Simple Analytics vs Google Analytics

Notice the third row. The two tools don't mean the same thing by "visitor", and the next section shows what that does to your totals.

Simple Analytics doesn't count people. Its docs say the number labelled visitors is unique page views. A page view counts when it has no referrer, or when the referrer is another site. A click to a second page, a reload or the back button doesn't count again (Simple Analytics).

There are no cookies, no stored IDs and no IP addresses to link one arrival to the next (what it collects). Here's one reader's day, counted by each tool's rules.

Notice the last row. By the next morning, one reader is 4 Simple Analytics visitors, 7 page views and 1 GA4 user.

You can see the rule at work on Simple Analytics' own site. Its public stats for 13 September to 11 October 2026 count 17,181 visitors from 22,002 page views (public JSON). So 78% of page views were arrivals, and each arrival read 1.3 pages on average. On a site where people read more pages per arrival, the two numbers sit further apart.

Two more rules move the totals. In a single-page app, every navigation after the first load is a repeat page view (docs). And visits from browsers with Do Not Track switched on aren't collected at all, unless you add data-collect-dnt="true" to the script tag (docs). The smaller light.js has no such option (docs). Those visits are missing from every report, page views included. Fathom, for comparison, counts them unless you choose to honour the setting (Fathom).

Countries come from the browser's time zone, not the IP address (docs). GA4 gets them from the IP address (Google), so country splits won't match either.

Our read: Simple Analytics' visitors and GA4's users measure different things, so don't hold one up against the other. Compare page views with page views, which is Simple Analytics' own advice (Simple Analytics).

We installed each tool alone on a test page and opened it in Chrome four ways: as a person who moves the mouse and scrolls, and as three kinds of bot. One announces it's automated. One hides that. One clicks without moving the mouse, the way AI agents click. No data reached Simple Analytics or Google: apart from fetching each script, our lab answered every request itself.

Notice the first bot row. Simple Analytics was the only tool in our lab whose request said the browser was automated. Its script checks navigator.webdriver, six older signs of automation tools, and user agents containing bot, spider or crawl, and adds bot=true to the page view (script). That caught Chrome as shipped. Its docs say robots are blocked by user agent and these browser traces (robot blocking). The disguised browser and the agent-style click carried none of those signs, so they went out as ordinary visits.

Checks like these can catch people too. In April 2025 a user found that the Phantom crypto wallet, which adds a window.phantom property, made their visits count as a robot. The issue was closed as fixed in May 2025. But the fix sits in a pull request that's still open, and the script served on 13 October 2026 still checks window.phantom (issue, pull request).

Our read: marking the page view is a sensible first check, and one GA4's tag doesn't make. But it only catches browsers that give themselves away.

The cookie row. GA4 left _ga and _ga_<property>, each for 400 days, which is Chrome's limit (Google sets two years). Simple Analytics and Nora left nothing, in cookies or in storage.

Ad blockers. EasyPrivacy blocks simpleanalyticscdn.com, so the script never loads for those readers, just as with GA4. Simple Analytics documents two ways around it, a custom subdomain and a proxy (docs). But in the pricing page's own plan data, neither is switched on for the Free or the Simple plan. In April 2026 the bypass came with the $40 Team plan, which has since gone, and the Enterprise offer mentions help with proxies and custom domains (April 2026, pricing). Ask before you count on it. Nora isn't on it.

Weight. The 3.8 KB we measured is latest.js, as downloaded. Its light.js is smaller, 1.9 KB by its makers' table, and leaves out time on page, scroll depth and single-page app support (docs).

GA4 is free at any size, with 30 key events and funnels; what it costs you is cookies that need consent in Europe. Nora is free up to 1 million visitors a month without a card, with goal pages, a weekly email, CSV export and report links.

Simple Analytics' free plan is free forever: 1 user, 5 websites and the latest 30 days of data, with anything older deleted. You have to show a Simple Analytics badge, and page views are unlimited under a fair-use policy (pricing). The plan data behind the page shows what else is off.

Notice the "Off" rows. Goals, events, the Stats API and the data-centre bot filter all need the paid plan, and the ad-blocker bypass and AI chat are off on both.

Our read: the free plan suits a small site that only needs the last month. Once you want goals, events or more than 30 days of history, you're on the paid plan.

The paid Simple plan rises with volume. Simple Analytics counts page views and events together, which it calls datapoints (docs). It moves you between steps on its own, first from last month and then from the average of the previous three (pricing).

What changed in 2026. In mid-April the page still showed Simple from $15 and a Team plan from $40, both at 20,000 page views. Team added a second user, 20 websites, 5 years of history, custom views, roles, the Export API and the ad-blocker bypass (April 2026). By mid-June the Team plan was gone, and the slider started at 100,000 page views and $20 (June 2026). Many comparison pages still quote the older plans.

Paid plans start with a 14-day trial without a card. Enterprise is priced on a call and comes with a free 30-day proof of concept run beside GA4 (pricing). You can pay in euros, dollars or pounds.

Run both and the totals won't line up. Here's what pulls them apart.

• Visitors. Arrivals in one, cookies in the other, as above. Compare page views instead.
• Do Not Track. Simple Analytics leaves those visits out unless you opt in.
• Consent. In Europe, GA4 counts only visitors who accept its cookies. In one study, GA missed 55.6% of visits when a banner was shown and 15.8% when none was (Orbit Media). Google estimates the missing visitors only for sites with at least 1,000 declined events and 1,000 consenting users a day (Google).
• Simple Analytics' own test. For one company's LinkedIn campaign, scaled to 1,000 clicks, Simple Analytics recorded 640 page loads and GA4 270 sessions. Simple Analytics itself warns that these are different metrics, not a count of visits lost to consent (Simple Analytics).
• Countries. Time zone in one, IP address in the other (Google).
• Ad blockers. EasyPrivacy blocks both scripts.
• Bots. Each drops different bots. GA4 never says how many, and Simple Analytics' robot-blocking page mentions no count. On the first site running Nora, 3.3% of the page views its script saw over nine days in October 2026 were bots or AI agents.

Simple Analytics blocks robots by user agent and by traces they leave in the browser. On the Simple plan, an advanced filter you switch on in account settings checks a list of data-centre IP addresses from Udger, Google Cloud, AWS and Azure before a page view is stored, and tags matches as robots (Simple Analytics). The Free plan doesn't have it (plan data).

It has no AI-assistant channel and nothing that tells AI agents from people; we found neither in its changelog or feature pages (changelog, features). On its own public stats, chatgpt.com, gemini.google.com and claude.ai sit among ordinary referrers: 42 of 17,181 visitors from 13 September to 11 October 2026, or 0.24% (public JSON). The JSON lists only the top 50 referrers, so smaller assistants may be missing.

Simple Analytics does have an AI of its own, a chat about your stats launched in December 2023 (AI chat, changelog). But the plan data has it switched off on both the Free and the Simple plan.

Google Analytics 4 drops known bots using Google's research and the IAB list. The filter is always on and shows no count (Google). Since 13 May 2026, an AI Assistant channel groups visits referred by ChatGPT, Gemini, Claude and others, by referrer (Google). Visits from agentic browsers such as ChatGPT Atlas and Perplexity Comet show up as ordinary Chrome, often as Direct (MarTech).

Nora keeps every page view and labels it person, AI agent or bot, with the test that caught it (a signature check, the name it gives, how it behaves, machine signs), and groups AI assistants as a source. In our lab it labelled the headless browser a bot and the agent-style click an AI agent. The disguised browser that never touched the page passed as a person, as it did everywhere.

None of the three sees crawlers such as GPTBot or ClaudeBot, which fetch pages without running scripts. That takes your server or CDN logs.

Where GA4 is ahead

• Users and sessions. Users followed across days, and sessions within a visit.
• Funnels in Explorations. Simple Analytics renamed its funnels to goals in July 2024, because the steps aren't tied to one visit (changelog).
• Revenue, Search Console and Google Ads. None of them appears on Simple Analytics' feature or integration pages (features, integrations).
• Raw data. Events in BigQuery, free up to 1 million a day (Google).
• An AI Assistant channel for traffic from chatbots.

Where Simple Analytics is ahead

• Nothing kept on the visitor. No cookies, nothing stored in the browser, and the IP address dropped from every request (docs).
• EU-based subprocessors only (list).
• A free plan that needs no consent banner.
• A bot flag in the script. It marks automated browsers before the page view leaves.
• Nothing hidden. No rows held back by thresholds, and 3 years of history on the Simple plan. GA4 keeps the event-level data behind explorations 2 or 14 months (Google) and hides small rows (thresholds).
• Events with metadata on paid plans (docs), and open-source tracking scripts (GitHub).

Features from each maker's pages on 13 October 2026. Simple Analytics' goals and events need a paid plan (goals).

We read reviews and threads about both. The same few things come up again and again. Here they are, in our words, with links to where people said them.

1. Run both for two to four weeks. The trial is 14 days without a card. Compare Monday with Monday.
2. Compare page views with page views. GA4's users and Simple Analytics' visitors will never line up.
3. Decide on Do Not Track. Add data-collect-dnt="true" if you want those visits counted.
4. Import your GA4 history. Only page_view events come across, by the hour, with page, source, device, country, browser and system. The import stops at your first Simple Analytics page view, so nothing is counted twice (import guide).
5. Rebuild goals and events on a paid plan; the Free plan has them switched off.
6. Plan for ad blockers. The bypass isn't on the self-serve plans, so ask about Enterprise or accept the loss.
7. Remove GA4, then check whether your cookie banner still has anything to ask about.

Nora is for you if you like Simple Analytics' privacy, want goals without paying, and want to know how much of your traffic is people. It's free up to 1 million visitors a month, sets no cookies, weighs 2.7 KB, and labels every page view person, AI agent or bot, with the test that caught it. AI assistants show as a source of their own. You also get live visitors, pages, sources, countries, goal pages, a weekly email, CSV export and report links.

It gives up things Simple Analytics has. Your data is in the US, while Simple Analytics keeps everything with EU-based providers. It can't import your GA history. Events with properties and agency features wait for Pro, which is coming. It launches on 1 November 2026, so it has no track record; Simple Analytics has run since 2018. If EU hosting matters most, Simple Analytics is the better pick.

Nora Lab, the research team at Nora Analytics, ran every test on this page the same way for each tool. Here is exactly how, so you can judge the results yourself.

The full method, with what each visitor stands for, is on How we test. Every result from the lab run is public as JSON.