How to Export Data from Umami Analytics

TL;DR: On Umami Cloud, go to Settings → Data → Export. Umami emails you a link to a gzipped CSV file, one website at a time. On a self-hosted Umami, your data sits in your own PostgreSQL database, so you export it with one SQL query. Both ways give you a CSV you can keep as a backup or import into Pieowl.

With web analytics, you want control over your data. Maybe you're switching tools, making a backup, or want to dig into your data somewhere else. Either way, you need a way to get it out of Umami.

How you do it depends on whether you use Umami Cloud or run Umami yourself. This guide covers both. It also shows how to bring the file into Pieowl if you decide to switch.

Export from Umami Cloud

Umami Cloud has a built-in export. Here are the steps from Umami's docs (checked October 2026):

1. Log in and open the profile menu in the side navigation.
2. Click Settings.
3. Go to Data and click Export.
4. Fill in the form and click Export.
5. When the export is ready, Umami emails you a link to download it.

The file is a CSV compressed with gzip. It holds your pageviews, events, sessions and event data for the website you picked.

A few things to know:

• One website per export. If you track many sites, you need one request per site. Umami asks you to contact their support if you have many sites or a lot of data.
• It runs in the background. You don't wait on the page. You wait for the email instead, and big sites take longer.

Export from a self-hosted Umami

Umami's export docs only cover Umami Cloud. When you self-host, though, all your data is already in your own PostgreSQL database. That means you can export it with a standard psql command. Nothing extra to install.

Umami keeps events in the website_event table and visitor details (browser, country and so on) in the session table. The query below joins them into one flat CSV, with the same column names Umami uses.

If you run Umami with its default Docker Compose file, the database service is called db and the user and database are both umami. Run this from the folder with your docker-compose.yml:

docker compose exec -T db psql -U umami -d umami -c "COPY (
  SELECT e.*,
         s.browser, s.os, s.device, s.screen, s.language,
         s.country, s.region, s.city, s.distinct_id
  FROM website_event e
  JOIN session s ON s.session_id = e.session_id
) TO STDOUT WITH CSV HEADER" | gzip > umami-export.csv.gz

This writes umami-export.csv.gz next to your Compose file. A few tips:

• Check your names first. If you changed the service name, user or database in your setup, use your own values.
• Export one site. Add WHERE e.website_id = 'your-website-id' before the closing bracket.
• Split big exports. Add a date filter such as WHERE e.created_at >= '2025-01-01' to export in chunks.
• It only reads. The query doesn't change anything, and Umami keeps collecting data while it runs.

The query uses the table and column names from Umami's current database schema (checked October 2026). If you run an older Umami version, check that the columns match.

Import your Umami data into Pieowl

Pieowl has a built-in Umami importer. It reads the CSV format above, plain or gzipped.

1. Upload your CSV file somewhere it can be downloaded over https://, such as a file link from your cloud storage. Pieowl only fetches files from public addresses.
2. In Pieowl, open your site and go to Settings → Imports.
3. Choose Umami and paste the file link.
4. If you exported more than one site, match each Umami website ID to a Pieowl site. Otherwise everything goes into the site you're in. Then start the import.

The import runs in the background. Pageviews, events, sources, UTM tags, pages, countries and devices come across, so your charts don't start from zero.

Why people move from Umami

Umami is a good, privacy-focused tool, and for many sites it is all you need. Its Hobby plan is free for 100K events a month on one site (checked October 2026), while Pieowl has no free plan: it starts with a 14-day free trial, then $9 a month for 10K events.

People tend to look at Pieowl when they want more product analytics in one place:

• Ask your AI assistant. Connect Claude, Cursor or another assistant through Pieowl's MCP server and ask about your data.
• Product analytics. Funnels, retention, cohorts and user profiles on every plan.
• Session replay on every plan. On Umami Cloud, session replays are part of the Business plan at $200 a month (checked October 2026).
• Revenue attribution. See revenue by source, campaign, landing page and country.

Pieowl sets no cookies, like Umami. Pieowl runs in our EU cloud. Unlike Umami, it has no self-hosted option.

Making the move decision

If you run a simple blog or portfolio, Umami may be all you need. Its simplicity keeps you focused on your content.

If you run a store, a SaaS app, or any business where user behavior drives revenue, compare the two on the reports you actually use. Export your data, import it into Pieowl during the 14-day free trial, and look at your own numbers side by side. Then decide based on real use, not feature lists.

Think about the full cost too. Self-hosting Umami is free in license fees, but you still pay for the server and your time. Pieowl has no self-hosted option, so there is no server for you to run.

Best practices for data export

• Pick a quiet time. The export only reads data, but a big query still puts load on your database.
• Check the file right away. Open the CSV and look at the date range, the websites included, and a few events.
• Split large exports. Export by date range so files stay manageable. You can import several files one after another.
• Keep exports safe. Umami doesn't collect personal data, but your traffic data can still be sensitive. Encrypt exports you keep for a long time or send over a network.
• Export on a schedule. Regular exports are a cheap backup, even if you stay with Umami. A monthly reminder or a small script does the job.