Viral coefficient (K-factor): what it is and how to calculate it

The viral coefficient, also called the K-factor, measures how many new users each current user brings, on average, by inviting other people into the product. It captures, in a single number, the strength of the word of mouth built into usage itself: the higher the K, the more each person who joins multiplies the base without the company paying for it.

The idea comes from epidemiology, where the same concept describes how many people a virus carrier infects. In a SaaS, the "contagion" is the invite: a user shares a link, refers a colleague, or exposes the product inside a workflow, and some of those people become users too. The viral coefficient turns that diffuse behavior into a metric you can track and try to improve.

The K-factor comes from multiplying two simple factors:

• Invites per user (i): how many invites, on average, each user sends in a period.
• Invite conversion rate (c): the share of those invites that turns into a new active user.

The formula is K = i x c. If each user sends 5 invites and 20% of them convert, K is 5 x 0.20 = 1.0, meaning each current user generates, on average, one new user. To raise K you have only two levers: get people to invite more, or make invites convert better. The two multiply each other, so small gains in each combine into a large effect.

The tipping point is K = 1. Above it, each generation of users produces an even larger one, and the base grows exponentially and self-sustains: 100 users become 120, which become 144, and so on, with no extra acquisition spend. Below 1, each generation is smaller than the last and the viral effect decays until it stops.

That does not make a K below 1 useless, far from it. With K = 0.5, every user you acquire brings roughly one more user for free across the whole cascade, because 1 + 0.5 + 0.25 + ... approaches 2. Virality below 1 is an amplifier: it multiplies the output of your other channels instead of replacing them. The common mistake is treating K under 1 as a failure, when in practice it makes all the rest of acquisition cheaper.

K alone does not tell the whole story. What is missing is the viral cycle time, which is how long it takes between a user joining and their invites bringing in new users. Two products with the same K grow in radically different ways if one closes the loop in days and the other in months.

Because the effect compounds, shortening the cycle is often worth more than raising K by a few points. An invite that lands on the first day of use, inside a natural product flow, spins many times in the span where an invite requested only weeks later spins once. That is why truly viral products do not just invite more: they invite early and with low friction, so each turn of the loop happens as soon as possible.

In practice, sustaining K above 1 for long is rare. Contact lists run dry, the network saturates, and novelty fades, so the K of almost every product ends up below 1. That is where virality shows its real value: as a reducer of acquisition cost, not as a replacement for paid channels.

The math is direct. If a paid channel brings users with K = 0.5, each purchased user yields about two users in total, which halves the effective CAC, the acquisition cost David Skok placed at the center of SaaS unit economics. The same logic applies to top-of-funnel metrics like cost per lead (CPL): each paid lead that becomes a user pulls more organic leads behind it. Treating the viral coefficient as part of the acquisition plan, rather than a side project, is what makes paid spend more efficient. Sources such as ForEntrepreneurs show how this effect changes the growth math of a SaaS.

Referral programs are the most direct attempt to engineer a higher K. By giving an incentive to invite (credit, discount, unlocked feature) and by making sharing easy, the company moves both levers at once: more invites per user and, with a good two-sided offer, better conversion of each invite.

Even so, few programs push K above 1 in a lasting way, and the industry track record shows sustained virality is the exception, not the rule. The honest read is to look at the viral coefficient alongside the overall growth rate: a good program usually earns its keep not by making the product self-exponential, but by lowering acquisition cost and speeding up cycle time. Bessemer and other reference firms treat virality as a multiplier of existing channels, a lever that compounds with the others, rather than the single engine of growth.