8 best Plausible alternatives in 2026, compared

The best Plausible alternative depends on which Plausible limit you hit. As of November 2026, Pirsch is the closest like-for-like switch and imports your Plausible history, Rybbit and Umami put funnels on plans that cost a fraction of Plausible’s Business plan ($139 a month at 1M page views), and Fathom suits agencies that outgrow Plausible’s 10-site cap.

The short answer by reason for leaving:

• The closest switch, history included: Pirsch. A German company hosting in Germany, cookieless, with event properties, an API, 50 sites and unlimited members from $6 a month.
• Funnels for half the price of Plausible Business, open source, history included: Rybbit, $69 a month at 1M events against $139.
• A free plan, or funnels on your own server: Umami.
• More than 10 sites, reported to clients by email: Fathom.
• Each visit step by step, heatmaps, and revenue read from Stripe or Paddle: invisit, the tool we build.
• Product analytics with session replay, feature flags and experiments: PostHog.
• Google Analytics depth on your own server: Matomo.
• Page views and Core Web Vitals for nothing: Cloudflare Web Analytics.

Stay on Plausible if what it leaves out is what you wanted left out. Plausible keeps every provider that touches visitor data European-owned, CDN included, loads about 2.5 KB of script, shows Search Console queries, sends scheduled reports by email and Slack, imports Google Analytics 4, and lets you self-host its Community Edition. If none of the six reasons below applies to you, switching costs more than it saves.

Why do people look for Plausible alternatives?

People leave Plausible for six reasons, and every one of them is documented on Plausible’s own pages: the features its Business plan gates, the caps on sites and seats, no hosted free plan, no view of a single visit, revenue that only comes from your code, and the price at high volume.

• Funnels, properties and the API need Business. Plausible Starter and Growth leave out funnels, custom properties, revenue goals and the Stats API. Business costs $39 a month at 100K page views, $139 at 1M and $339 at 10M, against $19, $69 and $169 on Starter (Plausible’s pricing).
• Sites and seats are capped. Starter covers one site and one login, Growth three of each, Business 10 of each. Past that is a custom Enterprise plan (plan guide).
• There is no hosted free plan. The trial lasts 30 days (trial). The free, self-hosted Community Edition leaves out funnels, user journeys, revenue goals, SSO and the Sites API (Community Edition).
• No visit view, heatmaps or web vitals. Plausible’s when not to use Plausible page rules out individual session views and heatmaps by design. The closest it gets is scroll depth and, on Business, aggregate user journeys.
• Revenue only comes from your code. Revenue goals record an amount your checkout sends, on Business, and Plausible’s docs say renewals, refunds and chargebacks cannot be tied to a visit, so expect less revenue than your billing tool shows (revenue tracking).
• High volume gets expensive. Plausible Business costs $339 a month at 10M page views and custom events; above 10M is a custom Enterprise plan, which is also the only plan with raw event exports (export options).

Plausible counts page views plus custom events across a team’s sites, and outbound link clicks, file downloads and form submissions count as custom events. Our Plausible vs invisit comparison walks through each plan’s limits at every volume.

What to look for in a Plausible alternative

Someone leaving Plausible should judge alternatives on seven points: the price of what you would buy Business for, history import, how long a visitor stays recognisable, where data lives, script weight, what Plausible cannot show, and self-hosting.

• The price of Business features. Compare the cheapest plan with funnels and event properties at your real volume. Some tools bill page views, some every event, one each stored property.
• History import. Plausible’s site-settings export is a ZIP of daily CSV files. Pirsch, Rybbit and DataFast import it as is; most others start empty.
• How long a visitor is recognisable. Plausible deletes the salt behind its visitor hash every 24 hours (data policy). Some alternatives link a visitor for a month or across days, and some set cookies that need consent in the EU.
• Data location and ownership. Plausible stores data at Hetzner in Falkenstein, Germany, with only EU-owned providers, CDN included (EU hosting).
• Script weight. Plausible’s per-site script measured 2,485 bytes gzipped; the tools below run from about 2.1 KB to 119 KB.
• What Plausible cannot show: each visit step by step, heatmaps, web vitals, revenue from your payment provider.
• Self-hosting, with or without the features you would otherwise pay for.

We left out Simple Analytics, which shares Plausible’s EU ownership and light script but has no funnels inside one visit, and Google Analytics 4, which brings back cookies and a consent banner. Both are covered in our Google Analytics alternatives.

Plausible alternatives compared: funnel prices, history import and privacy

Prices are monthly on monthly billing, for the cheapest plan with funnels, in each vendor’s own unit: page views plus custom events (Plausible, Pirsch), events (Rybbit, Umami, invisit, PostHog) or hits (Matomo). Fathom and Cloudflare have no funnels, so their column shows the plain price. PostHog sends at least two events per page view, so 1M page views is about 2M events. Umami bills each stored event property as one more event.

1. Pirsch: best like-for-like switch that keeps your Plausible history

Pirsch is the closest tool to Plausible with Plausible’s plan limits loosened: a cookieless dashboard from a German company, data stored in Germany, and a built-in import for Plausible’s export. Pirsch’s cheapest plan includes event properties, the API, 50 sites and unlimited members, which Plausible reserves for Business or does not offer at all.

Pirsch bills page views plus custom events, and 10% of session-extension requests. As of November 2026, Pirsch’s pricing is $12 a month at 100K and $54 at 1M on Standard, and $27 and $159 on Plus, the plan with funnels, revenue tracking, A/B testing and segmentation. At 10M, Standard costs $159 and Plus $479. A year costs ten months’ price. There is no free plan; the trial runs 30 days or 100,000 page views, whichever comes first, with no card (billing).

What Pirsch does better than Plausible: Standard includes events with up to 20 metadata pairs, the REST API, 50 sites, unlimited members, email reports, traffic spike alerts, Search Console, a URL shortener and session analysis, which Pirsch’s features page describes as in-depth session summaries of individual journeys. Pirsch hashes the IP address, User-Agent, date and a per-site salt, so a visitor is recognisable for 24 hours at most, and the IP address is never stored (privacy). Its DPA names one sub-processor, Hetzner, with processing in Nuremberg and Falkenstein.

What Pirsch does worse: funnels and revenue need Plus, which costs less than Plausible Business below 500K ($27 against $39 at 100K), the same at 500K ($99) and more above it ($159 against $139 at 1M, $479 against $339 at 10M). Revenue is a custom metric over an amount in event metadata, still sent from your code. Pirsch has no heatmaps or web vitals, self-hosting needs an Enterprise licence (its AGPL-3.0 core is a Go library), and pa.js measured about 3.5 KB gzipped against Plausible’s 2.5 KB. Over the plan limit, the dashboard stops at the day you hit it and Pirsch collects for five more days.

Moving from Plausible takes the least work of any tool here. Upload Plausible’s site-settings ZIP under Import / Export (up to 100 MB per upload) and Pirsch imports visitors, pages, entry and exit pages, sources, locations, devices, browsers and operating systems (Plausible import), though imported days cannot be filtered by page or referrer. In your code, props becomes meta (events).

Choose Pirsch if you like Plausible’s dashboard but want properties, more sites and teammates without paying for Business, and want your history to come with you.

2. Rybbit: best open-source alternative with funnels, session replay and your Plausible history

Rybbit puts what Plausible gates to Business on its cheapest plan: funnels, journeys, goals and an API, plus sessions, user profiles, retention, web vitals and error tracking. Rybbit is AGPL-3.0 like Plausible, self-hosted with Docker or hosted as Rybbit Cloud, and it imports Plausible’s export.

Rybbit Cloud bills events per organization, web vitals and autocaptured clicks included (billing). As of November 2026, Rybbit’s pricing is $19 at 100K and $69 at 1M on Standard (5 sites, 3 members, 3 years of data), and $39 and $139 on Pro (unlimited sites, session replay, 5 years). At 10M, Standard costs $249 and Pro $499. A year costs eight months’ price. There is a 7-day trial and no free plan, and an organization over its limit stops collecting until the month resets or it upgrades.

What Rybbit does better than Plausible: funnels for half the price of Plausible Business at 1M ($69 against $139), every session with its pages and events, session replay on Pro, Search Console, a hosted MCP server, and a Bots page on Cloud that shows which AI agents and crawlers read which pages (Bots). Rybbit Cloud runs on Hetzner in Germany and Finland (subprocessors).

What Rybbit does worse: the visitor ID is a hash of IP address and User-Agent that persists across days unless you turn on User ID Salting, and Rybbit’s site settings docs say the unsalted ID may count as personal data under the GDPR. Plausible’s privacy model is stricter out of the box. Rybbit’s script measured about 11.8 KB gzipped, close to five times Plausible’s, Standard caps you at 5 sites and 3 members, and there are no heatmaps or payment-provider revenue reports.

Moving from Plausible: upload the site-settings ZIP without unzipping it, on Standard (data up to 24 months back) or Pro (up to 60 months). On Cloud, one import runs at a time, and imported events count against the plan limit of the month they happened in, so events in a month busier than your plan are skipped (data import). Self-hosted imports have no limits. In your code, the props wrapper goes away.

Choose Rybbit for funnels and sessions at Plausible Starter’s price up to 1M, replay on Pro, and the option to self-host later.

3. Umami: best free or self-hosted alternative with funnels

Umami has funnels, journeys, retention and goals on every plan, the free one included, which makes it the cheapest way to get the reports Plausible keeps for Business. Umami is MIT-licensed: self-host it on Node.js and PostgreSQL, or use Umami Cloud.

Umami Cloud counts each page view and custom event as one event, and each stored property as one more. As of November 2026, Umami’s pricing has Hobby free up to 100K events (1 site, 6 months of data, no teammates), Pro at $20 for 1M events (20 sites, 10 members, 2 years) and then $0.00003 an event, and Business at $200 for 10M with heatmaps, session replays and unlimited sites and members. A site with 1M page views and 100K sign-ups carrying two properties each uses 1.3M events: $29 a month on Pro.

What Umami does better than Plausible: a free hosted plan, funnels everywhere, sessions with visitor profiles, opt-in web vitals, an LLM channel for six AI assistants, an official MCP server from Pro, and revenue imported from Stripe, Paddle and four more providers from Pro (payment providers). Self-hosted Umami has the same analytics as Umami Cloud, replays and heatmaps included, where Plausible Community Edition drops funnels.

What Umami does worse: the visitor hash salt rotates monthly by default, so a returning visitor stays linked for up to a month (visitor identification), against a day on Plausible. Umami Cloud’s EU or US region is chosen at signup and its docs do not name the EU country. Hobby keeps 6 months of data and Pro 2 years, where Plausible keeps everything while you pay, and heatmaps on Cloud mean the $200 plan.

Moving from Plausible: Umami Cloud imports CSV files by request from Pro, handled by Umami’s team (import data); its docs do not name Plausible as a source. Events map directly, without the props wrapper (track events).

Choose Umami for funnels without paying, or for a full-featured dashboard on your own server. See Umami vs invisit.

4. Fathom: best for agencies with more than 10 sites

Fathom Analytics fixes one Plausible limit well: sites. Every Fathom plan includes at least 50 sites, where Plausible allows 1 on Starter and 10 on Business, and Fathom emails reports to clients on a schedule and gives each one a share link. Fathom is a cookieless hosted service from Canada, with one plan that has every feature.

Fathom bills page views, custom events and API requests against one allowance. As of November 2026, Fathom’s pricing is $15 a month up to 100K, $60 at 1M and $200 at 10M, and 50 more sites cost $10 a month. There is no free plan, and the 7-day trial needs a card.

What Fathom does better than Plausible: at $60 for 1M it costs less than any Plausible plan at that volume, email reports are unlimited, per site or rolled up, and share links take a password and an expiry date. Fathom has an AI tab and Search Console terms, and its script measured about 2.1 KB gzipped, a little lighter than Plausible’s.

What Fathom does worse: it does less than Plausible Business, not more. Fathom has no funnels, an event is a name plus an optional value in cents with no custom properties (events), and an account has one login. EU visits are processed on EU servers, but Fathom stores the anonymized analytics in the US (data isolation), a step back from Plausible’s EU-owned setup.

Moving from Plausible: Fathom’s importer reads Google Analytics 4 only (GA importer), so your Plausible history stays in the CSV export. plausible('Sign Up', { props: { plan: 'pro' } }) becomes fathom.trackEvent('Sign Up') and the property is lost; a revenue amount becomes _value in cents.

Choose Fathom if you run many small client sites and report by email. See Fathom vs invisit.

5. invisit: best for visit timelines, heatmaps and revenue from your payment provider

invisit, the tool we build, answers the questions Plausible rules out by design: what one visit did, step by step, where people click and how far they scroll, and what your traffic earned according to your payment provider. invisit is hosted, cookieless web analytics with an MIT tracker and a proprietary backend, and it stores data in Falkenstein, Germany, like Plausible.

invisit bills page views, custom events (automatic outbound-link and download events included), conversions and short-link clicks, and never bills bots, crawler reads or web vitals. As of November 2026, the free plan covers 20,000 events a month on 4 sites, commercial use allowed, and paid plans cost $14 for 100K, $49 for 1M and $199 for 10M a month (pricing). Funnels, event properties (32 per event), the REST API and the MCP server are on every plan, free included.

What invisit does better than Plausible: every visit’s pages and events in order, live in the dashboard and at any age through the API and MCP; click and scroll heatmaps; web vitals from real visitors; and payments, refunds, fees and net read from Stripe, Paddle, Polar, Lemon Squeezy, Creem or Dodo Payments, the last 90 days backfilled on connect (connect a provider). invisit also has an AI tab for 19 assistants, crawler reads through @invisitdev/server, alerts to Slack, Discord, ntfy or a webhook, and unlimited members, at $49 for 1M against $139 on Plausible Business.

What invisit does worse: invisit cannot import your Plausible history. It has no self-hosting, no email reports, no Search Console, no session recordings or cohorts, and it does not attribute revenue to sources or campaigns, which Plausible’s revenue goals do. The core script is about 8.5 KB gzipped against Plausible’s 2.5 KB, the free plan’s 20,000 events suit small sites only, and Cloudflare, a US company, runs invisit’s DNS and CDN, so a policy requiring EU ownership end to end rules invisit out where Plausible passes.

Moving from Plausible: rename events without spaces, since names use letters, digits and _ . : - (events), drop the revenue option and connect your provider. See Plausible vs invisit, plan by plan.

6. PostHog: best when your site is the front door to a product

PostHog is a product suite with web analytics inside: session replay, heatmaps, feature flags, experiments, surveys and error tracking on one event stream. Pick PostHog when the questions that outgrew Plausible are about signed-in users rather than traffic.

PostHog bills every event: page views, page leaves, autocaptured clicks and web vitals. As of November 2026, PostHog’s pricing gives 1M events a month free, then $0.00005 each up to 2M, falling with volume. A page view sends at least two events by default, so 100K page views fit the free tier, 1M cost about $50 a month and 10M about $643. Seats are unlimited.

What PostHog does better than Plausible: depth. Funnels, retention, cohorts and person profiles, 5,000 session recordings a month free, an EU Cloud in Frankfurt, SQL over raw events, an official MCP server and a free managed proxy.

What PostHog does worse: by default posthog-js sets a first-party cookie for 365 days and uses localStorage (persistence). Cookieless mode needs settings in code and in the project, and turns off replay and surveys for visitors without consent. The main script measured about 119 KB gzipped, close to 50 times Plausible’s, and the MIT hobby self-host is unsupported.

Moving from Plausible: PostHog’s Plausible migration guide is marked experimental. Its Python script turns the export’s visitors and pages files into page view events with random person IDs, and leaves out UTM tags, sessions, bounce rate, custom events and entry and exit pages. Historical imports are free but require a paid product analytics plan.

Choose PostHog for a product team that needs replay, flags and experiments more than a light traffic dashboard. See PostHog vs invisit.

7. Matomo: best self-hosted alternative with Google Analytics depth

If Plausible Community Edition is too thin to self-host, Matomo On-Premise is the deep option: free under GPL v3 with unlimited hits, segments, goals, ecommerce, custom reports, a tag manager and scheduled reports. Matomo Cloud adds funnels, heatmaps, session recordings, A/B tests and cohorts.

Matomo bills hits: page views, events, downloads, outlinks and other tracking requests. As of November 2026, Matomo Cloud costs $42 a month for 100K hits, $204 for 1M and $975 for 5M, with prices above that quoted, for 30 sites and 30 users. On-Premise premium plugins cost extra per year, such as Funnels at $239 for 1 to 4 users (marketplace).

What Matomo does better than Plausible: a visits log and visitor profiles, heatmaps and recordings, A/B tests, ecommerce reports, a Google Analytics importer, official mobile SDKs, an official MCP server, and an AI Assistants channel. Matomo Cloud stores data on AWS in Frankfurt.

What Matomo does worse: the standard tracker sets first-party cookies, _pk_id for 13 months (cookie list). Turning cookies off is one setting, and the CNIL mode for consent-free tracking in France switches off the Visits Log, User ID, heatmaps and recordings. The core matomo.js measured about 22 KB gzipped, self-hosting needs PHP and MySQL, and at 1M Matomo Cloud is the most expensive hosted option here.

Moving from Plausible: Matomo documents no importer for Plausible’s export. Matomo events take a category, an action, a name and a value (event tracking), so plausible('Sign Up', { props: { plan: 'pro' } }) becomes _paq.push(['trackEvent', 'Account', 'Sign Up', 'pro']), and more context needs custom dimensions.

Choose Matomo if policy requires self-hosting and you need far more than Community Edition offers. See Matomo vs invisit.

8. Cloudflare Web Analytics: best free option if page views are all you read

If Plausible’s lowest price, $9 a month for 10K page views, is the reason you are leaving, and you only read page views, referrers, countries and Core Web Vitals, Cloudflare Web Analytics does that for free with no traffic cap. It needs a Cloudflare account and one script tag; your DNS can stay where it is.

Cloudflare’s docs list Web Analytics as free on every plan, and the limits page caps sites, not traffic: 10 per account for sites not proxied through Cloudflare. Cloudflare says Web Analytics uses no cookies or localStorage and does not fingerprint visitors.

What Cloudflare does better than Plausible: it costs nothing and reports Core Web Vitals, which Plausible does not.

What Cloudflare does worse: nearly everything Plausible adds on top of page views. Cloudflare’s FAQ lists no custom events, no UTM parameters, six months of data, and only the last 7 days unsampled before data is aggregated to around 10%. A visit is any page view whose referrer is not your own hostname, so there are no unique visitors. The beacon measured about 10.3 KB gzipped, and Web Analytics is not covered by Cloudflare’s EU Customer Metadata Boundary.

Moving from Plausible: there is no import and nothing to map, because Cloudflare has no custom events. Goals, properties and revenue have no home.

Choose Cloudflare Web Analytics for a personal site, docs or side project where page views and page speed are all you read.

When should you stay on Plausible?

Stay on Plausible when one of its strengths is a requirement, not a preference. None of the eight alternatives matches all of these at once.

• EU ownership end to end. Only European-owned providers touch Plausible’s visitor data: Hetzner, UpCloud and Bunny as its CDN. Pirsch comes closest, with Hetzner in Germany as the only sub-processor in its DPA.
• The lightest script. Plausible’s per-site script is about 2.5 KB gzipped; only Fathom, at about 2.1 KB, is lighter in this list.
• Search Console queries, scheduled email and Slack reports, and GA4 import in one simple dashboard.
• One site at high volume with core stats only. Plausible Starter costs $169 at 10M, less than every paid alternative here except Pirsch Standard ($159).
• Self-hosting a simple dashboard. You can move from Plausible’s cloud to Community Edition and keep your history, because Plausible’s export and CSV import use the same format. You lose funnels, journeys and revenue goals.

If your only complaint is the Business price on a small site, the saving from switching is small: at 100K page views Plausible Business costs $39 a month, $12 more than Pirsch Plus and $20 more than Rybbit Standard.

How do you switch from Plausible without losing your history?

Export from Plausible’s site settings first, import the ZIP where your new tool accepts it, then run both trackers side by side for two to four weeks before you remove Plausible’s script. Every step after the export can be undone; the export cannot be redone once Plausible deletes your stats.

1. Export from site settings, not from the dashboard. In Plausible’s site settings, Imports & Exports, Export to CSV queues a ZIP and emails you a link: one CSV per report, with a row per day for each page, source and so on (export guide). The dashboard’s own export is a different ZIP, with totals for the selected range and only the top 300 rows per report (top 100 for pages); the importers below expect the site-settings files.
2. Export imported data separately. The site-settings export leaves out anything you imported into Plausible, such as Google Analytics history. Download that from the dashboard with Include imported data switched on.
3. Import the ZIP where you can. Pirsch and Rybbit take it as is, and so does DataFast, which brings page views, visitors, referrers, devices, countries and UTM data but not goals or custom events (DataFast’s guide); DataFast’s default script sets a cookie, which kept it off this list (DataFast vs invisit). Elsewhere, keep the ZIP as your archive.
4. Install the new tracker next to Plausible’s; both can run on the same pages.
5. Rewrite your events with the table below, and rebuild goals and funnels.
6. Replace revenue goals. Pirsch sums an amount in event metadata on Plus, Fathom takes _value in cents, Umami takes revenue and currency properties or reads your payment provider from Pro, and invisit reads the provider instead of your code.
7. Compare both dashboards for two to four weeks. Expect small gaps: each tool filters bots its own way and defines a visit differently.
8. Remove Plausible’s script, any proxy rules for it, and cancel.

If you used Plausible’s class names to track clicks without JavaScript, Pirsch keeps the same idea with its own prefix, and Umami, Rybbit and invisit use data attributes:

<!-- Plausible -->
<button class="plausible-event-name=Sign+Up plausible-event-plan=pro">Start trial</button>

<!-- Pirsch -->
<button class="pirsch-event=Sign+Up pirsch-meta-plan=pro">Start trial</button>

<!-- Umami -->
<button data-umami-event="Sign Up" data-umami-event-plan="pro">Start trial</button>

<!-- Rybbit -->
<button data-rybbit-event="Sign Up" data-rybbit-prop-plan="pro">Start trial</button>

<!-- invisit -->
<button data-invisit-event="sign_up" data-invisit-prop-plan="pro">Start trial</button>

Imported history is daily totals in every tool that accepts it: single visits and events cannot be rebuilt from Plausible’s export, and raw event exports are Enterprise only.

Before you cancel Plausible

• The site-settings export is downloaded for every site, not only the dashboard export.
• Data you once imported into Plausible is exported from the dashboard with imported data included.
• The ZIP is imported into the new tool if it supports it, and any skipped months are noted.
• Every plausible() call and plausible-event-* class is replaced.
• Goals, funnels and revenue have a new home and are firing.
• Both tools ran side by side for two to four weeks.
• Plausible’s script and any proxy rules for it are gone.