Security and hosting

Your estate’s data, on its own.

epostle is deployed as a separate instance per client. There is no shared database, so there is no query, no configuration mistake and no software defect that could show one operator another’s trading.

Your own instance

A dedicated application and database per client. No shared tables, and no tenant identifier standing between your data and another operator’s.

Access under control

Named users with roles assembled from individual permissions — reporting, products, suppliers, imports, administration — so people see exactly what their job needs and nothing more.

Credentials kept out of configuration

EPOS keys and database passwords are supplied to the running service by the environment, never stored in configuration files or built into images.

Nothing thrown away

The original EPOS payload for every transaction is retained, so any figure can be traced back to what the till actually sent.

Configured, not customised

Trading-day cutover, till key meanings, discount structures, VAT and ledger mapping are settings. Your estate’s quirks do not become bespoke code.

Hosted and maintained

Deployment, monitoring, upgrades and support are handled for you, with the option to host in your own environment where policy requires it.

Questions we are asked

Where does the data live?
In your instance’s own database, hosted in the UK by default. Where policy requires it, epostle can run in infrastructure you control.
Does epostle write back to our tills?
No. Collection is read-only. The only data epostle writes outward is the ledger delivery into a database you nominate.
What happens if a night fails?
Nothing is silently skipped. Failed collections are logged, retried and re-collected, and a site reporting takings with no imported sales raises an alert rather than appearing as a quiet zero.