Run a referral program from design to first real results
Adding a "refer a friend" button and building a referral program that actually generates referrals are different projects.
- Outcome
- A referral program designed around why your specific customers would actually refer, an incentive structure that motivates without attracting the wrong kind of referral, and an honest read on early results before deciding whether to scale it.
- Time
- A few hours to design, then several weeks to gather real early results
This workflow is part of PromptKit Pro
A handful of prompts and workflows stay free for everyone — Pro unlocks the full library, including this one, plus the guided runner.
Referral programs often get built as a feature — a button, a discount code, a share link — without first understanding why a specific customer would actually want to refer someone, which produces a program that's technically functional but generates little real activity, or activity that's low quality (people referring themselves or strangers for the discount alone). This playbook designs around the real motivation to refer, builds an incentive structure that filters for genuine referrals, and checks early results honestly before scaling.
When not to use this
For a product where customers genuinely have no natural reason to refer others (a highly personal or embarrassing use case, say), a referral program may not be the right growth lever at all regardless of design.
Did this work?
You have an honest read on whether the program is generating real, quality referrals after the first real test, not just whether the feature exists and technically works.
Tested on claude-opus-5. Evidence status is draft; it moves to battle-tested only on recorded runs, never by hand.