Free Website Analytics in 2026: 10 Free Plans and Trials

Free website analytics in 2026 comes in two shapes: a free plan, or a free trial that ends in a paid plan. Seven of the ten services checked here publish a free plan, such as Umami Cloud's Hobby plan at "Up to 100K events per month" (Umami pricing), PostHog's "Analytics 1M events" (PostHog pricing) and Vercel Hobby's "50,000 events / month included" (Vercel Web Analytics pricing). Plausible, Fathom and Pirsch offer a trial only. Every figure here comes from the ZeroRaised analytics tiers dataset, built from each vendor's own pages and checked on 22 Oct 2026.

Disclosure: ZeroRaised publishes this guide and the analytics tiers dataset it draws on, and appears in the final section. Every limit, price and quote below comes from that dataset or the vendor pages it cites, and the dataset was built only from each vendor's own pricing, docs, privacy and terms pages, and each vendor fact links to the page it came from. Where the dataset says a value is not stated, so does this guide. ZeroRaised has no affiliate deal with any company named here.

TL;DR#

Free web analytics plans and trials for indie sites, from each vendor's own pages, checked 22 Oct 2026:

For a small product, the headline cap is rarely what decides things. Whether there is a free plan at all, how long data is kept, whether a commercial site is allowed and what the vendor says about cookies usually matter first.

What the analytics tiers dataset found, checked 22 Oct 2026#

On 22 October 2026, ZeroRaised checked the free plans and trials of 10 web analytics services that indie developers often use and published the results as the analytics tiers dataset. It records the free plan or trial, the event or pageview cap, sites, data retention, the vendor's own cookie and GDPR wording, commercial use, card requirement and any self-hosted option.

Three findings matter most if you are about to add a tracking script:

• Three of ten have no free plan. Plausible, Fathom and Pirsch publish a trial only. All ten candidate vendors could be checked on their own pages, so none was dropped.
• The units do not match. Events, pageviews, page views, events pooled across projects, or no number at all. Nothing is converted.
• Cookie wording is the vendor's own. Most say they do not use cookies, Google says its tags use first-party cookies, and GoatCounter hedges.

The dated event is the check itself. The vendor pages fetched on 22 Oct showed no dated free plan change between 8 and 22 October 2026.

Free website analytics compared: all 10 vendors#

The full dataset in one table, with values as each vendor writes them. "not stated" means the pages checked did not state the value; it does not mean "included", "not included", "allowed" or "not allowed". For the trial-only vendors, cells show the lowest paid plan as the vendor lists it. Cookie wording has its own table further down.

Two cells need care. Umami's card requirement is not stated, not "No": the Hobby card says "Get started" without saying whether a card is asked for (Umami pricing). And for Google Analytics the cap is not stated on the pages checked, so the dataset does not guess one.

The columns are not a league table. A trial row and a free plan row describe different things, and the cap column mixes events, pageviews and page views.

Events, pageviews and page views: why the caps do not line up#

The caps look like they sit on one scale. They do not, because each vendor decides what it counts.

Two rules catch people. Umami counts event properties as events, so a site that sends custom events with several properties uses its allowance faster than its pageviews suggest (Umami pricing). And Vercel and PostHog pool usage: Vercel says "events are shared across all projects" (Vercel Web Analytics pricing), so three side projects on one Hobby account share one allowance.

Umami's 100K event cap (Umami pricing) and Fathom's 100,000 pageview plan (Fathom pricing) are not the same size, and nothing on these pages converts one into the other. To compare two vendors honestly, count your own pageviews and custom events for a month, then read each vendor's rule against both numbers.

Free plans from analytics vendors: Umami, PostHog, Simple Analytics and GoatCounter#

These four sell or run analytics as their main product and publish a free plan you can stay on.

Umami Cloud. Hobby is "$0 / month" with "Up to 100K events per month", "1 website" and "6 month data retention" (Umami pricing). The next paid plan as the vendor shows it is Pro at "$20 / month" with "1 million events per month included", after a "14-day free trial". What happens when Hobby passes its limit is not stated. Umami "can be self-hosted for free", but "email reports and the streaming API" are "not available in the self-hosted version".

PostHog. PostHog is a product analytics suite rather than a pageview counter. Its free plan is "No credit card, not a trial." with "1 project, 1-year data retention" (PostHog pricing). The edge is hard: "Usage stops at the free tier limits, so you can't be charged by surprise." and "On the free plan, any additional events are permanently dropped". Session recordings sit beside the events allowance, but "Recordings on the free plan are retained for 1 month." Pay-as-you-go lists "6 projects – up from 1" (PostHog pricing). PostHog can be self-hosted, but its docs say "self-hosted deployments are officially unsupported" (PostHog self-host docs).

Simple Analytics. The free plan is "Free for hobby sites" with "1 user 5 websites 1 month history" and "Badge required" (Simple Analytics pricing). It "offers unlimited pageviews", but "we have a fair use policy". The trade is history: "Any traffic data older than 30 days will be automatically deleted from our servers." Sign-up starts a "14-day free trial (no credit card required)", after which you choose a paid plan or the free one.

GoatCounter. GoatCounter is "a free donation-supported hosted service or self-hosted app" with no numbered cap (GoatCounter). Its pricing note reads: "Running your personal website or small-to-medium business on it is fine, but sending millions of pageviews/day isn't." One account covers "any number of sites/domains" (GoatCounter sign-up). There is no paid plan; "Donations are accepted via Github Sponsors to cover server costs." Retention and card requirement are not stated.

Analytics that comes with a platform: Vercel, Google Analytics and Cloudflare#

These three come as part of a bigger platform, and their free terms follow that platform's rules.

Vercel Web Analytics. Vercel says "Web Analytics are free for Hobby users within the usage limits": "50,000 events / month included" and a "Reporting Window 1 Month" (Vercel Web Analytics pricing). "Hobby teams can't purchase additional events." Over the limit, "a three day grace period will start before Vercel will stop capturing events", then you "Wait 7 days" or upgrade. The rule that matters for anything you charge for is on the main pricing page: "Our Hobby plan is for personal, non-commercial use." (Vercel pricing).

Google Analytics. Google says it "gives you the tools, free of charge, to understand the customer journey and improve marketing ROI." (Google Analytics). The paid tier is "Analytics 360". Retention is a setting: "you can set the retention period of user-level data to: 2 months 14 months" (GA data retention), and "The two-month retention period is always applied to age, gender, and interest data regardless of your settings." Sites, card and self-hosting are not stated.

Cloudflare Web Analytics. The product page reads "Privacy-first, lightweight, accurate web analytics, for free" (Cloudflare Web Analytics). The limit is on sites: "Not proxied through Cloudflare 10", "Proxied through Cloudflare No limit" (Cloudflare limits). The FAQ says "you can access data for the previous six months" and "We retain unsampled beacon data for the past 7 days, after this point data is aggregated down to around 10%." (Cloudflare FAQ).

If your site already runs on Vercel or sits behind Cloudflare, you can switch these on without adding a vendor. Their terms are written for the platform first, though: Vercel's non-commercial rule covers the whole Hobby plan.

Trial only: Plausible, Fathom and Pirsch#

Three vendors in the dataset publish no free plan. Each offers a trial, then a paid plan sized by monthly pageviews or page views.

Plausible. Limits for the trial itself are not stated. The free route is self-hosting: Plausible says "The code is public and auditable" and that you can "run it yourself if you want to." (Plausible). Its self-hosting page adds: "Self-hosting is community supported only." (Plausible self-hosting).

Fathom. Fathom's FAQ is blunt: "Do you offer a free plan? Nope" (Fathom pricing). The trial runs on the plan you choose, with your card added, and Fathom says it will "email you before your plan starts billing". Extra sites come "in packs of 50 for $10/month".

Pirsch. At the page view limit, "We will continue to collect data for your sites for a further five days." (Pirsch pricing). Self-hosting appears only as "On-premise Installation" under Enterprise.

A trial that needs a card becomes a paid plan unless you cancel, so put the end date in your calendar.

Cookies and consent, in each vendor's own words#

Many makers look past Google Analytics because of cookie banners. The dataset does not assess any of these claims; it quotes what each vendor says on its own pages. Whether your site needs a banner depends on your setup and your visitors' law, which is a question for an adviser, not a pricing page.

Three rows are worded differently from the rest:

• GoatCounter hedges with "probably", and its terms put the duty on you: "It is your responsibility to ensure that all local laws regarding data collection are followed (e.g. GDPR)." (GoatCounter terms).
• PostHog's answer depends on setup. Its docs say to use a cookie banner "If you use PostHog with cookies on your website (for logged out users)" (PostHog GDPR docs). The pricing page says nothing on cookies.
• Vercel speaks of third-party cookies only. The page quoted does not say whether first-party storage is used (Vercel analytics privacy).

Read the table as what each vendor says it does. The claims are not identical, and none is a ruling on your site.

The small print on free website analytics plans#

The cap is rarely what moves a small site to a paid plan. Three patterns matter more.

What happens at the limit. Vendors either stop collecting, keep collecting for a while, or keep collecting and talk to you.

A launch is when this bites. If a post does well, a plan that drops events or pauses collection loses the days you most want to read later. The traffic spikes guide covers what else to check before a launch.

Commercial use. Only two vendors state it. Vercel's Hobby plan "is for personal, non-commercial use" (Vercel pricing), and GoatCounter says a "small-to-medium business" on it "is fine" (GoatCounter terms). Simple Analytics calls its free plan "Free for hobby sites". For the rest, commercial use is not stated, which is not the same as allowed.

Retention and sites. Simple Analytics keeps "1 month history" (Simple Analytics pricing) and Vercel Hobby has a one month reporting window, so export what you want to keep. If you run several small products, the site rule may matter more than the cap: Umami Hobby allows "1 website" (Umami pricing), PostHog "1 project", and Cloudflare's 10 site limit applies only to sites not proxied through it (Cloudflare limits).

How to choose free website analytics for a small product#

The dataset does not rank vendors. These questions sort the options for a particular site.

Free for good, or free for a month? Seven vendors publish a free plan and three a trial. A trial is a good way to run two tools side by side, but it ends.

Is the site commercial? If yes, Vercel's Hobby plan is out by its own terms, and for most others you will need to read the terms, because the pricing pages do not say.

How many pageviews and events do you send? Count both for a month, then read each vendor's unit against your numbers using the units table above.

How far back do you need to look? One month of history suits a quick look after launch, not a year-on-year comparison.

What will you say about cookies? Check the specific claim of the tool you pick, and your own setup, rather than a general idea that a tool is privacy-friendly.

Would you host it yourself? Plausible, Umami, PostHog and GoatCounter publish a self-hosted route; Pirsch lists "On-premise Installation" under Enterprise only (Pirsch pricing). You trade a plan limit for a server to maintain.

What does the next step cost? Know the next paid plan as the vendor shows it before you need it.

The rows mix monthly plans and per-event rates in different units, so they are not a price comparison. If you have not priced your own product yet, how to validate a SaaS idea before building is a better first read than any analytics price list.

What will you measure in the first month? Most makers want visits, where they came from and how many signed up. The first 100 users guide covers where early visitors tend to come from, and the launch checklist for bootstrapped makers puts analytics in order with everything else to set up before launch.

Where ZeroRaised fits#

Analytics is how you find out whether a launch did anything. ZeroRaised keeps the analytics tiers dataset for that reason: the free plans and trials of 10 web analytics services, every figure linked to the vendor's own page, with a visible check date and a CSV download. The ZeroRaised data hub lists it with the other datasets, including the free hosting and database tiers and the free auth and login tiers. If a vendor's page has changed since 22 Oct 2026, the vendor's page wins.

ZeroRaised is a free launch site for bootstrapped makers whose products have no institutional VC or accelerator equity. Once your analytics is in place and you can see where visitors come from, you can submit your product for review. The maker perks page lists separate offers for eligible makers, some of which cut running costs.

So, free website analytics in one line: seven of the ten vendors checked publish a free plan and three offer a trial, but they count usage in different units, keep data for different lengths of time and say different things about cookies, so match the small print to your site before you match the headline number.