Plausible runs on 3,257 Estonian sites, against 26 in Latvia and 8 in Lithuania
Plausible runs on 4.3% of live Estonian sites and on 26 Latvian and 8 Lithuanian ones, while Google Analytics runs on a fifth of each. What the split means.
Most owners never choose an analytics tool. A developer pasted a snippet years ago, and it is still there. The Baltic Web Audit, our near-census of the Latvian, Estonian and Lithuanian web in July and August 2026, looked at which snippets that is, and found one clear exception to the pattern: in Estonia, a different tool is running on a meaningful slice of the web.
In short
Plausible, a privacy-focused analytics tool whose company is based in Tartu, Estonia, was detected on 3,257 Estonian sites, 4.3% of the living .ee web, against 26 Latvian sites and 8 Lithuanian. Google Analytics runs on about a fifth of each countryâs living web: 20.9% in Latvia, 20.0% in Estonia and 21.4% in Lithuania. So Estonia did not leave Google. It has a visible minority using a local, cookieless tool, and its two neighbours have almost none. For an owner the useful question is not which country you resemble but what your numbers are for, because the answer decides whether you need a consent banner at all.
What the audit found
Every figure here is from the Baltic Web Auditâs chapter on Google. Counts are detections, so each is a minimum.
| Latvia | Estonia | Lithuania | |
|---|---|---|---|
| Google Analytics | 10,282 sites (20.9%) | 15,213 (20.0%) | 20,339 (21.4%) |
| Google Tag Manager | 5,276 | 8,149 | 10,530 |
| Plausible | 26 | 3,257 (4.3%) | 8 |
Matomo, the self-hosted alternative, is in single digits everywhere. The auditâs own reading is that Plausible is âthe single clearest case in the study of a country preferring its own softwareâ, and it reports the same pattern in Lithuania for e-mail marketing, where Omnisend and MailerLite, both founded there, are far more common than in the other two countries.
What the data cannot show is why. A detection says a script is on a page, not what the owner was thinking when it went there. The report reads it as a country choosing its own software. The data fits that, but it fits other explanations too, so we leave the motive alone. The more practical question is what the choice changes. We covered the wider Estonian picture in the state of the Estonian web.
What Plausible says it does
Plausibleâs company page gives the data controller as Plausible Insights OĂ in Tartu, Estonia (its privacy page). Its data policy makes four claims worth knowing:
- No cookies and no local storage. âOur script stores nothing on the visitorâs device. It sets no cookies and writes nothing to local storage or the browser cache.â
- A daily identifier instead of a persistent one. Visitors are counted using a hash of a daily salt, the websiteâs domain, the IP address and the browserâs User-Agent. The salt is rotated, and the policy says raw IP addresses and User-Agent strings are never stored, so a visitor cannot be recognised on another day or on another site.
- EU processing. Data is âprocessed and stored in the EU on infrastructure owned by European companiesâ.
- No banner needed. Its about page lists âNo cookie banner requiredâ and says the code is open source under the AGPLv3, so what it collects can be inspected.
All four are the vendor describing itself, checked on 12 November 2026. Open source is what makes the claims testable: anyone can read the code, or look at what the script sends in a browserâs network panel.
What Google says Google Analytics does
Googleâs own consent mode documentation lists analytics_storage as the setting that âenables storage, such as cookies (web) or device identifiers (apps), related to analyticsâ. It describes two ways of running behind a banner:
- Basic consent mode. Google tags are blocked until the visitor answers the banner, and âno data is sent before a user consentsâ.
- Advanced consent mode. Tags load at once with consent denied by default. While it stays denied, they send âmeasurements without cookiesâ, and only after consent are cookies written and full data sent.
So the trade is real on both sides. Behind a banner, Google Analytics sees fully the visitors who accept and, depending on the mode, nothing or a reduced signal from the rest. Plausibleâs comparison page says the same thing from its side, listing Google Analytics 4 data as modelled or estimated under consent mode. That is a vendorâs comparison and worth reading as one, but the underlying mechanism is in Googleâs own text.
When each one fits
- Plausible fits when the figures are pageviews, sources, top pages, and whether a campaign or an article moved anything. It counts everyone who loads the page and, by its own account, needs no banner of its own.
- Google Analytics fits when the figures feed Google Ads. Googleâs consent types include
ad_user_data, which âsets consent for sending user data to Google for online advertising purposesâ, so that link runs through Googleâs tags and Googleâs consent settings, not through a cookieless tool. - Both can be on one site, at the price of the banner. The banner then has to block the Google tag until consent, and it has to be tested to do so.
One thing holds either way: the tool on the page must match what the banner promises. In the audit, 91.5% of rendered Latvian shops, 91.1% of Estonian and 91.8% of Lithuanian set cookies before the visitor consents, and replacing the analytics tag does not fix that if an advertising pixel or a chat widget is still setting them.
What you can check yourself
- See which tool you run. Open your homepage, choose View page source and search for
googletagmanager.com,google-analytics.comandplausible.io. Whatever you find is what is measuring you today. - See what it sets. In the browserâs developer tools, open the Application tab, then Cookies, load the page in a private window without touching the banner, and look at what is already there. Any cookie from an analytics or advertising tool that appears before you have answered is that tool running early.
- Decide what the numbers are for. If nobody has made a decision from them in the last year, the tool is not the problem.
- Export your history before changing anything. A new tool starts from zero, and year-on-year comparisons are what you lose.
Changing the tool is a small job. Getting the banner, the tag manager and the tracking to agree afterwards is the part that needs a developer, and it is the part we are asked about. That is covered by our web development work.
Why we can say any of this
The figures come from a near-census rather than a sample. We measured 293,806 of the 306,872 Baltic domains in our frame across July and August 2026, from one fixed location, with a crawler that identified itself honestly on every request. Everything is published as an aggregate and no individual site is named beside a weakness.
If you want to know what your own site loads and sets before a visitor says yes, send us the address and we will tell you what we see.