K-factor: what a viral coefficient below 1 is actually worth
Almost nothing sustains a k above 1. Here is what a k of 0.4 is genuinely worth to your blended CAC, and why chasing true virality burns product cycles.
the short answer: k above 1 means each user brings more than one more user and the thing grows without you. almost nothing sustains that. a k of 0.4 is not a failed viral loop, it is a permanent discount on every user you pay for, and it is the outcome worth actually building toward.
The arithmetic, briefly
k is invites sent per user multiplied by the share of invites that convert. ten invites at a 4% conversion is a k of 0.4.
above 1, growth compounds on its own. below 1, each cohort produces a smaller next cohort and the loop fades out. that sounds like the difference between success and failure, which is why founders fixate on crossing 1, and why so many product cycles disappear into trying.
What below 1 is genuinely worth
it is a multiplier on everything you buy.
at a k of 0.4, every hundred users you pay for bring forty you did not, who bring sixteen, who bring six. the loop converges to roughly a 1.67x multiple on paid acquisition, which means your effective CAC is about 60% of what you actually paid per install.
that is a very large, very permanent discount. it does not make headlines because it is not a growth story, it is a cost story, and cost stories do not get written up.
Why chasing k above 1 usually costs more than it returns
sustained virality above 1 is mostly a property of the product category, not of how hard the growth team tried. products where using the thing requires other people present have it structurally. products people use alone do not, and no amount of invite flow polish converts the second kind into the first.
the expensive failure mode is spending quarters of product time on referral mechanics for a single player product, while the acquisition channel that was already working goes untended. i have watched that trade get made and it is almost never worth it.
worse, the aggressive version of the chase degrades the product: forced invite walls, contact list harvesting, notification pressure. these can move k briefly and cost you retention permanently, which is a bad trade at any coefficient.
What moved the number on a real app
scaling ZuAI from 10,000 to 2 million users at a blended $0.02 CAC, the cheap traffic did not come from a viral coefficient above 1. it came from Reddit seeding and raw UGC creative volume driving acquisition costs down far enough that paid could scale behind proven winners.
the honest framing is that we made acquisition absurdly cheap rather than free. a loop below 1 helped by making the cheap thing cheaper. it was never the engine, and treating it as the engine would have taken product attention away from the channels that were.
What I would do instead
measure k honestly, per cohort, and accept the number. then treat it as what it is: a discount rate on paid, which makes paid scale further, which is worth real effort at the margin.
build the share flow because sharing genuinely helps the user, not because a spreadsheet needs a number above 1. a loop that exists because the product wanted it tends to hold. a loop bolted on to hit a target tends to get removed a year later when someone checks what it did to retention.
the k-factor calculator runs the arithmetic on your real invite and conversion numbers, including what your effective CAC becomes once the loop is counted.
want to know whether your loop is worth building before you spend a quarter on it? the breakdown call is free. Let’s talk Growth.