Keep the health model deterministic so every score can be explained.
PROJECT / 01 · Customer success / self-hosted
OpenRetention
OpenRetention is a source-available customer-success system for teams that want a focused, inspectable risk workflow instead of a large enterprise suite.
RISK / RENEWALexplainable signal stack
Early B2B SaaS teams can have product, billing and renewal data without having a usable daily queue for customer risk. The useful question is not “what is the score?” but “who needs attention first, why, and how much revenue is exposed?”
HOW I BUILT AROUND IT01 02 03 04
Separate usage, inactivity, payment and renewal signals instead of hiding them inside one opaque model.
Order the risk inbox by commercial exposure so the operator sees what matters first.
Keep deployment self-hosted and the core intentionally small enough to inspect.
CAPABILITIESWhat it does,
What it does,
without the theatre.
01Deterministic health scoring02Usage decline and inactivity signals03Payment and renewal signals04Revenue-at-risk prioritisation05Risk inbox and account inspector06Node.js 24 + SQLite07Docker configuration08Public seeded product preview
ARCHITECTURE
What the system is meant to preserve.
- 01Explainability over opaque scoring
- 02Workflow density over dashboard theatre
- 03Revenue context beside every risk signal
- 04Accurate source-available licensing language
Is OpenRetention open source?+
Not before its Change Date. It is source-available under Business Source License 1.1 and the repository states the future Change License.
What does the health model use?+
The deterministic model uses observable signals including product-usage decline, inactivity, payment state and renewal proximity.
Is it a full enterprise customer-success suite?+
No full feature-parity claim is made. The current product is deliberately narrower and focused on self-hosted, inspectable risk and renewal workflows.