Related Ideas
eyko Ideas
Referral programs aimed at the full customer base produce thin response. A Referral Likelihood Prediction Playbook reads NPS, advocacy behavior, network position, and segment fit to surface the customers most likely to refer and the moment in their lifecycle when the ask should land.
The Challenge
A referral prompt sent to every active customer produces a fractional response rate. The customers who would have referred warmly get the same campaign as the ones who would not, the message has to be generic, and the conversion stays well below where targeted asks could put it.
A customer at month two is rarely ready to refer; the same customer at month nine after a successful renewal is highly ready. Without a lifecycle-aware ask, the program either fires too early or sits as a static link in an email footer.
A small subset of customers sit in network positions where a single referral leads to multiple introductions. The standard NPS list does not capture network reach, and the program never identifies the high-reach candidates worth a personal ask from the account team.
How eyko Solves It
A Referral Likelihood Prediction Playbook reads NPS history, content engagement, advocacy activity (public reviews, references), product success milestones, and network signals (LinkedIn reach, ICP-fit network density) to score each customer on referral likelihood. It segments candidates by readiness type, surfaces high-reach referrers worth a personalized ask, and recommends the lifecycle moment when each candidate is most likely to respond.
The Playbook scored 4,200 active customers on referral likelihood and identified 480 high-likelihood candidates, 142 of whom have substantial ICP-fit network reach. 240 of the 480 have not yet been asked. The current referral program currently reaches 4,200 with a generic prompt; replacing it with targeted asks projects a 4.6x lift in referral close rate.
| Metric | Current | Benchmark | Status |
|---|---|---|---|
| Primary indicator | Flagged | Target | Action needed |
| Secondary indicator | Monitoring | Within range | On track |
| Trend direction | Declining | Stable | Review required |
Referral Likelihood Prediction ranks every customer on probability of referring based on NPS, advocacy activity, recent success, and network position. The Playbook surfaces the high-likelihood candidates, separates the high-reach referrers worth personalized asks from the broader targeted cohort, and recommends the lifecycle moment when each candidate is most likely to respond so referral programs run on evidence rather than a generic prompt to the whole customer base.
Related Ideas
FAQ
Everything you need to know about Referral Candidate Ranking.
Referral Likelihood Prediction is an AI-driven ranking of every customer on probability of referring new business. The Playbook reads NPS history, advocacy activity, product success milestones, and network signals to score each customer, surfaces the high-likelihood candidates, separates the high-reach referrers worth personalized asks from the broader cohort, and recommends the lifecycle moment when each candidate is most likely to respond.
The Playbook reads from your survey platform (NPS and CSAT history), CRM (account context, prior reference activity, lifecycle stage), product analytics (success milestones and adoption depth), marketing automation (referral program engagement, content interactions), and network signals where available (LinkedIn reach, public review activity). At least 12 months of historical NPS and referral outcomes anchors the model in real referral behavior.
A high-NPS list captures customers who said they would recommend. Referral Likelihood Prediction captures who is actually likely to refer when asked and which ones have the network reach to deliver multiple introductions. The Playbook also recommends the lifecycle moment for each ask so the timing matches the customer's readiness rather than firing as a static link in an email footer.
Yes. For high-reach candidates the Playbook recommends a personalized account-manager outreach with the recent success milestone attached as context. For the broader high-likelihood cohort the Playbook recommends a targeted automated ask matched to the customer's milestone. Each recommendation projects expected response and close rate so referral leadership can prioritize the highest-leverage touches first.
Join the enterprises replacing weeks of manual analysis with a single prompt. See what eyko Playbooks can do with your data.
