Is self-hosted Matomo or Plausible cheaper than Google Analytics?
If you are weighing a renewal, an Analytics 360 contract or a data residency concern, the license fee disappears but infrastructure, implementation, upgrades and internal skills take its place. Matomo On-Premise Community and Plausible Community Edition cost nothing to license, so the budget question becomes what you pay to run them. GA4 is free to use but not free to operate once you need its exports, and Analytics 360 is a paid contract that buys service levels and scale. These two differ in kind, so compare each against its own cost base rather than against a single headline number. Start with hosting, which scales with traffic. Matomo's requirements FAQ sizes a single server with at least 4 CPUs, 8 GB of RAM and 250 GB of SSD for up to 1 million pageviews a month, and recommends separate app and database servers with 8 CPUs and 16 GB of RAM each for up to 10 million pageviews a month. Plausible CE needs Docker, Docker Compose and at least 2 GB of RAM, with a CPU that supports SSE 4.2 or NEON. As a reference point only, Matomo Cloud starts at 29 per month (excluding tax), and Plausible Cloud lists $9, $14 and $19 monthly plans for 10k pageviews. Ask your team to price their own pageview volume against these tiers before deciding. The recurring costs you will not see on a license invoice are the ones to budget. Your own staff or partner will handle installation, upgrades, backups, uptime and security patches, and Plausible's self-hosted page notes that Community Edition operators manage these themselves. Matomo's paid premium plugins, such as Heatmap and Session Recording, are sold per number of users under a separate EULA, so price them as an ongoing line if you need them. Also ask whether your data already flows through the GA4 BigQuery export, because storage and query costs for that export stay on your Google Cloud bill whichever analytics tool you choose.
Which cost components change when you leave Google Analytics
This table lines up the commercial costs of GA4 or Analytics 360 against the self-hosted costs of Matomo or Plausible, marking each row as one-time or recurring. Use it to build your own cost model, since most line items depend on your traffic, plugin choices and internal staffing rather than on a fixed price. The main takeaway is that the software licence is free on both sides, so the difference lies in infrastructure, plugins, implementation and ongoing operations.
| Cost component | Commercial side (GA4 or Analytics 360) | Self-hosted side (Matomo or Plausible CE) | One-time or recurring |
|---|---|---|---|
| Software licence | GA4 is free to use. Analytics 360 is a paid contract with terms set by Google, so confirm the fee and SLA in your agreement. | Matomo On-Premise Community and Plausible Community Edition are both open source with no licence fee under GPLv3 and AGPLv3. | Recurring for Analytics 360, none for the open-source core |
| Servers or cloud instances | None for the tag itself. Your site's pages load Google's script. | Matomo sizing is set by monthly pageviews, starting at 1 million for a single server. Plausible CE needs Docker, Compose and at least 2 GB of RAM. Confirm sizing for your traffic. | Recurring |
| Premium Matomo plugins | Not applicable. | The Heatmap and Session Recording plugin is paid, priced per number of users under the InnoCraft EULA, with a 30-day free trial. Quote it separately. | Recurring, if licensed |
| BigQuery export and storage | GA4 standard exports are capped daily, and the BigQuery export is a separate cost in your Google Cloud bill. Confirm current storage and query charges. | Not needed for reporting. Raw data lives in your own database, so you pay for its storage and backups instead. | Recurring |
| Consent tooling | Consent banners and a consent management platform are usually required for GA4 in the EU, which adds licence and integration work. | Matomo configured to CNIL conditions may avoid a consent banner for analytics. Plausible uses no cookies, so no banner is needed. Verify with your legal team. | One-time setup, with some recurring review |
| Implementation and migration | Initial GA4 tagging, tag manager work and goal setup in place. | Installing the tracker, importing history with the Google Analytics Importer or Plausible's GA4 import, and rebuilding goals. Raw visit history does not carry over. | One-time |
| Upgrades, backups and monitoring | Handled by Google. Universal Analytics data was permanently deleted after the week of July 1, 2024. | Your team applies Matomo and Plausible releases, runs backups, watches uptime and patches security fixes. Plausible CE gets fixes only when you apply releases. | Recurring |
| Staff and agency time | Agency retainers or internal analyst time for reporting and tag changes. | Platform engineering or a managed implementation partner for hosting, upgrades and support. Ask any partner for a written, fixed scope. | Recurring |
What moves the cost of self-hosted analytics up or down
Self-hosted analytics costs are set less by the license, which is free for Matomo On-Premise and Plausible Community Edition, than by traffic volume, the number of sites and users, paid add-ons and the skills your team already has. The cards below show which inputs to fix in writing before you compare any quote or internal estimate.
Monthly pageviews set the server size
Pageviews and hits are the main sizing input. Matomo's requirements FAQ sizes a single server with at least 4 CPUs, 8 GB of RAM and 250 GB of SSD for up to 1 million pageviews a month, and recommends separate app and database servers with 8 CPUs and 16 GB of RAM each for up to 10 million pageviews a month. Ask your team for a 12-month traffic forecast, including campaign peaks, before you size hardware or cloud instances.
Sites and users multiply the footprint
Each additional website, language version or subdomain adds tracking configuration, reports and storage, while every extra analyst account adds administration and access reviews. Count the sites and named users in scope before you agree a deployment model. Also confirm whether reporting is shared across business units, since that changes how you split servers and roles.
Paid plugins priced per user
Matomo's Heatmap and Session Recording plugin is a paid premium plugin sold per number of users under the InnoCraft EULA rather than the GPL, with a 30-day free trial. Its cost recurs as headcount grows, so model it separately from the free core. Ask the vendor for the current per-user rate, the renewal terms and whether it applies to read-only stakeholders.
Availability and backup targets
Self-hosted operators own uptime, backups and restore testing. Define a recovery point and recovery time target, then price redundant database hosting, offsite backup storage and monitoring to meet it. Missing data during an outage affects your numbers directly, so confirm how long collection can be interrupted without losing trust in the dashboards.
Plausible CE needs Docker and memory
Plausible Community Edition requires Docker and Docker Compose, a CPU that supports SSE 4.2 or NEON (required by ClickHouse) and at least 2 GB of RAM. Plan a host that meets these requirements with headroom, and budget the time to maintain Docker images and the ClickHouse data store. Check that your hosting provider exposes the CPU features ClickHouse needs before you commit to a virtual machine type.
Security patches depend on your team
Plausible Community Edition is a long-term release published twice per year, is community supported only, and gets security fixes only when the operator applies the release, while Plausible Cloud ships updates multiple times per week. The recurring cost is the staff time to track releases and apply them quickly. Agree a patch window and an owner before go-live, and budget for an upgrade plan on each release.
In-house skills in PHP and databases
Matomo runs on PHP with MySQL or MariaDB behind a web server, and Matomo 6.0.0 requires PHP 8.1 or newer and MySQL 8.0 or newer or MariaDB 10.6 or newer. Plausible adds Docker and ClickHouse. If your team lacks these skills, the gap becomes a recurring service fee or a retainer, so ask any implementation partner to split its quote into setup and ongoing operations.
How to build the comparison with your own numbers
This sequence turns vendor claims into a total cost of ownership comparison built from your own traffic, feature usage and written quotes, so the numbers you compare reflect your environment.
Export current GA4 volumes and BigQuery usage
Pull your monthly pageview and event counts for the last 12 months from GA4 reports, and record how much you send through the BigQuery export and what that export costs today in your Google Cloud billing account. Your analytics or data engineering lead usually owns this, and finance should confirm the BigQuery line items. You know this step is done when you have one table of monthly pageviews, events and BigQuery spend, with the source of each figure noted. Those pageview counts become the input for server sizing in a later step.
List the reports and features you actually use
Ask the people who open analytics each week which reports, segments, funnels, audiences, conversions and dashboards they rely on, and which ones they have not opened in months. Mark anything that depends on raw events, such as exploration analysis or custom segmentation, because those behave differently after migration. Product, marketing and compliance should each sign off on the list. The check is a short inventory where every feature is tagged as needed, nice to have or unused.
Decide whether heatmaps or session recording are needed
Decide early whether you need heatmaps or session recordings, because Matomo's ePrivacy FAQ says that features such as heatmaps or session recordings remove the CNIL consent exemption for audience measurement. Heatmaps and session recording are also paid premium plugins for Matomo, sold per number of users under the InnoCraft EULA, so they add a recurring license line. Legal or privacy counsel should confirm the consent position for each market you serve. The output is a documented yes or no for each feature, with the consent consequence recorded.
Price server options from your own infrastructure
Take the monthly pageview figure from step one and compare it with the sizing guidance in Matomo's requirements FAQ, which describes a single server for up to 1 million pageviews a month and separate app and database servers for up to 10 million. Plausible Community Edition requires Docker, Docker Compose and at least 2 GB of RAM, so size it against your own volumes. Request prices for compute, storage, backups and bandwidth from your cloud provider or data center team, since those costs are yours to pay. The result is a costed server option for each tool, tied to a stated traffic level.
Add implementation, migration and upgrade effort from written quotes
Collect written quotes for installation, tag replacement, historical import, consent configuration, backups and upgrades, from your internal team and from any implementation partner you are evaluating. Ask each provider to state scope, deliverables and who carries the work of each upgrade, since Community Edition operators handle upgrades and security fixes themselves. Include the importer limits in your scope: Matomo's Google Analytics Importer and Plausible's GA4 import bring over aggregated report data, not raw hits, so the Visitor Log and segmentation will not cover imported history. The check is a quote per tool that lists one-time and recurring effort separately.
Account for what GA4 or Analytics 360 costs you today
Record the current GA4 cost in its real form: zero license fee for standard properties, or the contracted Analytics 360 fee if you pay for it, plus the BigQuery, consent tooling, tag management and staff time you already spend. Mark which of these you could retire after leaving, such as a cookie banner vendor that a cookieless or consent-exempt configuration would make unnecessary. Note that Plausible states it uses no cookies, so no cookie banner is required for its script, while Matomo's exemption depends on configuration. Finance and procurement should agree on the figures. The output is a baseline with each item marked as retained, reduced or removed.
Compare total cost over a fixed period you choose
Choose one fixed comparison period, and put one-time costs such as setup, migration and parallel running in the first period, then add recurring costs such as servers, backups, premium plugins, support and upgrade work for each period. Run the comparison for the GA4 baseline, self-hosted Matomo with and without premium plugins, and self-hosted Plausible Community Edition, using the same traffic assumption across all three. Show which features are missing in each option, because a cheaper tool that drops a needed report is not a like-for-like comparison. Leadership should review the table with the feature gaps listed next to it. Done means one table, one period and one stated traffic level.
Matomo or Plausible: which cost profile fits your team
Marketing teams that rely on Visitor Log, segmentation and custom reports usually fit Matomo On-Premise best. The Community edition is free to license under the GPLv3, so the recurring cost is the infrastructure and the people who run it. Matomo's requirements FAQ sizes a single server with at least 4 CPUs, 8 GB of RAM and 250 GB of SSD for up to 1 million pageviews a month, and recommends separate app and database servers with 8 CPUs and 16 GB of RAM each for up to 10 million pageviews a month.
- Server sizing: Pageview volume is the main driver of hardware cost, so measure your monthly traffic before you size anything.
- Import limits: The free Google Analytics Importer brings in aggregated reports only, so Visitor Log, segmentation and funnels do not work on imported history.
- Premium plugins: The Heatmap and Session Recording plugin is a paid premium plugin sold per number of users under the InnoCraft EULA, with a 30-day free trial.
- Parallel running: Plan a period where GA4 and Matomo run side by side, since historical raw detail stays in GA4.
- Operating work: Your team handles upgrades, backups and PHP and database maintenance.
Before committing, confirm which current reports your marketers use and whether they depend on raw visits that the import cannot carry.
Content, publishing and privacy-focused teams that want a simple traffic dashboard often fit Plausible Community Edition. It is licensed under the AGPLv3 and needs Docker, Docker Compose, a CPU with SSE 4.2 or NEON support and at least 2 GB of RAM. Plausible states that its script uses no cookies, so the cookie banner question is simpler, though you should still confirm the legal position for your own jurisdiction with counsel.
- Feature gaps: Marketing funnels, user journeys, ecommerce revenue goals, single sign-on, the Sites API, the Looker Studio connector and advanced bot filtering are Cloud-only.
- Release cadence: Community Edition is a long-term release published twice per year, with security fixes only when you apply the release. Plausible Community Edition v3.2.1 was a security release published on May 15, 2026.
- Historical import: The GA4 import brings historical stats as daily aggregates, cannot import ecommerce revenue and approximates unique visitors by summing daily uniques.
- Google integration: Search Console keyword data and GA4 import are configured with GOOGLE_CLIENT_ID and GOOGLE_CLIENT_SECRET.
- Cloud option: Community Edition users can export and import data to and from Plausible Cloud, which is a route if you outgrow self-hosting.
Ask whether your team needs funnels or revenue goals, since those features decide whether the Community Edition covers your reporting.
Teams that need event-level data that no other party can export or reuse, such as an internal data science group or a regulated business that must keep every visit under its own control, should weigh the storage and pipeline costs more heavily. Both tools keep data on your servers, but the ownership question depends on what you store and where your backups go. Matomo's importer and Plausible's import both work from aggregates, so the raw history of the old tool must be exported separately if you need it.
- Raw history: Export GA4 data through the BigQuery route before cancellation, because Universal Analytics data can no longer be exported.
- Storage growth: Raw visit and event storage grows with traffic, so budget disk and backup capacity for the retention you choose.
- Backups and access: You own backup, restore and access control, which become your team's duties.
- Warehouse access: Matomo and Plausible do not replace a warehouse, so decide where analysts will query raw data.
- Retention policy: Set retention periods in writing, since CNIL's exemption conditions cap tracker lifetime at 13 months.
Confirm with your data owner which raw fields must be kept and for how long before selecting an architecture.
Questions buyers ask before comparing analytics costs
How much effort does a self-hosted Matomo or Plausible install take, and who does it?
Plan for an in-house platform or DevOps owner to do the install, plus a named person who handles upgrades, backups and uptime. Plausible's Community Edition documentation says operators manage installation, upgrades, capacity, uptime, backups and security themselves, and the repository requires Docker, Docker Compose and at least 2 GB of RAM. Matomo On-Premise needs a PHP and MySQL or MariaDB stack behind a web server. The effort depends on your hosting model, your existing monitoring and whether you front the tool with a tag manager. Confirm the exact steps against the current release notes in your own environment before estimating hours.
Can our team run a self-hosted analytics tool without outside help?
Often yes for a small site on a single server, provided someone owns patching and restores backups. Teams usually need outside help for initial sizing, the consent configuration, the GA import and a documented upgrade runbook. Plausible Community Edition gets security fixes only when the operator applies a release, and its releases come out twice per year, so your team must watch for them. Matomo has a free On-Premise edition with forum support, so paid support is an optional decision. DeployOpen plans, implements, hardens, backs up, upgrades and documents self-hosted platforms in your infrastructure, and most engagements are fixed-scope with the scope agreed in writing before work starts.
What drives the server bill for self-hosted Matomo or Plausible?
Server cost is driven mainly by monthly pageviews, retention and how many sites you run. Matomo's requirements FAQ sizes one server with at least 4 CPUs, 8 GB of RAM and 250 GB of SSD for up to 1 million pageviews a month, and recommends separate app and database servers with 8 CPUs and 16 GB of RAM each for up to 10 million pageviews a month. Your bill then adds backups, monitoring, storage growth and staff time. Get quotes for your own cloud or data center rates, since list prices vary by provider and region, and confirm the sizing against your actual traffic and the vendor's current requirements.
Can Plausible Cloud serve as a cost comparison point?
Yes, as a reference point for a managed price, but not as a like-for-like feature match. Plausible's homepage lists Cloud plans starting at 10k monthly pageviews, with Starter, Growth and Business tiers and custom Enterprise pricing, and a 30-day free trial with no credit card required. Its self-hosted page lists features missing from Community Edition compared with Cloud, including marketing funnels, user journeys, ecommerce revenue goals, single sign-on, the Sites API, the Looker Studio connector and advanced bot filtering. Compare the subscription against your server, staff and upgrade costs, and check which of those missing features you actually use.
Can we export our GA4 data before we cancel Google Analytics?
Export it before you cancel, because the exports and their limits decide what history you can carry over. Google states that Universal Analytics data was permanently deleted after access ended the week of July 1, 2024, so GA4 history should be treated the same way once you stop paying or disable the property. Matomo's Google Analytics Importer and Plausible's GA4 import bring over aggregated report data, not raw hits. Raw history for long-term analysis usually depends on the BigQuery export, so confirm which export your property has, how far back it reaches and what your storage costs will be in the warehouse.
Can we drop the consent banner if we switch to self-hosted analytics?
Possibly, but only if the configuration meets the CNIL conditions for the consent exemption. CNIL's sheet 16 requires, among other things, that purposes stay limited to audience measurement and A/B testing, that the tracker is used for a single publisher scope without cross-checking with other processing, that the last byte of the IP address is truncated, and that the tracker lifetime is at most 13 months. CNIL also says most large audience measurement offerings do not qualify regardless of configuration. Matomo's FAQ notes that adding User ID, heatmaps or session recordings removes the exemption. Have your legal or privacy team confirm the setup against the current CNIL guidance and your other national authorities before removing any banner.
Which costs are one-time and which recur after a migration?
Migration work is mostly one-time, while hosting, upgrades, backups and any paid plugins recur. One-time items include scoping, installing the tracker or tag manager, importing history, rebuilding goals, running a parallel period and documenting the setup. Recurring items include servers, storage, staff time for patching and monitoring, and paid premium features such as Matomo's Heatmap and Session Recording plugin, which is sold per number of users under the InnoCraft EULA with a 30-day free trial. Ask any quote to separate the one-time work from the monthly run cost and to name each paid plugin.
What should we ask for in a written scope before comparing quotes?
Ask for a written scope that names the tool, the version, the hosting target and the features in use. Include the expected monthly pageviews, the number of sites, the retention period, the consent approach, the GA history to import and the list of reports, goals and segments to rebuild. Ask what is excluded, such as paid plugins, Enterprise support and BigQuery costs, and how upgrades and security fixes will be handled after go-live. DeployOpen does not publish a price list, and a quote follows from the written scope, so a clear scope is what makes two quotes comparable.
What a fixed-scope quote for leaving Google Analytics should include
A fixed-scope quote for leaving Google Analytics starts with an inventory, not a price. The written scope should list every property, web stream, app stream and tag manager container in use, with the sites and subdomains each one covers, so the implementer knows which tags are replaced and which must stay on GA4 for Google Ads or other reasons. It should also name the data import method and its limits in plain terms: Matomo's Google Analytics Importer and Plausible's GA4 import both bring over aggregated report data, not raw hits, so Visitor Log, segments and funnels will not work on imported history. Ask the vendor to confirm which reports and dimensions carry over in your configuration, and record that decision in the scope before work starts.
The second part of the scope covers the server and the people who run it. Hosting size depends on monthly pageviews, so the quote should state the pageview range it assumes, the Matomo or Plausible edition, and the hosting location, then name who provisions and patches the database, web tier and backups. Plugin licences belong here too. Matomo's Heatmap and Session Recording plugin is a paid premium plugin under the InnoCraft EULA, so the scope should say whether it is in or out and who holds the licence. Upgrade and backup responsibilities need explicit owners, since Plausible Community Edition gets security fixes only when the operator applies a release. Finally, require a testing and cutover plan with parallel running, a rollback step, consent settings reviewed by your legal team, and a training session for analysts. Confirm each item with the implementer before signing.
How DeployOpen scopes a fixed-price analytics migration
DeployOpen plans, implements, hardens, documents and supports self-hosted analytics inside your own cloud accounts or data centers. Engagements are fixed-scope, so the cost comparison between GA4 or Analytics 360 and self-hosted Matomo or Plausible is turned into a written scope before any build work starts.
How we scope
- We inventory the GA4 properties, tags, events and any BigQuery export your team depends on today.
- We size the self-hosted server against your actual monthly pageview volume and growth expectations.
- We list which paid Matomo plugins, such as heatmaps or session recording, are in or out of the written scope.
How we build
- We install Matomo On-Premise or Plausible Community Edition in your environment, with separate app and database hosts where volume requires it.
- We configure the tracker, consent settings and IP truncation so the setup matches your legal review, not a default assumption.
- We run the historical import and check which reports carry over and which need to be rebuilt from raw data.
How we hand over
- We document the architecture, sizing assumptions, configuration keys and the cost components your finance team will review.
- We deliver a runbook covering upgrades, backups and the security release process for your chosen edition.
- We run both tools in parallel for an agreed period and hand over the cutover checklist before retiring the old tag.
How we support
- Ongoing upgrades, monitoring and capacity reviews can be covered by a retainer after handover.
- Security fixes are applied on a schedule you approve, which matters for Plausible Community Edition's twice-yearly cadence.
- We can prepare your configuration and evidence for compliance reviews, though independent auditors issue any formal assessment.
