# From Google Analytics to Plausible > Source: > Published: 2026-07-20 12:00:00+00:00 # Why We Switched from Google Analytics to Plausible Over the past couple of years, Lightning Jar has replaced Google Analytics with [Plausible](https://plausible.io/) on our own applications and websites, and on most of the client sites we build and maintain. Like our [move away from WordPress](/blog/leaving-wordpress), this was not a snap decision. It came from watching how the tool actually got used, what it cost our visitors, and what it cost our clients in legal and operational overhead. Here is the reasoning, in roughly the order it mattered to us. ## Privacy: Measure Traffic, Not People Google Analytics is free because the data is valuable to Google. Visitor behavior collected through GA feeds an advertising ecosystem, and the tracking works by attaching identifiers to individual people: cookies, device identifiers, IP-derived signals. That model made us increasingly uncomfortable. Our clients are B2B companies who want to know whether their content is working. None of them need a behavioral profile of an individual visitor, and neither do we. Plausible takes the opposite approach. It counts page views, referrers, countries, and devices without cookies, without persistent identifiers, and without storing personal data. You learn what content resonates and where visitors come from, which is the question marketing teams are actually asking. What you give up is the ability to follow one person around, which is a capability we are glad to be rid of. ## Less Invasive Tracking Is Also Lighter Tracking The privacy difference shows up in the page weight. The GA4 tracking script is roughly 135 KB gzipped. Plausible's script is about 2.5 KB, more than fifty times smaller. On sites where we work hard to keep every page fast, shipping a large tracking bundle to every visitor so that we can read a bar chart never made sense. There is an accuracy angle too. Because GA is the backbone of the ad-tech ecosystem, ad blockers target it aggressively, and consent banners let a large share of visitors opt out entirely. GA then fills the gaps with modeled estimates. Plausible sits outside the ad-tech ecosystem, so it is blocked less often, and what it reports is a direct count rather than a model. In our experience the numbers are simpler to explain to a client precisely because there is less machinery behind them. ## Simpler Privacy Policies, Fewer Banners This was the sleeper benefit. When your analytics stores nothing on the visitor's device and collects no personal data, the compliance picture gets dramatically simpler: - No cookie consent banner is required for the analytics itself. For several of our clients, analytics was the only thing forcing a banner onto the site. - Privacy policies shrink. There is no need to explain Google's data processing, advertising signals, or international data transfers on behalf of your analytics vendor. - The regulatory ground stops shifting under you. Between 2022 and 2023, data protection authorities in Austria, France, Italy, Denmark, Finland, Norway, and Sweden ruled that the use of Google Analytics violated GDPR because of data transfers to US servers. The EU-US Data Privacy Framework later eased that specific conflict, but the episode made the point: building your measurement stack on contested data flows is a liability. Plausible's cloud service runs entirely on EU-owned infrastructure, and the data never leaves the EU. For a small marketing site, the difference between "we use a cookieless analytics tool that collects no personal data" and a multi-page disclosure about Google's processing is real money in legal review time, and a better experience for every visitor who no longer has to dismiss a banner. ## Open Source: Software We Can Audit and Keep Lightning Jar has a long-standing commitment to [open-source technologies](/blog/open-source-b2b-advantage). Our stack is SvelteKit, Vite, UnoCSS, and a long list of other open tools, and we think the reasons apply to analytics with particular force. Analytics is exactly the kind of software you should be able to audit. Plausible's entire codebase is [public on GitHub](https://github.com/plausible/analytics) under the AGPL-3 license, so the claim "we do not collect personal data" is verifiable rather than a marketing promise. Google Analytics is a closed, proprietary product; you take the vendor's word for what it does. Open source also changes the exit math. If Plausible the company disappeared tomorrow, Plausible Community Edition can be self-hosted, and the data export is straightforward. Compare that with the forced migration from Universal Analytics to GA4, where Google retired a product on its own schedule and deleted years of historical data on its way out. We would rather pay a modest subscription to an independent company whose code we can read than be the product in a free tool we cannot. ## What We Gave Up The honest accounting: **Plausible costs money.** GA is free at the point of use. For the traffic profiles of our sites and most client sites, the subscription is small, and we consider it the fair price of not paying with visitor data.**Fewer knobs.** GA4 offers deep segmentation, user-level exploration, and tight integration with Google Ads. Plausible is deliberately simpler. For teams running large paid-acquisition programs inside Google's ecosystem, GA still has a case. We[do not sell digital advertising](/blog/no-digital-advertising), and most of our clients' sites are content and lead-generation driven, so the trade cost us little.**A learning moment.** Some client stakeholders were attached to GA's interface. In practice, the one-page Plausible dashboard has been easier to hand to a marketing team than GA4 ever was. The most common reaction to the switch has been relief. ## Where That Leaves Us Our default for every new build is now Plausible, loaded as a tiny script on a fast page, with no consent banner required for analytics and a privacy policy a person can actually read. A handful of client sites remain on GA where advertising integration genuinely earns its keep, and we support those thoughtfully. Analytics should answer the question "is our website working?" It should not require surveilling the people the website is for. Plausible answers the question, and it lets us keep a promise we like making to our clients and their visitors. If you are weighing a similar move, we are happy to share the migration playbook: what to export from GA before you lose it, how to map your old reports onto Plausible's model, and what to tell your legal team.