Alternatives to Umami

What changes when you move off Umami

Umami comes in two editions, and leaving them is not the same job. Umami Cloud is a subscription: you pick a data region at sign-up, US or EU, the plan sets retention, and the documentation calls account deletion permanent and says to export first. Self-hosted Umami is the same code under the MIT license, on Node.js 18.18 or newer with PostgreSQL 12.14 or newer, or the Docker image. A subscription carries an end date, an installation does not, so a server you keep running holds yesterday’s reports while a new tool starts counting.

The report set is the other half of the decision. Umami’s plan table puts breakdown, funnel, retention, user journey, goals, revenue, attribution and UTM reports on every cloud plan, the free Hobby one included, with session replays and heatmaps on Business. Several tools in the table above stop at pages, sources, countries and devices.

What to check before you switch

1. What you can take out, and from which side. In Umami Cloud the export sits in Settings, Data: choose a website and a period, and the vendor emails a link to a gzip-compressed CSV of page views, events, sessions and event data. The plan table lists it on every plan. On your own server the database is the archive.
2. The unit on the bill. A pageview counts as one event, a custom event as one more, and each stored event property as another; above the included volume the vendor bills per event and says collection is not interrupted. A plan sold by pageviews counts something narrower, so a property-heavy setup sits elsewhere on the other ladder.
3. Retention, which decides what an export can hold. Hobby keeps six months, Pro two years and Business five; older data is not there to move.
4. What happens to the history. Deleting a cloud account takes the subscription and the data with it. Where the next tool has no importer, keep the old dashboard reachable and run both scripts for a few weeks before you remove the old snippet.
5. Who holds the data. Umami Cloud runs servers in both regions, which is why its row carries EU and US at once; on your own server the location is yours.

Which of these can import your data

Three tools read a Umami export. Swetrix takes the archive as it arrives, unopened, under Settings, Import data. Rybbit expects a CSV under Site Settings, Import, and its cloud plans cap how far back an import reaches, 36 months on Standard and 60 on Pro, with imported events counted against the monthly allowance and neither limit on a self-hosted instance. TinyAnalytics parses the file in the browser tab, and because a Umami export holds individual events, timestamps survive the move.

Inside Umami the documented route runs one way: the cloud imports a PostgreSQL export of a self-hosted instance or a CSV file, for Pro customers, with the vendor’s team running the job. Nothing is documented in the other direction.

The rest of the table reads other sources: Plausible’s CSV import takes files in its own format, Pirsch documents imports from Google Analytics, Plausible and Fathom Analytics, and PostHog lists Google Analytics, Matomo, Mixpanel, Amplitude and Plausible.

Questions and answers

Can I take my Umami history with me?

Yes. In the cloud the export runs from Settings, Data and arrives as an emailed link; on your own server you also hold the database. Three tools here read that export directly; the rest start counting on install day.

Is leaving Umami Cloud the same as leaving an installation I run?

No. A cloud account has dates on it: the plan sets retention, and account deletion is permanent. Your own installation keeps working after you point the script elsewhere, so a long overlap costs nothing extra.

I run Umami myself. What changes if I move to a hosted tool?

Updates, backups, the database and the bot list stop being your work, and volume limits, retention periods and per-plan features start to apply. The open source and self-hosted list shows which tools here still run on your own server.