Viral Coefficient and Viral Cycle Time

Description

Compute how many new users each user brings, invitation rate times acceptance rate, and how long one invitation cycle takes, to judge viral growth.

The viral coefficient is the invitation rate (invites sent divided by users) multiplied by the acceptance rate (sign-ups divided by invites). A value above 1 means each cohort recruits a larger one. Cycle time, the delay between a user joining and their invitees joining, often matters more: with the same coefficient, a two-day cycle yields about 20,000 users after 20 days while a one-day cycle yields over 20 million. Croll and Yoskovitz distinguish inherent virality (the product needs other users), artificial virality (incentivised invites) and word of mouth, which behave differently and should be measured separately. The viral loop is instrumented as a funnel so that each step (prompt, send, open, accept, activate) can be diagnosed.

Attributes

Loss type
Siloed
Stage
Post-launch
Effort
Hours

Steps

  1. Define the viral loop and classify it as inherent, artificial or word-of-mouth.
  2. Count users and invitations sent in a period and compute the invitation rate.
  3. Count sign-ups from those invitations and compute the acceptance rate.
  4. Multiply the two rates to get the viral coefficient.
  5. Measure cycle time from a user's sign-up to their invitees' sign-ups.
  6. Instrument each step of the loop as a funnel and track coefficient and cycle time by cohort.

Sources

  • Lean Analytics (Croll & Yoskovitz, 2013)
  • Product Management for UX People (Crumlish, 2022)

Also known as

  • K-factor
  • Viral cycle time

Categories

  • Analytics & Behaviour
  • Product Analytics & Strategy