Session Usage shows where your session time went: by person or API key, by day, by installation key, or by system. Use it to find what is using your hours, and to check your bill.

Make sure you have the following:

HyperView bills session time, so the question behind most usage work is the same one: where is the time going? Session Usage answers it by grouping every session in a window and showing what each group consumed.

The Period filter sets the window: recent windows such as the last 7 days, or the billing period, which is the one to use when you are reconciling against an invoice. Kind and System Type narrow it further, and the search box filters within the result.

The footer states the totals for whatever you have selected (sessions, time consumed, and the dates covered), so you can always see what the numbers on screen actually cover.

Whichever grouping you choose, each row carries Sessions, Consumed, Share of the total, Avg per session and Last active. Switch grouping from the menu at the top left.

The one to start with. People and API keys appear in the same list, and the Kind column labels each row HUMAN or AUTOMATION, so the split between hands-on work and automation is readable at a glance:

Here one person accounts for 86% of the time across three sessions, and an API key for the remaining 14% in one. That is a fleet where automation is still an experiment. The same view on a mature setup usually inverts.

Whether consumption is steady or spiky, and whether a change you made stuck:

Which fleet, environment or customer the time belongs to, because the key a system installed with is the cleanest tag it carries:

A single row at 100%, as here, means every system in the window came from one installation key. That's fine for a small account, and the thing to change if you need usage attributed per team or per customer. See Installation keys.

Which individual machines dominate. Usually a short list, and the fastest way to find a runaway job.

The breakdowns are more useful in sequence than individually. Start by actor. If automation is consuming the time, go by system to find where it is concentrated, then by day to see whether it is steady or a one-off. Use by installation key when the answer needs to be attributed to a team, an environment or a customer.

Share is the column to read first. A single group at 67% tells you where the next hour of investigation belongs; a flat distribution tells you the usage is structural rather than a runaway job.

Avg per session separates two different patterns that produce the same total: many short sessions, which usually means automation, and a few long ones, which usually means people.

Once a grouping and set of filters is useful, save it from the group-by menu with Save current as…, or make it the view you land on with Set as default. A monthly reconciliation is much easier when the report opens on the billing period already grouped the way you check it.

The report can only attribute what your setup makes attributable:

All three are cheap to do at the start and awkward to retrofit, because the attribution applies from the moment a system registers.

Set the period to the billing period, leave the filters clear, and read the footer total. That is the number to compare against the invoice. Then group by actor to explain it.