Security
Accolgo runs on trust and transparency. A new starter’s line manager sees their check-in answers as part of how the product works, and the starter knows that from the first email: it is disclosed, not secret. That visibility is the whole point, it is what lets a line manager spot someone struggling and have a conversation in time.
The one deliberate exception is the private word. If a starter asks to speak with someone senior instead of their line manager, that request is walled off in the database itself: the line manager is never told it was raised and cannot reach it through the app, an export, or a direct query. Here is how both halves are built.
Row-level security on every table, zero anonymous table access, and organisation scoping enforced again in the application layer. One organisation can never see another.
Check-in links are randomly generated, single-use tokens that expire after 14 days. A guessed or shared URL opens nothing.
Line managers see their own team. The Head of People sees their organisation. Platform operators see activation metadata only, never check-in content.
A private word with someone senior reaches only the Head of People. A line manager cannot read it through the app, an export, or a direct query. This is enforced at the database itself.
Encryption in transit everywhere and at rest at the database provider. Secrets live in environment configuration, never in code or the repository.
Migrations are numbered, reviewed and applied with explicit grants, each checked against prior revocations. Error monitoring is tagged per organisation, with an append-only email delivery log.
The detail, and reporting a concern
The full security policy, our data processing terms and our retention schedule set this out in detail and are linked below. If you believe you have found a vulnerability, email dermot@clickonic.co. We welcome responsible disclosure and are glad to hear from you. We acknowledge reports promptly, we do not take action against good-faith researchers, and we simply ask that you steer clear of anyone else’s real data while demonstrating an issue.