Guide · 33

Subscription metrics

Computes MRR, new customers, departures, churn, ARPA and LTV month by month from your list of subscriptions.

In one line

node scripts/33-subscription-metrics.mjs abonnements.csv --from 2026-06

What it's for

If you sell a subscription (software, maintenance, coaching), five numbers say whether the business is healthy: monthly recurring revenue (MRR), new customers, departures, churn rate and customer lifetime value (LTV). This script computes them month by month from your list of subscriptions.

Before you start

  • A CSV: customer, mrr (monthly amount), start (start date), end (end date, empty if still active).

Step by step

  1. Run over a period:
    node scripts/33-subscription-metrics.mjs samples/abonnements.csv --from 2026-06 --to 2026-10
    
  2. Read the MRR column: its monthly growth is the first thing to track.
  3. Watch churn: above a few percent a month, growth comes at the price of a leaky bucket.

Reading the output

A subscription counts for a month if it started before the end of that month and is not finished by then. Month churn = departures ÷ customers at the start of the month. ARPA = MRR ÷ active customers. LTV = ARPA ÷ monthly churn; it stays empty until a departure has been observed (an infinite value teaches nothing).

Automating it

Export your subscriber list at the start of each month and re-run; keep the outputs to trace the trend.

Pitfalls

  • With few customers, a single departure makes churn jump: read trends, not an isolated month.
  • The script assumes a constant price per subscription; an upgrade is recorded as a new line.

This guide goes with a free script (MIT license). Try it on a copy of your data first.