---
title: "Exporting your data to Excel or CSV"
description: "Monthly trend, Sentry quota, crash-free by release, stores, server metrics: take out the history StatHall keeps, ready for a pivot table or a presentation."
lang: en
canonical: "https://www.stathall.com/en/docs/exporter-ses-donnees/"
updated: "2026-10-09"
---

# Exporting your data to Excel or CSV


## What StatHall keeps and the services don't give back

Sentry keeps only **90 days** of project statistics: after that, your
month-by-month quota consumption no longer exists there. StatHall collects it
daily and **keeps** it, along with the monthly issue trend, crash-free rate
release by release, daily store metrics or, for a server, 400 days of hourly
system metrics. The **Exports** screen hands them
back to you as a file.

Datasets labelled **StatHall exclusive** exist nowhere else in this form.

## Two entry points

- **An app's or a server's Exports tab**: that target's data only.
- **Exports**, in the *My targets* menu: the whole fleet visible under the
  scope chosen at the bottom of the sidebar, or one target picked from the
  list. Every row carries the target name, so a pivot table compares your apps
  or servers as is.

You only export targets whose data you can see: your own, or those shared
with you. A dataset that does not apply to the target (server metrics in an
app's tab) is not offered.

## Period and granularity

- **Period**: 3, 6 or 12 **whole** months (from the 1st), *Since the
  beginning* (the oldest known data), or custom dates.
- **Granularity**: *Month* or *Day*, for time series (quota, events per
  issue, sessions, stores). Lists (issues, reviews, versions, anomalies) are
  filtered on the period, without grouping.

Each card shows the range where data exists; a card without data in the
chosen period has its buttons disabled.

## Excel or CSV

- **Excel**: one sheet per dataset, with **filters** and a frozen header,
  typed dates and numbers. Click inside the data, then *Insert › PivotTable*:
  Excel picks up the whole range. *Ctrl+T* formats it as a table if you wish.
  An **About** sheet dates the extraction, restates the period and defines
  every column.
- **Everything in one workbook** gathers every dataset with data in the
  period into a single Excel file.
- **CSV**: in English, `,` separator and decimal point; in French, `;` and
  decimal comma. UTF-8 encoded.

## Available datasets

| Dataset | Content |
| --- | --- |
| Monthly trend — issues and Sentry quota | new issues, events received, dropped and filtered, reviews and average rating, per month |
| Sentry quota consumption | events received, dropped, filtered, by day or month |
| Crash issues | issues active during the period, with their counters |
| Events per issue | events and affected users, issue by issue |
| Sessions and crash-free by release | sessions, crashed sessions, crash-free rate |
| Store metrics | downloads, revenue, active users, stability, ratings |
| Store reviews | the period's reviews, **without their author** |
| Versions and release health | adoption, health verdict, crash-free |
| Server system metrics | CPU, memory, disks, swap…: minimum, weighted average, maximum per series |
| Probe availability | checks, failures, availability and latency per probe |
| Server events | systemd units, updates… reported by the agent |
| Component end of support | current state of the Support scale, whatever the period |
| Detected anomalies | every anomaly opened during the period, apps and servers |

Monthly volumes are **summed** and rates **averaged**; the *Current month*
column flags a month whose totals are still partial.

## Limits

A file holds at most **200,000 rows** per dataset. Beyond that it is
truncated and the screen tells you: narrow the period or switch to months.
Every export is recorded in your account's audit log.