Referral and Virality Loops for SaaS: Design for Repeatable Pull
Most referral programmes are a coupon with a share button. The ones that compound are built into how the product delivers value, so every successful use shows the product to someone who needs it. Here is how to design one, and how to tell whether it is working.
TL;DR
- A loop is a repeatable path from one user's success to a new user's first session. A one-off campaign is not a loop.
- Inherent loops (collaboration, sharing outputs) beat incentive loops for B2B because the invite is useful to the recipient.
- Viral coefficient (invites per user times conversion) rarely exceeds 1 for SaaS. A coefficient of 0.3 still cuts paid acquisition cost materially.
- Align the reward with the job the product does, and measure the retention of referred users, not just signups.
- Guard against fraud, spam and consent issues before you launch.
A campaign is not a loop
A loop has four parts: a user gets value, a moment arises to share, the recipient arrives with context, and the recipient gets value fast enough to repeat the cycle. If any step is missing, you have a campaign that fades when you stop pushing it.
Dropbox’s early programme is the textbook case. Both the inviter and the invitee received extra storage, a reward that was only valuable to people who used the product. The reward matched the job.
Three kinds of loop
| Type | How it works | Typical in | Strength | Weakness |
|---|---|---|---|---|
| Inherent | Using the product requires or benefits from involving others | Collaboration, scheduling, sharing tools | Invite is genuinely useful | Only fits some products |
| Content | Outputs are shared publicly or sent to third parties | Forms, reports, proposals, public pages | Free distribution, brand exposure | Needs visible branding |
| Incentivised | Rewards for referring | Consumer and prosumer apps | Works for almost any product | Attracts bounty hunters |
For most B2B tools, start with inherent and content loops. If your product is a form builder, every submitted form shows your brand to the respondent. If it is a planning tool, inviting teammates is how the work gets done.
The arithmetic
The viral coefficient is invites sent per user multiplied by the share of invites that convert to active users. If an average user sends 4 invites and 10% become active, the coefficient is 0.4. Add the cycle time, meaning how long it takes a new user to send their own invites. A coefficient of 0.4 with a 7-day cycle compounds much faster than 0.6 with a 60-day cycle.
You do not need a coefficient above 1. A coefficient of 0.4 means roughly 40% more users from every cohort without paid spend, which can lower blended CAC meaningfully.
Designing the loop
- Find the share moment. It is usually right after a win: a report generated, a task completed, a goal reached. Prompt then, not at signup.
- Make the invite useful on its own. A colleague invited to a shared workspace gets immediate value. A discount code for a stranger does not.
- Shorten the path. One click to invite, a pre-filled message, and a landing page that already knows who invited them.
- Reward the job, not the act. Extra seats, usage credits or premium features that help the team do more. Avoid cash unless your margins and product justify it.
- Instrument everything. Log every step so you can see where the loop leaks.
An example
A scheduling tool adds a booking page. Each booking sends the invitee an email footer, “Scheduled with” plus a link. The product team measures:
- Bookings per active user per month.
- Share of invitees who click the footer.
- Share who sign up, and share who create their own booking page within a week.
If the final number is small, the loop leaks at activation, so the fix is onboarding, not another reward.
Where loops go wrong
- Spam. Pushing users to invite their whole address book damages trust and deliverability.
- Fraud. Self-referrals and multi-accounting appear as soon as rewards have cash value. Add limits and review.
- Rewards that outweigh value. If a customer joins only for the credit, expect weak retention.
- Counting signups, not retention. Compare referred cohorts with other channels at 30 and 90 days.
- Consent. Emailing contacts on a user’s behalf has legal limits in many jurisdictions.
How it connects to the rest of growth
Loops work best alongside community and the experimentation discipline in the growth playbook. They amplify a product people already like and do nothing for one they do not, which is why fit comes first.
Instrumentation checklist
You cannot improve a loop you cannot see. Log these events with user and account IDs:
- Invite prompt shown, clicked and sent, with the channel used.
- Invite landing page viewed, signup started and signup completed.
- First value reached by the invited user, and by the inviter’s team.
- Rewards granted, redeemed and reversed.
- Opt-outs and spam reports.
Build one dashboard that shows invites per active user, invite-to-signup rate, signup-to-activation rate and retention for referred versus other users at 30 and 90 days. Review it weekly while you tune the prompt, the message and the reward. Pause the programme if spam reports rise or referred cohorts retain markedly worse.
Your next step
Find the single moment in your product where a user has just succeeded and test a one-click invite there. Measure invites sent and activated invitees for two weeks before building anything more elaborate.
Bottom line
Build referral into the product’s natural moments, make invitations valuable to the recipient, reward the job rather than the act, and judge the loop by the retention of the users it brings. A modest coefficient with a short cycle and real retention is worth more than a viral spike that churns.
Frequently Asked Questions
What viral coefficient should a SaaS product aim for?
Above 1 means each user brings more than one new user, which almost no B2B product sustains. Aim for a coefficient that meaningfully reduces paid acquisition, such as 0.2 to 0.5, and focus on cycle time as much as the number.
Do cash rewards work for B2B referrals?
They can, but they attract low-quality referrals and can feel awkward between professionals. Product credit, extra seats or storage usually align better because the reward is only useful to someone who values the product.
How do I measure a referral loop?
Track invites sent per active user, invite-to-signup conversion, signup-to-activation, and the retention of referred cohorts compared with other channels. The loop is healthy only if referred users stay.
Are there legal limits on referral invitations?
Yes. Email consent rules such as GDPR, CAN-SPAM and the Swiss data protection act apply to invitation emails, and incentive programmes may fall under advertising rules. Get local legal advice before launching.
