# Reading Your Usage

The **Usage** panel sits below the summary on **Settings → Billing**. It answers one
question — *what is costing this much* — by re-pivoting the same period across the attributes
Suger records as consumption happens.

![The Usage panel — group-by tabs, filters, the daily spend chart, and the period grouped by resource with quantity, cost and share columns](images/billing-usage-by-resource.png)

---

## Choose the month

The period selector at the top of the page drives **both** panels, so the summary and the
usage table always describe the same month. It offers the last **12 months**, newest first,
with the current month marked *(current)*.

Selecting a closed month shows that month's settled figures. The credit balance above is
always your balance *now*.

## Group the period

The tab strip re-pivots the same consumption. Every pivot totals the same amount.

| Group by | Answers |
|---|---|
| **Resource** | What kind of thing did the money go on — tokens, sandbox, storage |
| **User** | Which member of the organization caused it |
| **App** | Which Insulin app it ran under |
| **Model** | Which model was used |
| **Provider** | Which provider served it |

Each row shows a **quantity**, the **cost**, and the row's **share** of the period. The unit
is named only in the **Resource** grouping: every other grouping mixes resources measured in
different units — tokens with processor-seconds — so a single unit label would be wrong.

### Rows with nothing to attribute

Some consumption genuinely has no value for the dimension you picked: a scheduled job has no
user, a direct API call has no app, sandbox storage has no model. Those rows are **named
rather than dropped** —

| Grouping | The unattributed row reads |
|---|---|
| User | *No user — scheduled and system runs* |
| App | *No app* |
| Model | *No model — sandbox and storage* |
| Provider | *No provider — sandbox and storage* |

![The same period grouped by user, with three named members and an italic 'No user — scheduled and system runs' row carrying the remainder](images/billing-usage-by-user.png)

This is why the pivots reconcile. If unattributed consumption were left out, the by-user view
would quietly total less than the by-resource view, and neither would match your bill.

## Narrow it with filters

Four multi-select filters — **users**, **apps**, **models**, **providers** — sit under the
tabs. Grouping and filtering answer different questions: grouping is *what did we spend it
on*, filtering is *what did this team spend it on*.

A filter's choices come from the grouping currently shown, so **the filter matching the
active tab is the one with values to pick**. To narrow by user, switch to the **User**
grouping first and the Users filter fills with the people who actually used AI this period;
the selection then stays applied as you switch to another grouping. This is also why no
filter ever offers someone who never used AI — the list is built from the period's own
results, not from your member list.

With any filter applied the panel says so, because **a filtered total is a subset of the
period, not the whole bill**. **Clear filters** returns to the full period.

## The daily spend chart

One bar per day of the period, with the peak day highlighted and labelled above the chart.
Days with no spend render as an empty slot rather than being closed up — a chart that
bridged a quiet weekend would draw a slope across spending nobody can account for.

## Why the table can look emptier than your balance

The usage breakdown is re-aggregated **every couple of hours**. The credit balance above it
comes straight from the ledger and moves within about a minute of a request.

So this sequence is normal:

1. You run something expensive.
2. The **Credit balance** drops immediately.
3. The **Usage** table still shows the period without it.

The panel says as much above the table. If the two disagree, **the balance is the current
one** — the table is catching up, and nothing has been lost or double-counted.

An empty table therefore means *nothing has been aggregated for this period yet*, which is
not the same as *you have used nothing*.

## What this panel does not do

There is no CSV export and no per-request drill-down. The finest grain available here is a
grouped row for the period. If you need the underlying detail — to reconcile a specific job
or a specific day — contact [support@suger.io](mailto:support@suger.io) while the
consumption records are still inside their 90-day retention window.