Documentation

Reporting

Four kinds of report, answering four different questions. Three go to your clients; one is for you.

The four, and when each is used

They are separated by nature, not by subject matter, which is why they live on separate tabs rather than as options on one form:

Health reports

Eleven sections, each of which can be switched off per client, covering devices, tickets, patching, security, Microsoft 365, alerts, service performance and asset lifecycle. Helio drafts the executive summary at the top; you can edit it, and an edited summary sticks for future sends if you save it.

Set the frequency (weekly or monthly), the day, and the hour. The hour is read in the client's own timezone, so a report scheduled for 9am arrives at 9am where they are. Monthly sends are capped at the 28th so the date exists in February.

Reports are addressed to the client's contact, plus any extra recipients you list. Wording is deliberately generic — "your IT team", never "Helio" — because the person reading it has never heard of the platform.

The baseline audit

A baseline is recorded automatically 48 hours after a client's first machine enrols, once the rest of the rollout has had time to check in, and separately for each machine a couple of hours after it first reports. Nothing is emailed to anyone automatically.

It is deliberately exhaustive rather than flattering: the complete machine list rather than a top ten, and every problem found, grouped by kind so an estate where nine machines share a fault does not repeat the same paragraph nine times.

Why it exists

The value of a baseline is the comparison, not the document. Helios stores the numbers alongside the report, so months later Progress shows where each figure started and where it is now — from your own records rather than from memory. If no baseline exists, Progress says so rather than comparing today against today and reporting no change.

What it counts as a backup, and what it does not

A Windows shadow copy (VSS) sits on the same disk as the data it copies. It will recover a file someone deleted this morning, but it is lost with the drive, it goes with a stolen laptop, and it is the first thing ransomware deletes. Where a shadow copy is the only signal on a machine, the baseline counts that machine as having no backup and says why, in those words, to the client.

A machine that has never reported is recorded as unknown, never as clean, and unknowns are excluded from both sides of the count rather than folded into the good column. The number of them is stated.

Retaking one

You can retake a baseline, but it replaces the starting position every later comparison is measured against, so it is an admin action, it asks for confirmation, and it never happens automatically.

Internal reports

These run across every client at once and name them alongside each other, so they are admin-only, scoped to your own tenant, and marked as internal on every page. Sample and demo clients are excluded from all of them.

Estate reports the agent versions running but does not label any of them out of date. The platform has no record of which release is current, so calling a version old would be a guess rather than a fact.

Sample data never counts

Demo clients are visible everywhere in the platform but are excluded from every figure presented as fact, including the reporting summary counts and all internal reports. A schedule cannot be created for a sample client at all.