Stop using Google Analytics on your indie project

If you've ever opened Google Analytics 4, stared at the dashboard for ten minutes, and closed the tab without learning anything useful — you're not bad at analytics. You're using the wrong tool.

GA4 was rebuilt from the ground up for enterprise marketing teams running multi-channel attribution models across paid media budgets in the millions. The event-based data model, the exploration reports, the conversion paths, the predictive audiences — none of it was designed for someone who just wants to know whether people are reading the blog post they wrote last week.

Yet indie founders reach for GA4 by default because it's free and because they've heard of it. Then they spend three hours reading documentation to understand why their pageview count doesn't match what Cloudflare shows, give up, and make product decisions based on gut feeling anyway.

This post is about the gap between what Google Analytics actually is and what indie founders actually need — and what to do about it. I'll cover the specific ways GA4 fails you, the privacy and legal reality that most founders ignore, and a fair review of the alternatives before you decide what to use.

I have a stake in this topic, which I'll be transparent about at the end. The alternatives section covers tools I compete with.

What GA4 was actually built for

Google Analytics 4 replaced Universal Analytics in July 2023. The migration was so disruptive — breaking every existing report, changing fundamental concepts like sessions and bounce rate — that thousands of businesses delayed as long as possible and many didn't finish before the deadline.

That disruption wasn't accidental. Google was following money that flows from its biggest customers.

Google Analytics 360 — the enterprise version of GA4 — costs around $150,000 per year. The organizations paying that price need things that you do not need:

• Cross-device tracking that stitches together sessions across Google's signed-in ecosystem
• Attribution modeling that connects YouTube ads, Display ads, and Search ads to downstream conversions
• Raw BigQuery export for data science teams running SQL on event streams
• Predictive audiences ("users likely to churn in 7 days") for remarketing campaigns
• Consent mode v2 integration with Google's ad measurement infrastructure
• Server-side tagging for enterprise GTM setups

These are legitimate enterprise needs. A retail brand spending $10M/year on Google Ads needs to know which ad creative drove which purchase across which device. The GA4 data model — where everything is an event, sessions are reconstructed after the fact, and users are stitched together across devices using Google's identity graph — exists to serve that use case.

That use case is not your use case.

When you install GA4 on your indie project, you're deploying infrastructure built for a team of analysts with six-figure tool budgets, pointing it at a product with a few hundred daily active users, then wondering why it feels like overengineering.

Because it is.

What founders actually need from analytics

Before getting into what's wrong with GA4 specifically, it's worth being honest about what a solo founder or small indie team actually checks on a regular basis.

I've asked this in a few founder communities and the list is remarkably consistent:

• How many people visited my site today? This week?
• Which pages are getting traffic?
• Where is traffic coming from? (search, Twitter, direct, referral)
• Is this week better or worse than last week?
• Did that Product Hunt launch actually drive signups?
• Is the blog post bringing in search traffic yet?
• Where do people drop off before signing up?

That's it. That's the entire list for most indie projects at the stage where analytics actually matters.

You don't need attribution modeling. You don't need predictive churn scores. You don't need a BigQuery export. You need about 40 data points, updated daily, that you can read in under 60 seconds before deciding what to build next.

The problem isn't that founders need less data — it's that GA4 optimizes for a completely different set of questions that cost millions of dollars to need to answer.

The specific ways GA4 fails indie founders

The interface is optimized for analysts, not decisions

Open a fresh GA4 property and you're presented with a realtime overview, a reports snapshot that requires configuration, and a left sidebar with nine categories: Reports, Explore, Advertising, Configure, Admin — plus sub-items inside each.

The information you actually want — pageviews, top pages, referrers — is buried. The default "Acquisition overview" report shows Sessions (not pageviews), New users (vs. users — two different definitions of the same concept), and Engaged sessions (a metric that didn't exist before GA4 and that most founders couldn't define without Googling).

Finding your top pages requires going to Reports → Engagement → Pages and screens. Finding your referrers requires Reports → Acquisition → Traffic acquisition, then segmenting by session source. These are not hard for an analyst who uses the tool daily. They're friction for a founder checking in once a week.

Every minute you spend navigating a tool is a minute you're not spending on the product. This matters more than it sounds at early stage, when your time is the most constrained resource you have.

Pageview numbers are unreliable

GA4 uses data sampling in its reports once you exceed a certain event volume — the threshold varies but is lower than most founders expect. When sampling is active, GA4 estimates your metrics rather than counting them. The estimate can be directionally correct but numerically wrong by 10–30%.

For a paid analytics tool, this would be unacceptable. For GA4, it's documented behavior.

There's also a structural reason GA4's pageview numbers rarely match other sources. GA4 tracks sessions through JavaScript that fires after the page loads. Any visitor who leaves before the script executes — including many bot visits, but also fast human exits — doesn't register. Cloudflare and your server logs count raw HTTP requests. The methodologies are fundamentally different, and there's no universally "correct" answer, but GA4's number isn't the authoritative one.

GA4 tracks for Google, not just for you

When you install GA4, you're installing Google's data collection infrastructure on your users' browsers. Google's privacy policy makes clear that data collected through their products may be used "to improve Google's products and services." Specifically, GA4 can use behavioral signals to improve Google's advertising models.

You are not the customer of Google Analytics. You are the distribution channel. Your users are the data source.

This isn't a conspiracy theory — it's how the business model works, and it's disclosed in the terms. But it means that when you install GA4 on your indie project, your users' behavioral data potentially contributes to Google's advertising intelligence. Most founders building privacy-conscious products haven't thought through this implication.

The GDPR situation is worse than you think

In March 2023, Austria's data protection authority ruled that Google Analytics violates GDPR by transferring personal data to the United States without adequate protection. Italy, France, Finland, and Denmark reached the same conclusion. The Irish Data Protection Commission later issued guidance that effectively makes standard GA4 use non-compliant for EU users without a significant consent infrastructure.

Google has since introduced features — consent mode v2, server-side tagging, data redaction — that attempt to address this. But implementing these correctly requires technical work most indie founders aren't set up for, and the regulatory picture is still evolving.

The practical consequence: if you have EU visitors and you're running standard GA4 without a properly implemented cookie consent banner and data processing agreement, you may be in violation. The fines for GDPR violations start at €10M or 2% of global annual turnover.

In practice, small indie projects are not the enforcement priority. But it's disingenuous to call GA4 "free" when you factor in the compliance work required to use it legally at any scale in Europe.

You own less of your data than you think

Google has deleted historical Universal Analytics data. If you relied on GA3 reports, they're gone — permanently. GA4 data has similar retention limits: by default, event data is only retained for two months, with a maximum of 14 months at the account level.

After 14 months, your historical data in GA4 is deleted. If you want long-term retention, you need to set up BigQuery export — which is free but requires technical setup and ongoing storage costs.

The privacy-first alternative ecosystem

In response to GDPR (2018), the Schrems II ruling (2020), and growing awareness of surveillance capitalism, a set of privacy-first analytics tools emerged that take a fundamentally different approach: collect only what you need, store no personal data, require no cookie consent, and give you full ownership of your data.

The core insight is that you can answer almost every question founders care about without collecting any personal data at all. You don't need to know who visited. You need to know how many visited, from where, and what they did.

The technical approach varies by tool, but the common thread is:

• No cookies — session data is never persisted to disk on the visitor's device
• No IP storage — IPs may be read to derive country, but are never written to a database
• No cross-site tracking — each site's data is completely isolated
• No fingerprinting — no browser attributes combined to create a pseudonymous ID

This means no cookie consent banner is required under GDPR, CCPA, or PECR, because no personal data is being collected in the first place.

The alternatives, reviewed fairly

Plausible Analytics

Plausible is the most widely known privacy-first alternative. Founded in 2019, bootstrapped, profitable, and open-sourced in 2021. The dashboard is clean and fast — pageviews, visitors, bounce rate, top pages, referrers, countries, and devices all on one screen. The script is lightweight (~1kb).

The main limitation: pricing starts at $9/month for 10,000 monthly pageviews. Most indie projects won't hit 10k pageviews in their first few months, so the floor price is essentially a product tax for validation-stage projects.

Plausible is also positioning toward marketing teams — their feature roadmap and copy reflect this. Self-hosting requires running your own server.

Best for: Established indie projects with consistent traffic who want maximum simplicity and are willing to pay a premium for it.

Fathom Analytics

Often compared directly to Plausible. Also bootstrapped, similarly priced. The UI is arguably even simpler — one screen, minimal to the point of spartan.

Fathom's differentiating feature is EU isolation: European visitor data never leaves EU infrastructure, making them the most defensible GDPR option. They've invested significantly in privacy legal compliance.

Pricing starts at $14/month for 100,000 monthly pageviews. Better value than Plausible at higher volumes, more expensive at the floor. No self-hosted option.

Best for: Projects with EU users where GDPR compliance is genuinely load-bearing.

Simple Analytics

Netherlands-based, leans hard into the "no personal data, ever" promise. Publishes infrastructure details, stores data only in the EU, and offers a public dashboard feature. Pricing starts at $9/month (annually) for 100,000 pageviews with no meaningful free tier.

Best for: Open-source projects or public products where traffic transparency is an asset.

Umami

Open-source and self-hostable. Deploy it on a $5 Railway or Render instance and you own everything: the data, the schema, the retention policy. The UI is straightforward. The real cost is operational — you manage the database, upgrades, and uptime.

Best for: Developers already running their own infrastructure who want full data ownership at zero ongoing cost.

Matomo

The established open-source alternative since 2007. Feature-comparable with GA4 — heatmaps, session recordings, A/B testing, a large plugin ecosystem. Both its strength and its weakness: you're back to a tool built for comprehensive marketing analytics. Cloud starts at €23/month for 50,000 hits.

Best for: Larger projects migrating from GA4 that need feature parity.

How to pick the right tool

Three questions determine the answer:

Do you have EU users and need airtight GDPR compliance? Fathom is the safest choice — their EU isolation and legal investment is the deepest in the market.

Do you want zero ops overhead? Rule out Umami and self-hosted Matomo. Pick from the cloud-hosted options. Your time is worth more than the $9/month delta between them.

Are you at validation stage with minimal traffic? The $9–14/month floor price of most cloud options is a flat tax for knowing whether your product exists to anyone. Look for a tool with a meaningful free tier, or use Umami self-hosted if you have the infrastructure.

What all of these options share: they will tell you everything you actually need to know, require no cookie consent banner, and won't share your users' behavioral data with Google's advertising infrastructure. Any of them is a meaningful upgrade over GA4 for an indie project.

What we built, and why I'm telling you this

I started working on this problem because none of the tools above fit exactly where I was. Plausible and Fathom are excellent products, but they're increasingly aimed at marketing teams and content sites. The pricing doesn't reflect the reality of early-stage indie projects. And the feature sets don't prioritize what product-focused founders actually care about.

So I built Statjot — a privacy-first analytics tool specifically for indie founders and solopreneurs. One JS snippet. No cookies. No config. No consent banner required.

I'm not going to claim it's better than Plausible or Fathom across every dimension. Both have been around longer, have more users, and have features Statjot doesn't have yet. What I can say is that it's built specifically for the use case this post describes.

If that's you, the free plan is a reasonable place to start. If it's not what you need after trying it, Plausible, Fathom, and Umami are all worth your time. The goal here is that you stop using GA4 on your indie project — not specifically that you use mine.

The short version

• GA4 was designed for enterprise marketing teams. The interface, the data model, and the feature set reflect this.
• The specific failures for founders: an interface built for analysts, unreliable pageview counts, data that may be used by Google's ad infrastructure, GDPR exposure most founders ignore, and 14-month data retention limits.
• Privacy-first alternatives — Plausible, Fathom, Simple Analytics, Umami, Matomo — answer every question you actually have without these downsides.
• The right pick depends on your traffic, your ops tolerance, and your GDPR exposure. Any of them is better than GA4 for this use case.
• If you're at validation stage with minimal traffic, look for a tool with a real free tier or self-host Umami.

The best analytics tool is the one you actually check. If you've been ignoring GA4 because it's too confusing to use — that's not a personal failure. That's a product-market fit failure on GA4's part. Something simpler will serve you better.