesn't grow on its own.
The three canonical shapes
Incentivised referral. Dropbox gave storage to both the referrer and the invitee. The incentive was the product itself, which meant it cost Dropbox marginal storage rather than cash, and it selected for people who actually wanted more Dropbox.
Embedded distribution. Hotmail appended a signup link to every outgoing message. Nobody chose to refer anyone — using the product broadcast it, which is the cheapest loop that exists.
Paid referral. PayPal's $20 for signup and $20 for referring was expensive and blunt, and it bought a payments network in a market where the network was the entire product. It worked because the lifetime value of an early network node justified it; the same tactic bankrupts a business with weaker economics.
The maths people skip
K = invites sent per user × conversion rate of those invites. Both halves are addressable, and the second is usually where the leverage is — teams obsess over prompting more invites while the invite landing page converts at 8%.
Cycle time then determines whether K matters. Compounding is a function of how many cycles fit in a period, so a loop that completes in days is worth far more than the same K completing in months.
Where it goes wrong
A viral loop on a product that doesn't retain accelerates failure. You spend acquisition budget to introduce more people to something they'll abandon, and you burn the goodwill of everyone who was referred.
The sequencing that works: retention first, then activation, then the loop. Dropbox's referral programme is famous because the product underneath it was genuinely sticky — the loop distributed something worth having.
Seen in practice
Case studies where this shows up as a real decision, not a definition.
Related questions
What is a good viral coefficient?
Anything above 1 means self-sustaining growth, which is rare and usually temporary. Most durable products run well below 1 and treat virality as a discount on paid acquisition rather than a growth engine on its own.
Why does cycle time matter?
Because compounding depends on frequency as much as rate. A K of 1.2 with a two-day cycle grows explosively; the same K with a two-month cycle is nearly flat over the same period. Shortening the loop is often easier than raising K.
What's the difference between a viral loop and word of mouth?
A viral loop is built into product usage — sharing is how the product works. Word of mouth is people choosing to recommend you separately. Both are valuable; only the loop can be engineered and measured directly.
More on growth
Last reviewed 2026-09-07