Nikita Boyarkin
← Volta Neobank

RAT v2 · project 18 of 23

Does the 45+ trust track pay off: 45+ LTV/CAC = 0.66 against a ≥3 gate; assisted CAC €120 is ~3× referral and doesn't pay back (50-month payback).

Situation
First project in the validation round: the earlier recommendation of a separate trust track for 45+ is put to a money test
Task
Own the money test of the earlier recommendation — a separate trust track for the senior segment
Action
LTV per user, blended CAC by segment and channel, LTV/CAC with bootstrap CIs, payback, Welch t-test on assisted LTV
Result
The anchor clears the ≥3 gate only via referral at LTV/CAC 3.62; 45+ clears none: assisted 0.41; blended 0.66; 50-month payback
Stack
Pythonpandas / NumPySciPy / Statsmodelsscikit-learnMatplotlib / Seabornuv + ruff
Source View on GitHub →
On this page
  1. Situation
  2. Task
  3. Actions
  4. Result
  5. Recommendations
  6. Documentation

Volta — Assisted CAC vs LTV

Situation

The first RAT v2 validation project: the v1 recommendation ‘a separate trust track for 45+’ is tested in money.

Task

I owned the first check in the validation round that put the earlier recommendation — a separate trust track for the senior segment — to a money test.

Actions

  • LTV per user (ARPU × contribution margin × retention months).
  • Blended CAC by segment × channel, LTV/CAC with bootstrap CIs, payback.
  • Welch t-test anchor vs 45+ on assisted LTV.

Result

  • The anchor clears the ≥3 gate only via referral: LTV/CAC 3.62.
  • 45+ clears no channel: assisted 0.41; blended 45+ 0.66.
  • Assisted gives the best retention (churn ×0.75), but the €120 CAC outweighs it — 50-month payback.
  • Welch t-test: anchor − 45+ = €43.99 (p≈2e-13).

Recommendations

  • The v1 ‘trust track’ is right in direction but not yet affordable — hold.
  • Cut assisted CAC (remote video KYC, partner cost-sharing).
  • Combine with a 45+ monetization lever before scaling.

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).

Assisted CAC vs LTV by segment and channel

LTV/CAC across three segments and five acquisition channels. The red gate is 3.0. Assisted onboarding gives the best retention but a costly CAC.

0 1 2 3 4 Young Professionals — Referral: 3.63 3.63 Digital Newcomers 45+ — Referral: 1.72 1.72 Family Budgeters — Referral: 2.12 2.12 Young Professionals — In-app: 2.28 2.28 Digital Newcomers 45+ — In-app: 1.1 1.1 Family Budgeters — In-app: 1.35 1.35 Young Professionals — Partner: 1.65 1.65 Digital Newcomers 45+ — Partner: 0.77 0.77 Family Budgeters — Partner: 0.95 0.95 Young Professionals — Paid social: 1.08 1.08 Digital Newcomers 45+ — Paid social: 0.56 0.56 Family Budgeters — Paid social: 0.62 0.62 Young Professionals — Assisted: 0.77 0.77 Digital Newcomers 45+ — Assisted: 0.41 0.41 Family Budgeters — Assisted: 0.48 0.48 Young Professionals Digital Newcomers 45+ Family Budgeters Segment LTV/CAC Referral In-app Partner Paid social Assisted
Key takeaways
  • The anchor clears the ≥3 gate only via referral: LTV/CAC 3.62; 45+ clears no channel (assisted 0.41).
  • Assisted onboarding gives the best retention (churn ×0.75), but the €120 CAC outweighs the gain for 45+ (50-month payback).

Volta project map

Volta overview →

Core

Extended

Market & Jobs

RAT v2

Causal