Google Analytics alternatives for a small product

Google Analytics 4 is free, event-based, and shows sessions, pages, events and conversions after the click. Nothing about it is broken. The complaints are consistent and they are about fit.

• Time to an answer. "Which landing pages brought signups last month" is a report you build rather than a page you open.
• The event model. Everything is an event with parameters, which is powerful and unfamiliar if you arrived from the old session-and-pageview way of thinking.
• Consent and blockers. A banner people decline and a blocker in the browser both remove visits from the count, so the totals are already partial.
• Sampling and thresholds on smaller reports, which make a small site's numbers feel unreliable.
• Data ownership and hosting, which some teams need to answer for in a security review.

• What you count: page views and sources, or product events and the people behind them.
• How fast you get an answer: one dashboard, or a report you assemble.
• Where the data lives: a vendor's cloud, a region you choose, or your own servers.
• What it costs: free, a subscription that grows with traffic, or hosting you maintain.
• What it must join to: your signup event, your billing, your Search Console figures.

1. 1

   You want one simple dashboard: Plausible or Fathom

   Both are lightweight, privacy-focused web analytics: visitors, sources, top pages, countries, on one screen you can read in a minute. They answer "what is happening on the site" without a training course.

2. 2

   You want to follow a person from landing page to signup: PostHog

   PostHog is product analytics with a web analytics dashboard on top. It records events and the people behind them, so a funnel from a search landing page to a signup, broken down by the first referring domain, is a few clicks rather than a project.

3. 3

   You need the data on your own servers: Matomo

   Matomo is open source and can be self-hosted or used as a hosted service, with a feature set close to the traditional analytics model. It is the usual answer when a security review or a policy requires the data to stay somewhere specific.

4. 4

   You need the Google ecosystem: stay on GA4

   If you run Google Ads, need the Search Console link inside analytics, or report into Looker Studio with Google's own connectors, leaving costs you more than it saves.

For a product whose next question is always "did this search click become a customer", product analytics is the category that answers it, and simple web analytics is the category that will not.

• Free at any scale. GA4 costs nothing however much traffic you have; most alternatives price by events or visits.
• The Google connections: the Search Console link inside analytics, Google Ads reporting, and the first-party Looker Studio connector.
• Years of documentation and answers. Every GA4 question has been asked publicly by someone already.
• Your history. Analytics data does not migrate; the old account keeps the past and the new tool starts empty.
• In the simpler tools, the depth: no person-level analysis, so "which landing pages produce customers" stays out of reach.

The practical move is to run both for a month. Install the new tool, keep GA4, and compare one number you care about. If the two disagree by a lot, you have learned something about consent and blockers, not about the tools.

Every tool on this page measures what happened after the click. None of them can see the search.

• Search Console is pre-click: the queries, the impressions, the position, the clicks Google recorded. No analytics tool has that data, and no analytics tool ever will.
• Analytics is post-click: the session, the pages, the events, the signup.
• They never reconcile exactly, because of consent, blockers, redirects and the different moment each one counts.
• They meet on one key: the landing page. That is the join that answers which searches bring people who sign up.

So the honest stack for a small product is two tools, not one: Search Console for the search, and whichever analytics tool answers your product question fastest.

1. 1

   Write the three questions first

   For example: how many people arrived from search last month, which landing pages produced signups, and what did people do before cancelling. Choose the tool that answers them fastest.

2. 2

   Install alongside, do not cut over

   Run the new tool beside GA4 for a month. Two scripts for a month is a fair price for a comparison you can trust.

3. 3

   Mark the signup event on day one

   An analytics tool without your key event is a page-view counter. Send the event, and record the landing page with it.

4. 4

   Keep the old data readable

   Export what you will want later before you stop paying attention to the old account. History does not migrate.

5. 5

   Re-point the reporting

   If a dashboard read from GA4, decide what replaces that connection before you turn the old tag off.