This essay is part of the Solving Distribution chain framework. If you haven’t read the main pillar article, start there to see how Referral turns a linear funnel into a loop.
The Distribution Chain: Link 7 of 7
You have reached the end of the chain.
The bucket holds water (Link 1), the positioning is sharp (Link 2), sign-ups reach the aha moment (Link 3), activated users pay (Link 4), you have validated a channel (Link 5) and you are scaling it with healthy economics (Link 6).
The engine works. Every euro and every hour you put into the top of the funnel produces revenue at the bottom.
Now comes the last upgrade: getting your existing users to bring you the next ones.
This is Link 7 of the Distribution Chain: Referral.
Referral is the multiplier. It takes a linear system, where growth is capped by how much money and time you spend on acquisition, and bends it into a loop. Every user acquired through paid or organic channels brings a fraction of another user with them, which pulls down your blended acquisition cost and accelerates everything.
It is also the most common victim of premature optimisation. Bolting a referral programme onto a product people do not love yet is just spam with a coupon attached. You cannot pay people to recommend something they feel lukewarm about, and zero times anything is still zero.
What Referral Actually Answers
Referral measures whether your product generates enough genuine delight, collaboration or utility that word of mouth sustains itself. It answers two questions:
- The Organic Love Question: do happy users recommend this to peers without being bribed?
- The Viral Coefficient Question: does using the product naturally expose it to non-users as part of how it works?
Real compounding growth comes from loops woven into the core mechanics of the product. An affiliate link buried in the account settings page is not one.
Model Your Viral Loop
The viral coefficient is simple arithmetic: K = invites sent per user × the share of invites that convert. What surprises most founders is how much a sub-viral loop is still worth.
- every paid customer really brings
- –
- 1,000 users become, in 90 days
- –
Notice what the numbers say. You do not need K above 1 to win. At K of 0.4, every customer you pay for effectively becomes 1.67 customers, which is the same as cutting your acquisition cost by 40%. Almost every successful software company lives below 1.0 and treats the loop as an amplifier on paid and organic acquisition.
Then play with loop time. Shortening the days per loop does more for compounding than another few points of invite conversion, because speed multiplies the number of cycles you get in a quarter.
How to Diagnose a Broken Link 7
A referral failure shows up as growth that stays strictly dependent on spend. Turn off the paid channels and growth stops dead.
Quantitative Signals
- A viral coefficient well under 1: on average a new user invites fewer than one additional user who joins, and often far fewer.
- Negligible organic share: word of mouth, direct traffic and referred sign-ups account for less than 10% to 15% of new acquisition.
- Dead referral programme: under 1% to 2% of active users ever click or share their referral link, and the links that do get shared convert poorly.
Qualitative Signals
- Embarrassed sharers: you offer a real cash payout and users still hesitate, because sharing it would reflect on them.
- Artificial virality: your mechanism reads as a gimmick, like tweeting about the product to switch on a basic feature.
- Word-of-mouth disconnect: users like the tool privately but never mention it to colleagues, because nothing in the product is collaborative or visible.
Why Referral Breaks (Root Causes)
1. The Zero Times Anything Problem
Founders reach for a referral programme to fix an upstream retention or activation problem. If people are not getting repeatable value from the tool, no affiliate link will make them champion it. Love comes first, and recommendation follows it.
2. Misaligned Incentives
A €10 gift card in a B2B setting creates awkwardness, because it makes the referrer look like an affiliate marketer to their own colleagues. B2B referrals work when the incentive is in-product utility, like extra capacity or credits, or social capital, where recommending the tool makes the referrer look useful and well-informed.
3. Friction on the Receiving End
A user shares a link, their colleague clicks it, and lands on a generic homepage with a standard signup form that knows nothing about who invited them or why. The loop breaks at the last step, which is the most expensive place to break it.
The Fix: 3 Tactical Playbooks for Link 7
Tactic 1: Pick the Right Viral Engine
Match the mechanism to your product rather than building a generic programme:
| Engine | How it works | Fits |
|---|---|---|
| Inherent, workflow-driven | The product has to be shared to be useful at all | Figma, Loom, Calendly, DocuSign |
| Two-sided value loop | Referrer and invitee both get something immediately | Dropbox storage, PayPal credit |
| Social proof and badges | The output carries a visible mark back to the source | “Powered by Typeform”, “Sent via Superhuman” |
Two notes on making these work. For inherent virality, design features where collaborating requires inviting someone, like sharing a read-only view. For two-sided loops, always reward both sides: giving the invitee a perk turns the referrer from a salesperson into someone doing a favour.
Tactic 2: Treat Invited Users as Guests, Not Strangers
When someone arrives through a referral link, the page should know it.
- Show who invited them, by name and avatar: “Daniel invited you to collaborate.”
- Say what they are joining, naming the specific project, document or workspace.
- Offer one-click signup through Google, GitHub or Apple, so they land in the shared thing immediately.
The difference is stark. A generic “Welcome, create an account” drops most of the traffic your existing users worked to send you. “Alex invited you to view the Q3 growth dashboard” converts, because it arrives with context and a reason.
Tactic 3: Ask at the Moment of Delight
Timing decides whether the ask feels natural or grabby. Never trigger a referral prompt while someone is entering billing details or fixing a setting.
Ask right after a win: publishing their first project, sending their first invoice, hitting a milestone. Or ask right after they tell you they are happy, when someone rates a support interaction or an in-app survey 9 or 10: “Glad that worked. Want to give a colleague a free month?”
Real-World Example: How Loom and Figma Built Loops
Loom did not grow on ads or referral bounties. The loop is the product. A user records a video and sends the link to a teammate or client. The recipient watches it and experiences the entire core value, that this replaced a 30-minute meeting, without ever creating an account. At the end of the video sits one line: record your own Loom for free. The recipient becomes a user, records a video for someone else, and the loop turns again across organisations.
Figma did the same thing to design software by making files URL-based. Designers could not work with developers or managers without sharing a link, which turned every developer, every client and every stakeholder into a potential account.
Neither company bolted a programme on. In both, sharing is how the product works.
The Takeaway: Compounding Is the Reward for a Solved Chain
Referral is the last link, and it depends completely on every link above it holding.
If your messaging is vague (Link 2), referred visitors will not understand what they were invited to. If activation is clunky (Link 3), invited colleagues abandon onboarding. If retention is poor (Link 1), referred users churn as fast as they arrived.
When every link works, distribution stops being an uphill hunt for new traffic and becomes a closed loop: usage creates awareness, awareness creates sign-ups, and sign-ups create more usage.
Fix Link 7 systematically:
- Match the viral engine to how your product actually gets used.
- Make incentives two-sided and useful rather than cash-shaped.
- Ask for the referral only at moments of genuine delight.
That is the whole chain. Eight questions, asked in order, each one answerable.
One link at a time.
What to Read Next
- The full Distribution Chain framework: the master blueprint, and where to start over.
- Link 1: Retention: because a loop only multiplies a product people already love.
- Link 6: Reach: how to scale the channel that the loop amplifies.