Churn rises with ticket count: 37.1% at zero contacts vs 81.1% at 3+. Support is a measurable retention lever, not just a cost center.
Volta — Support & Churn
Situation
We test whether a bad support experience drives churn and whether it can be influenced.
Task
My job was to check whether a poor support experience actually drives churn, and whether that link is something the team can act on.
Actions
- Ticket–churn merge per user.
- Churn by ticket count, by unresolved, by CSAT band.
Result
- Churn rises with ticket count: 37.1% (0) → 46.3% (1) → 65.2% (2) → 81.1% (3+).
- Users with 3+ tickets churn 2.2× more than the base.
- CSAT bands barely discriminate churn (52–55%) — a weak signal.
Recommendations
- Cut ticket volume via self-serve and clearer errors.
- Resolve unresolved tickets faster — they amplify churn.
- Don’t rely on CSAT as a retention predictor.
Documentation
Charts
Source: github.com/NikitaBoyarkin/volta-banking — 22 projects; figures recomputed from the repo's own datasets (data/*.csv) via its analysis scripts. Funnel counts from data/volta_funnel_data.csv (10,000 users); A/B, retention, segmentation, churn, RFM, CLV, attribution, anomalies, spend, support, NPS, JTBD, unit economics, premium, KYC deep-dive, referral, assisted CAC, FX sourcing, premium offers, anchor CAC and dormant win-back follow the published project narrative (README + part pages).
Churn by support-ticket count
Share of churned users by number of support contacts. More tickets — higher churn, linking support quality to retention.
Key takeaways - Churn rises with ticket count: 37.1% at zero contacts vs 81.1% at 3+.
- Users with 3+ tickets (185 people) churn 2.2× more than the base — support is a retention lever.