Cookieless analytics: the GA4 alternative that needs no banner and sees your whole audience
Cookieless analytics measures your traffic without storing cookies on visitors’ devices or collecting personal data — it counts visits, pages and sources in aggregate rather than tracking individuals across the web. The reason to consider it is that Google Analytics isn’t actually free: it costs you a consent banner, an ongoing compliance burden, and — the part that surprises people — more than half your data, since an independent study found GA4 misses about 55.6% of traffic once you account for consent rejections and ad blockers. Cookieless tools flip all three problems at once. Because they set no cookies, they fall outside the consent requirement and need no banner, so they see your complete audience; and their tiny scripts make your site faster rather than slower. The trade is real and worth naming plainly: you give up deep funnels, session replay, user-level journeys and tight Google Ads attribution, so if your business runs on behavioural advertising, GA4 with proper consent or a fuller tool like Matomo is still the right call — and we’ll tell you so. But for the large majority of sites, “how many visitors, which pages, where they came from” is exactly what a cookieless tool provides, more completely and more privately. That’s why the sites we build use cookieless analytics by default: it’s the measurement that matches an architecture which doesn’t leak — no banner, complete data, faster pages, and numbers that are actually yours rather than fuel for someone’s advertising business. None of this is legal advice, the rules vary by where your visitors are, but for most sites the honest summary is that “free” analytics is the expensive option.
The hidden cost of “free” analytics
Google Analytics is free to install, which is precisely why its true costs go unexamined. The first is legal: GA4 loads a script that sets the persistent _ga cookie on a visitor’s device, and under Article 5(3) of the ePrivacy Directive, storing that identifier requires prior consent — which is why using Google Analytics without a consent banner is non-compliant in the jurisdictions where the rule applies (EU Biz Tools, 2026). Data protection authorities across Austria, France, Italy, Denmark, Finland, Norway and Sweden have ruled that GA use violated the GDPR, and the 2023 EU-US Data Privacy Framework resolved the transfer issue without removing that underlying consent requirement (Plausible, 2026).
The second cost is the data itself. An independent study found that GA4 fails to capture an average of 55.6% of traffic compared with cookieless analytics, with the consent banner accounting for most of the gap — every visitor who declines is simply never counted (Plausible, 2026). Ad blockers compound it, since they target Google Analytics specifically and can block the majority of a tech-savvy audience (Plausible, 2026). The uncomfortable result is that a tool chosen for accurate measurement can be missing more than half of what it’s meant to measure — while adding a banner, a privacy policy and often a paid consent platform to your costs. This is the analytics-shaped version of the point our pillar on respecting visitor privacy makes: the tracking you add is the cost you carry.
How cookieless analytics works — and why it needs no consent
The mechanism is the whole reason the consent problem disappears. A cookieless tool sets no cookies and reads nothing from the visitor’s device; it measures traffic in aggregate, typically deriving a daily-rotated, unrecoverable hash so that a visitor can’t be re-identified from one day to the next (Analytics Alternatives, 2026). Because nothing is stored on or read from the device, Article 5(3)‘s consent trigger is never pulled, and because no persistent personal identifier survives, the aggregate outputs — pageviews, sources, durations — have been accepted by EU authorities as processable under legitimate interest without consent, including a German DSK review of Plausible in 2022 (Analytics Alternatives, 2026).
That design has one honest limitation worth stating. Because the identifying hash rotates daily, a visitor who returns the next day is counted as new, so “unique visitors” over long periods is an approximation rather than a precise headcount (Snappa, 2026). For most sites that’s an acceptable trade for numbers that are complete, private and banner-free — but it’s the kind of limit worth knowing before you switch, and it’s why the honest framing throughout this guide is that cookieless is the right default, not a universal law.
The tools, compared
The category has matured to the point where the choice is about features and sovereignty, not whether the tool is credible. A few options cover most needs.
| Tool | Cookieless | Self-host | From | Best for |
|---|---|---|---|---|
| Plausible | Yes | Yes (free) | ~$9/mo | Simplicity, open source |
| Fathom | Yes | Lite | ~$14/mo | Agencies, unlimited sites |
| Matomo | Yes (mode) | Yes (free) | Free / €19+/mo | Feature depth, data ownership |
| Umami | Yes | Yes (free) | Free | Developers, high traffic |
Plausible is the lightweight, open-source, EU-hosted option with a single clean dashboard and a script under a kilobyte, ideal when you want the core numbers and nothing to configure (Vucense, 2026). Fathom — a Canadian company — sits close in spirit but bundles unlimited sites on every plan and stronger campaign tracking, which suits agencies and marketers (SegmentStream, 2026). Matomo is the feature-rich choice that comes closest to Google Analytics feature-for-feature, can import your historical GA data, and can be self-hosted for complete data ownership, at the cost of more setup (Vucense, 2026). Self-hosting any of them is roughly one to four hours of setup plus ongoing maintenance — free in licence, a few dollars a month in server in practice (Analytics Alternatives, 2026).
What you keep, and what you give up
Being honest about the trade is what separates advice from a sales pitch. The useful question, as the analysts put it, is whether your problem is privacy or measurement. Lightweight cookieless tools deliberately don’t offer multi-step funnel analysis, session replay, user-level journey mapping, detailed e-commerce attribution, or the tight Google Ads conversion integration GA4 provides (SegmentStream, 2026). If your business genuinely runs on behavioural advertising or needs revenue attributed across complex funnels, a minimalist tracker is the wrong tool, and the right answer is GA4 with proper consent management or a fuller platform like Matomo or PostHog (EU Biz Tools, 2026).
What you keep is, for most sites, everything you actually use. Visitors, popular pages, referral sources and basic goal or event tracking are exactly what cookieless tools are built to report, and for a marketing or informational site those are the numbers that inform decisions (Vemetric, 2026). The mistake the industry encourages is assuming you need GA4’s full behavioural apparatus because it’s there; most owners open GA4 to see traffic and sources and never touch the rest. If that’s you, the trade isn’t a loss at all.
The speed and honesty bonus
Two advantages come along for free once you switch. The first is performance: a cookieless script is a fraction of the size of Google’s analytics library — Plausible’s is under a kilobyte — and every script you remove from a page has a real cost in load time, so lighter analytics makes a site measurably faster, which feeds directly into the Core Web Vitals that affect ranking (Plausible, 2026). Privacy, in this one respect, is a speed feature.
The second is ownership. Google Analytics is free because your visitors’ behavioural data serves Google’s advertising business rather than you — the same logic that saw Universal Analytics shut down with no real migration path and years of history deleted when it stopped serving that purpose (Plausible, 2026). A cookieless tool, especially self-hosted, keeps your data in your account or on your own servers, audited by no one’s ad model, and answerable only to you. For a business that’s meant to own its website outright, owning its analytics is the consistent next step.
Why we use it by default
Stated plainly as our position: the sites we build use cookieless analytics by default, because it’s the measurement that matches the rest of the architecture. A site we build already self-hosts its assets and makes no third-party requests, so pairing it with analytics that sets no cookies keeps the whole thing consistent — no banner to interrupt visitors, complete data rather than a consent-shaped hole, a faster page, and numbers that belong to the owner. It’s the analytics equivalent of the no-banner outcome that the cookie question resolves to, and it’s why the two decisions are really one.
The honest gate applies here as everywhere, in two parts. We’re not lawyers, and while genuinely cookieless tools generally need no consent, your obligations depend on your jurisdiction and exactly what you measure, so a professional should confirm the specifics. And if your business truly needs behavioural analytics — funnels, session replay, Google Ads attribution — we won’t pretend a lightweight tracker covers it; we’ll set up GA4 with proper consent management, or a fuller privacy-first tool like Matomo, and do it correctly. For the large majority of sites, though, cookieless analytics is simply the better default: it tells you what you need to know, respects the people who visit, and costs you neither a banner nor half your data.
”Free” analytics is the expensive option
Step back and the case reduces to a single comparison. Google Analytics costs nothing to install and then charges you in the currencies that matter — a consent banner that annoys visitors, a compliance burden you carry indefinitely, roughly half your traffic data lost to rejections and blockers, and a dependence on a tool built for someone else’s business. Cookieless analytics costs a few dollars a month or a little server time, and pays you back in complete data, a faster page, no banner, and ownership of your own numbers.
So the useful way to decide is to ask what you genuinely need to measure, not what a free tool happens to offer: if the answer is visitors, pages and sources — which for most sites it is — a cookieless tool is more accurate, more private and less trouble, and you can run it alongside GA4 for a month to prove it before switching. Do that and analytics stops being a privacy liability bolted onto your site and becomes a quiet, honest instrument that respects the people it counts — the same “do it right once” logic that runs through our pillar on respecting visitor privacy and the whole way we build.
Frequently asked
- What is cookieless analytics?
- Cookieless analytics measures your website's traffic without storing or reading cookies on a visitor's device and without collecting personal data about individuals. Instead of tracking people across the web with a persistent identifier, tools in this category count visits, pages and referral sources in aggregate — often using a daily-rotated, unrecoverable hash so no one can be re-identified from one day to the next. Because nothing is stored on the device, these tools fall outside the consent requirement of the ePrivacy Directive, which is why they generally need no cookie banner. Plausible, Fathom, Matomo in cookieless mode, and Umami are common examples.
- Does cookieless analytics really need no cookie banner?
- For genuinely cookieless, aggregate-only tools, generally yes. The ePrivacy Directive's consent requirement is triggered by storing or accessing information on a user's device; a tool that sets no cookies and reads nothing doesn't trigger it, and the aggregate data it produces has been accepted by several EU data protection authorities as processable under legitimate interest without consent. So there's nothing to gate behind a banner. You still keep a short privacy or cookie policy describing what you measure, because transparency is a separate obligation — but the interruptive analytics banner disappears. This is general orientation rather than legal advice, and the details vary by jurisdiction.
- Is Plausible, Fathom or Matomo better?
- They suit different needs. Plausible is the lightweight, open-source, EU-hosted option with a single clean dashboard — ideal when you want the core metrics and nothing to configure. Fathom, a Canadian company, is close in spirit but includes unlimited sites on every plan and stronger campaign tracking, which suits agencies and marketers. Matomo is the feature-rich, self-hostable choice that comes closest to Google Analytics feature-for-feature and can be run entirely on your own servers for complete data ownership, at the cost of more setup. For a typical business site any of the three is more than enough; the choice is about how many features and how much data sovereignty you need.
- What do I lose by leaving Google Analytics?
- The deep, user-level features. Lightweight cookieless tools don't offer multi-step funnel analysis, session replay, individual user-journey mapping, detailed e-commerce attribution, or the tight Google Ads conversion integration that GA4 provides. If your business runs on behavioural advertising or needs to attribute revenue across complex funnels, that's a real trade-off, and the right answer is either GA4 with proper consent management or a fuller tool like Matomo — not a minimalist tracker. But for the large majority of sites, the metrics you actually use are visitors, popular pages and traffic sources, which cookieless tools provide completely.
- Is cookieless analytics more accurate than GA4?
- Often, and by a surprising margin. An independent study found that GA4 fails to capture an average of about 55.6% of traffic compared with cookieless analytics, with the consent banner accounting for most of the gap because visitors who decline are never counted. On top of that, ad blockers specifically target Google Analytics, so with tech-savvy audiences the blocked share can reach the majority. A cookieless tool that needs no consent and isn't part of the ad-tech ecosystem sees traffic those factors hide, which is why sites often report noticeably higher — and more complete — numbers after switching.