Saturating · 2025-11SecondaryReferralProduct

Engineer viral factor and cycle time

Treat virality as a measurable loop: invites per user times conversion, and how fast one cycle completes; improve each step in the loop.

Also called: K-factor, viral coefficient

Claim and evidence

The claimTreat virality as a measurable loop: invites per user times conversion, and how fast one cycle completes; improve each step in the loop.

No dated example yet. Treat this tactic as a hypothesis.

What the numbers show

No published number for this tactic has passed our word-for-word check yet, so there is no chart.

Why it works

Diagram, not data
New user joinsSends invites(invites per user)Invitees convert(rate per step)Cycle time untilthe next invite
New userjoinsSends invites(invites peruser)Inviteesconvert (rateper step)Cycle timeuntil thenext invite

If the factor stays below 1 the loop fades, and a faster cycle compounds sooner. Chen stresses loops must be designed into the product (invites, tagging, referral links) rather than bolted on.

How to run it

  1. Map the loop from first session to next invite
  2. Measure invites per user and conversion per step
  3. Shorten time between join and invite
  4. Remove registration friction in the loop
  5. Treat k as a multiplier on other channels, not a standalone engine

Ways to run it

Design options to test, not promised results.

  1. Option AFix the weakest step

    Measure conversion at every step of the loop and work on the one that loses the most people.

  2. Option BShorten the cycle

    Move the invite moment closer to signup so each round completes sooner.

  3. Option CDesigned-in invites

    Build invites, tagging and referral links into the product rather than bolting them on later.

  4. Option DMultiplier on other channels

    Treat the viral factor as a boost on paid and organic traffic, not a standalone engine.

Dated examples

No dated example yet. Treat this tactic as a hypothesis.

Does it still work?

Saturatingas of 2025-11

Chen (Nov 2025) argues mobile made first-session viral factor above 1 almost impossible; loops now work mainly as multipliers. source

Risks

  • Chasing k>1 on a low-retention product
  • Spam-style invite flows get blocked by platforms
  • Metric ignores non-invite loops

Sources

  1. andrewchen.substack.com · Andrew Chen's Substack · Andrew Chen · 2025-11-05
    “When it’s <1 then your viral loop eventually stops working.”
    ✓ quote checked word-for-word on the page
  2. andrewchen.substack.com · Andrew Chen's Substack · Andrew Chen · 2025-11-05
    “The viral loop I’m talking about is designed into a product, with invite features, tagging, referral links, and so on.”
    ✓ quote checked word-for-word on the page

Change log

  • 2026-11-20 — harvest v0.1 (referral-loops)