I have a lot of complaints about GA4. The UI is hostile. The default reports hide more than they show. Data thresholding swallows information you actually need. Anyone who tells you they "love GA4" is either lying or has never used Universal Analytics.
And yet. After years of pushing clients to evaluate alternatives — Plausible, Matomo, Mixpanel, Adobe, the whole list — I keep landing in the same place for performance marketing accounts: GA4 wins. Not because it's the best analytics tool. Because it's the best analytics tool for performance marketers. Those are different things.
This post is the honest comparison I wish someone had handed me three years ago. What the alternatives are actually good at, what GA4 still does better than all of them, and when you should genuinely switch.
The TL;DR: Privacy-first tools (Plausible, Fathom) are better for content sites. Product analytics tools (Mixpanel, Amplitude) are better for SaaS. Open-source tools (Matomo, PostHog) win on data ownership. But if your job is to spend money on Google Ads and report ROAS, the integration depth GA4 has with Google Ads is a moat nothing else can cross.
Why people are leaving GA4 (and why some of it is fair)
Let's get this out of the way. The complaints are real:
- The UI is genuinely bad. Reports take three clicks to find. Filters reset. Exploration interface is its own learning curve.
- Data thresholding hides rows when sample sizes are small. Useful for privacy. Painful for analysis.
- 14-month default data retention on the free tier. You lose year-on-year comparisons unless you export to BigQuery.
- Schrems II / GDPR concerns in the EU. Several countries' DPAs have ruled Google Analytics non-compliant under specific configurations.
- Slow data processing. Up to 48 hours for full data to appear.
- Modelled data mixed in with observed data, without clear labels.
- Custom dimensions are limited on the free tier (50 event-scoped, 25 user-scoped).
- It's still Google collecting the data.
None of those are made up. They're all legitimate reasons to look at alternatives. The question is whether the alternatives solve your actual problem.
The main GA4 alternatives, grouped by what they're actually for
Privacy-first, lightweight analytics
The headline category. Tools built around "cookieless, GDPR-safe, simple."
- Plausible — Open-source option, EU-hosted, beautifully simple dashboard. ~$9-$19/month for small sites.
- Fathom — Cookieless, privacy-compliant. $15/month entry tier.
- Simple Analytics — Similar pitch. No cookies, no personal data.
- Cloudflare Web Analytics — Free, server-side, very basic. Bundled if you're already on Cloudflare.
What they're good at: Clean traffic data, page-level performance, referrer tracking, no cookie banner needed in most jurisdictions, fast page load impact.
What they're not good at: E-commerce tracking, conversion-value attribution, Google Ads integration, audience sync, complex funnels.
Open-source / self-hosted analytics
For teams that want full data ownership.
- Matomo — The longest-standing GA alternative. Can be self-hosted free or cloud-hosted (paid). Feature-rich, includes heatmaps and session recordings on higher tiers.
- PostHog — Open-source product analytics. Strong on feature flags, A/B testing, session replays. Can be self-hosted.
- Umami — Lightweight, open-source, similar pitch to Plausible. Simpler.
- Snowplow — Enterprise-grade event collection. Raw data into your warehouse. Powerful, not for beginners.
What they're good at: Data sovereignty, no per-event pricing, customisation, integration with your own data warehouse.
What they're not good at: Out-of-the-box Google Ads integration, plug-and-play setup, ongoing maintenance cost (self-hosted requires DevOps).
Product analytics tools
Built for software products, not marketing sites.
- Mixpanel — Event-based, strong on funnels, retention, cohorts.
- Amplitude — Similar. Excellent for user behaviour analysis. Free tier exists but limited.
- Heap — Auto-captures every interaction. Less setup, more storage cost.
- PostHog — Mentioned again because it straddles both worlds.
What they're good at: User journey analysis inside a logged-in product, behavioural segmentation, retention curves, feature usage.
What they're not good at: Marketing channel attribution, e-commerce GA4 standard reports, Google Ads conversion import.
Enterprise analytics
- Adobe Analytics — Powerful, expensive, painful to implement. Best when you're already inside the Adobe Experience Cloud.
- Piwik PRO — Privacy-compliant enterprise analytics. EU-hosted. Mid-market positioning.
What they're good at: Customisation, governance, enterprise integrations.
What they're not good at: Speed of implementation, cost, simplicity, Google Ads integration depth.
Heatmap and session tools (not really GA4 replacements)
People sometimes lump these in. They shouldn't:
- Microsoft Clarity — Free heatmaps and session recordings.
- Hotjar — Same category, paid.
- FullStory — Premium tier.
These are complements to your analytics, not replacements. They tell you how people interact with a page, not which channel sent them.
Why GA4 still wins for performance marketers specifically
Now the actual point.
If your job is to spend money on Google Ads (or YouTube, or Demand Gen, or Performance Max) and prove ROAS, there are six things GA4 does that nothing else does as well.
1. Conversion import into Google Ads is one-click
In GA4, you mark an event as a key event. In Google Ads, you import it as a conversion. Done. Smart Bidding can now use it.
With every other tool, you're either:
- Sending conversions to Google Ads via the offline conversion upload (manual, slow, lossy).
- Building a custom Measurement Protocol bridge.
- Living with the fact that Google Ads is blind to your real conversions.
That single integration is worth more than every UI complaint about GA4 combined.
2. Audience sync for remarketing and Smart Bidding
GA4 audiences flow directly into Google Ads as remarketing lists. So do predictive audiences (likely to purchase, likely to churn). Other tools either can't do this at all or require a CDP in the middle.
If you're running Performance Max or value-based bidding, this signal matters. A lot.
3. Google Signals = cross-device attribution
Google Signals stitches sessions across devices using signed-in Google users. No other analytics tool has access to this data. For a buyer who researches on mobile and purchases on desktop (i.e., most buyers), GA4 will credit the right channel where alternatives can't.
4. Free BigQuery export
GA4 exports raw event data to BigQuery for free. That's a five-figure annual feature in enterprise analytics tools, given away.
It means:
- You own the data forever, not just 14 months.
- You can run your own SQL on every event.
- You can blend Ads data, CRM data, and analytics data in one warehouse.
- You can build attribution models that aren't GA4's defaults.
This single feature undercuts most of the "but I don't own my data" argument for switching.
5. Data-driven attribution (DDA) is now the default
GA4's DDA model uses machine learning to assign conversion credit across touchpoints. It's not perfect, but it's significantly better than last-click, and it's free.
Building equivalent attribution outside GA4 requires either an enterprise tool (six figures) or a data team with months of work.
6. The ecosystem moat
This one is underrated. Because GA4 is so widely used:
- Every freelancer, agency, and contractor knows it.
- Every CMS, e-commerce platform, and marketing tool has a native GA4 integration.
- Every tutorial on YouTube, every Stack Overflow thread, every "how do I..." search points at GA4.
- Hiring an analyst who knows your tool is easy.
Switch to Matomo or Mixpanel and that ecosystem shrinks dramatically. Not unmanageable, but every problem becomes harder to solve.
When you should genuinely switch
Performance marketers shouldn't switch lightly. But there are honest reasons to:
- You're a content site with no ads, no e-commerce, no Google Ads spend. Plausible or Fathom will give you cleaner numbers, faster load times, and no cookie banner.
- You're an EU government / public-sector entity where Google Analytics has been ruled non-compliant by your DPA. Matomo (self-hosted) or Piwik PRO is the safer call.
- You're a SaaS product where the question is "what features are users using?" not "what channel converted them?" Mixpanel, Amplitude, or PostHog will run circles around GA4 here.
- You're a Fortune 500 with Adobe Experience Cloud already in place. Adobe Analytics integrates with everything else you own.
- You need 100% data ownership for regulatory or strategic reasons. Self-hosted Matomo or Snowplow is the answer.
If none of these describe you and you spend money on Google Ads — stay on GA4.
The hybrid approach most pros actually use
The dirty secret: most experienced marketers don't run one analytics tool. They run a stack:
- GA4 as the source of truth for Google Ads attribution and Smart Bidding inputs.
- A privacy-friendly tool (Plausible, Fathom) for public-facing dashboards and quick traffic checks without a cookie banner.
- Microsoft Clarity (free) for heatmaps and session recordings.
- BigQuery + Looker Studio for blended reporting that mixes GA4, CRM, and ad spend.
- A product analytics tool (Mixpanel, PostHog) for in-app behavioural analysis if there's a product side.
This isn't bloat. Each tool answers a question the others can't. Trying to make GA4 do all of them is what creates the "GA4 is bad" feeling. Trying to make Plausible do all of them is worse.
Use case: a multi-brand events business reconsidering its stack
A composite based on patterns from multiple performance marketing teams I've consulted with.
A mid-sized events business with three brands and ~£200k/month in Google Ads spend was tempted to migrate from GA4 to Matomo. The pitch was data ownership and "no more GA4 headaches." Their team had spent 18 months struggling with GA4's reports.
We mapped what they'd actually lose:
- Google Ads conversion import would break. They'd need to migrate to offline conversion uploads or rebuild via Measurement Protocol.
- Smart Bidding signal quality would degrade. Cancellation tracking, value adjustments, predictive audiences — all needed to be rebuilt outside GA4.
- The agency they worked with had two GA4 specialists. Zero Matomo specialists. Onboarding new people would cost 3-6 months.
- Their attribution reporting in BigQuery was already running. Migration meant rebuilding from scratch.
We ran the numbers. Migration cost was ~£40k in dev time. Annual savings (Matomo cloud hosting at their volume vs free GA4): roughly £0. The "win" was data ownership and cleaner privacy posture, both real, both not worth £40k for them.
Instead, we did three things:
- Added Plausible for their public marketing dashboards (the ones execs looked at). Cleaner, faster.
- Tightened GA4 consent mode v2 implementation so the privacy concern was largely addressed.
- Built BigQuery + Looker Studio dashboards for the analyses GA4's UI couldn't surface cleanly.
Six months later, the same team that was about to migrate said GA4 was "actually fine now." Nothing changed about GA4. Their relationship with it changed.
Common mistakes when evaluating GA4 alternatives
- Comparing GA4's worst UX moments against an alternative's marketing site. Every tool looks great on its homepage. Try a 30-day trial on real data before deciding.
- Forgetting the integration tax. "Free" tools become expensive when you have to rebuild Google Ads attribution.
- Mistaking page views for performance marketing data. Plausible is amazing for blog traffic. It cannot tell you that Campaign A drove £18k in profitable revenue while Campaign B drove £40k in cancellations.
- Switching without exporting historical data. If you leave GA4, export the BigQuery data first. You won't get it back.
- Assuming the privacy compliance gap is just GA4. Most "private" tools still need a cookie banner if you load them client-side and tie them to a logged-in user. Read the legal fine print.
Bottom line
GA4 is not a great analytics tool. It's a great Google Ads attribution tool that also happens to do analytics, sometimes well, often badly.
If you're a performance marketer:
- Stay on GA4 for attribution, conversion import, and Smart Bidding signal.
- Use BigQuery export to fix the data ownership and retention problems.
- Add a lightweight tool for daily traffic checks and exec dashboards if the GA4 UI is grinding your team down.
- Use Microsoft Clarity alongside it for heatmaps. It's free and complementary.
- Only switch fully if you genuinely don't run Google Ads or you have a regulatory reason.
The alternatives aren't bad. They're just optimised for a different job. For the specific job of "spend money on Google Ads, prove ROAS, feed Smart Bidding," GA4 still wins. Probably for a few more years yet.
Sources and further reading: