Best cookieless analytics 2026: 25 tools tested, brutally honest

Let's be real. The cookieless analytics market is a mess in 2026. Mixpanel had a massive November 2025 security breach (ShinyHunters, ~28M SoundCloud accounts exposed, OpenAI publicly removed Mixpanel from production). Statsig got acquired by OpenAI in September 2025 for $1.1B and then in May 2026 Amplitude took over the brand and customers while OpenAI kept the engineers. Piwik PRO sunset its free Core plan in February 2026, leaving small users orphaned. CNIL fined Google EUR 325M in September 2025 for consent violations, which means even GA4 sitting next to a cookie banner is now legal exposure if you do not enforce Consent Mode v2. And every 'best privacy analytics 2026' page on the internet pitches one tool at #1.

I ran four weeks of side-by-side testing on 25 tools. SaaS dashboards, ecommerce stacks, indie blogs, EU-strict shops. What follows is the honest version. Including where each tool is actually wrong for most readers.

Quick read: Plausible, Fathom, Umami, and Rybbit own the indie/SMB privacy-friendly tier. Microsoft Clarity is the best free heatmap tool in the world (just do not expect deep analytics). Mixpanel and Amplitude still do funnels and retention better than anyone but the November 2025 breach and renewal pricing are real. PostHog is the all-in-one for technical teams. Adobe Analytics, Contentsquare, and Pendo own enterprise. DataCops is not a Plausible replacement, it is the layer underneath that adds CNAME tracking + CAPI + bot filtering + first-party consent.

---

Quick stuff people keep asking

What does cookieless analytics actually mean? It means analytics that work without setting a third-party tracking cookie. Most of the tools in this list use either no cookie at all (Fathom, Plausible, Cloudflare Web Analytics) or a server-side salted hash that rotates regularly (Umami). Cookieless does not automatically mean GDPR-exempt. You still need to be honest about what you collect.

Do I still need a cookie banner with cookieless analytics? Often no. Plausible, Fathom, Simple Analytics, Umami, Rybbit, Friendly Captcha-style tools all run without a banner in most jurisdictions. The exception: if you also run advertising pixels, Stripe checkout cookies, or any third-party cookie, you still need a CMP for those. CNIL's EUR 325M Google fine in September 2025 made that real.

Is GA4 actually that bad? GA4 is free and dominates. The UI is widely hated (Search Engine Land literally published an article called 'Why people hate the Google Analytics 4 user interface'). Reports take 10+ clicks where UA took 2. UA historical data cannot be migrated. Most teams keep GA4 for Google Ads attribution and BigQuery export, then run a real analytics tool alongside.

Mixpanel had a breach. Should I switch? Mixpanel disclosed the November 2025 ShinyHunters smishing attack. Names, emails, and analytics data exposed across customers including OpenAI, SoundCloud (~28M accounts), CoinTracker, PornHub Premium. OpenAI publicly removed Mixpanel from production. If you are in a regulated industry, the renewal conversation just got harder. If you are a B2C startup, the product is still best-in-class for funnels and you can stay if your security team accepts the disclosure.

What is the cheapest cookieless analytics that actually works? Microsoft Clarity (free, unlimited, real product) for heatmaps and recordings. Cloudflare Web Analytics (free, unlimited) if you just want a server-log-style traffic dashboard. Umami Hobby (100K events/mo free) for proper privacy analytics. Plausible at $9/mo if you want a polished SaaS.

---

Tier 1: Privacy-first SaaS analytics (the indie/SMB sweet spot)

1. Plausible

The Good: Genuinely simple single-page dashboard. No cookie banner needed. GDPR/PECR/CCPA-friendly out of the box. Open source and self-hostable. Trusted brands include Hugging Face, 37signals, Ghost, Penpot, Tor Project. Lightweight script (<1KB).

Frustrations: Funnels and Looker Studio export are paywalled to the $39+ Business tier. Starter at $9/mo caps at 1 site. Trustpilot/Reddit reports of dashboards being locked when prepaid-annual customers exceed pageview cap.

Wish List: Soft limits instead of dashboard lockouts. Built-in funnels on the entry tier.

Value for Money: 7.5/10. One of the cleanest privacy-first analytics tools out there. Pricing tiers eroded some love.

Pricing: Starter $9/mo, Growth $19/mo, Business $39/mo.

---

2. Fathom Analytics

The Good: Privacy-first by design. Cookieless. GDPR/CCPA/PECR/ePrivacy compliant out of the box. EU-only data processing. Single-founder product, sustainable indie business.

Frustrations: Thin feature set. No funnels, cohorts, or proper user-journey analysis. No white-label or agency multi-client reporting. Limited segmentation.

Wish List: Funnels and basic retention/cohort views. Agency white-label.

Value for Money: 7.5/10. Cleanest privacy-first analytics for indie creators and SMBs who want pageview-level truth.

Pricing: From $15/mo (100K pageviews).

---

3. Simple Analytics

The Good: Minimalist single-page metrics. Cookieless, GDPR/CCPA/PECR compliant. EU-based company. Free forever plan (30-day retention). 50% non-profit discount. Strong transparency culture.

Frustrations: 30-day retention on free plan. Intentional simplicity hits a ceiling fast (no cohorts, weak funnels). Reviewers cite occasional UI bugs and slow page loads. Hard to understand user journeys.

Wish List: Optional power-user mode with funnels/cohorts. Longer free-tier retention to compete with Umami/Rybbit.

Value for Money: 7/10. If 'one page of metrics, no fuss, EU-hosted' is what you want, lovely. Anyone needing real product analytics outgrows it in a quarter.

Pricing: Free (30-day retention), paid usage-based slider.

---

4. Umami

The Good: Genuinely cookieless (server-side salted hash, rotates monthly). Free Hobby cloud tier (100K events/mo, 3 sites, no card). MIT-licensed self-host runs on a $5/mo VPS. Mainstream customers include AMD, Accenture, GM, ESPN, Siemens, Intel.

Frustrations: Hits a ceiling fast for advanced cohort analysis, revenue attribution, behavioral segmentation. Self-host requires Docker/Postgres ops knowledge. Limited integrations vs full analytics platforms.

Wish List: Native funnels and cohort segmentation in core. More polished UI to match Plausible/Rybbit.

Value for Money: 8/10. Best free open-source web analytics for indie hackers and small SaaS. Unbeatable for the price.

Pricing: Free Hobby, paid cloud from $9/mo, self-host free.

---

5. Rybbit

The Good: Genuinely cookieless. GDPR/CCPA-compliant. EU-hosted (Germany), no banner needed. Free tier (3K pageviews/mo, 1 site, 6 months retention). Cult-favorite UX, 0 to 10K+ GitHub stars in under a year. Reputation as 'simpler than Plausible, prettier than Umami'.

Frustrations: Very young product (founded January 2025). Feature gaps vs mature platforms. Limited integrations. Self-host still requires Docker/infra knowledge. Lifetime AppSumo deals signal early-revenue stage.

Wish List: Deeper funnels, cohorts, attribution. Native CDP/CAPI hooks for ecom teams.

Value for Money: 7.5/10. One of the best new privacy-first analytics tools to watch in 2026. Fast, cheap, well-designed, but young.

Pricing: Free 3K pageviews, paid tiers usage-based, self-host free.

---

6. Cloudflare Web Analytics

The Good: Genuinely free, no usage tier, unlimited pageviews. Privacy-first by default (cookieless, no fingerprinting, no PII in URLs). Lightweight beacon (~1KB) or server-side via Cloudflare proxy. GDPR-friendly without a CMP.

Frustrations: Only 30 days of data retention. YoY comparison impossible. Server-log-style accuracy: bot traffic pollutes stats. Reviewers report 'top OS unknown', 'top browser unknown', wp-login.php as a top page. Visitor counting is naive.

Wish List: Longer retention (at least 13 months). Real bot filtering and proper unique-visitor de-duplication.

Value for Money: 6.5/10. Free 'is the site up' dashboard. As actual analytics, it is a server-log viewer.

Pricing: Free (with any Cloudflare account).

---

7. Matomo

The Good: Open-source self-host gives 100% data ownership, no sampling. Privacy-first by design, cookieless tracking, EU residency, GDPR/CCPA workflows. Cloud plan from EUR 22/mo for 50K hits. Going through a public 2026 rebrand to fix UX.

Frustrations: Self-hosted requires running your own infra and paying separately for premium plugins. UI historically clunky (rebrand explicitly fixing this). Overage pricing (EUR 2.20 per 5K extra hits) catches people off guard.

Wish List: Bundle most-requested premium plugins into base tiers. Lower-friction self-hosted upgrade path.

Value for Money: 7.5/10. Best privacy-first GA alternative if you self-host or pay for Cloud. 2026 rebrand finally addresses UX.

Pricing: From EUR 22/mo Cloud, self-host free + paid plugins.

---

8. Piwik PRO

The Good: EU-hosted. Strong privacy/compliance posture (GDPR, HIPAA-friendly). Bundles analytics + tag manager + consent + CDP. Granular consent-mode integration and audit trails for enterprise compliance teams.

Frustrations: Free Core plan ended February 28, 2026. Major bait-and-switch complaints from users who lost dashboard access and historical data. Business plan jumps to ~EUR 35/mo minimum. Enterprise from ~EUR 10,995/yr.

Wish List: An honest mid-tier (sub-EUR 100/mo) for the small businesses orphaned by the Core sunset. Modern UI matching PostHog/Mixpanel.

Value for Money: 6.5/10. Solid EU-residency analytics for compliance enterprises. 2026 Core sunset burned a lot of goodwill.

Pricing: Business EUR 35/mo+, Enterprise EUR 10,995/yr+.

---

Tier 2: Free, dominant, lossy by design

9. Google Analytics 4

The Good: Free for the vast majority of sites. Generous limits before GA360 upsell. Native Google Ads, Search Console, BigQuery export (free). Unbeatable for paid-search-driven sites.

Frustrations: UI widely hated. UA historical data cannot be migrated/imported into GA4. CNIL fined Google EUR 325M in September 2025 for consent violations, which puts GA4 at the center of consent-enforcement scrutiny. Sampling kicks in on free tier at scale.

Wish List: A genuinely usable default UI. Importable historical UA data, even read-only.

Value for Money: 6/10. Free, dominant, disliked. Most teams keep it for Google Ads attribution and BigQuery export, then run a real tool alongside.

Pricing: Free, GA360 enterprise.

---

10. Microsoft Clarity

The Good: Genuinely free, no session caps, no recording limits. Heatmaps + session replay + AI insights + dead-click/rage-click detection. One-click Shopify install. No card ever.

Frustrations: 30-day retention only, no paid tier to extend. Heatmaps capped at 100K pageviews. Privacy posture mixed (US servers, EU regulators now treat with caution). Lazy-loaded pages produce incomplete screenshots.

Wish List: Longer (90+ day) retention as a paid add-on. Funnel/path analysis.

Value for Money: 8/10. Best free heatmap + session replay on the market.

Pricing: Free.

---

Tier 3: Product analytics (funnels, retention, cohorts)

11. Mixpanel

The Good: Best-in-class event analytics. Funnels, retention, flows, cohorts, formulas. Free plan generous (1M events, 10K session replays/mo). Pay-as-you-go ($0.28/1K events on Growth) more transparent than most.

Frustrations: Massive November 2025 ShinyHunters smishing breach exposed names, emails, analytics data across OpenAI, SoundCloud (~28M accounts), CoinTracker, PornHub Premium. OpenAI publicly removed Mixpanel from production. Costs balloon at scale. Add-on tax (pipelines, experiments, feature flags as separate SKUs).

Wish List: Hardware-key MFA across all employees. Roll add-ons into Growth instead of stacking SKUs.

Value for Money: 6.5/10. Most powerful in the category. November 2025 breach is a real conversation before renewal.

Pricing: Free 1M events, Growth $0.28/1K events, Enterprise custom.

---

12. Amplitude

The Good: Best-in-class for funnels, retention, pathfinder/journey reports. Gold standard for PM-led teams. Free Starter (50K MTUs, 12-month retention). Plus self-serve at $49/mo for 300K MTUs is one of the cheapest entry points.

Frustrations: 2-5x Mixpanel for equivalent volume per Reddit/HN. Growth/Enterprise pricing custom and opaque, quotes vary 5-10x. MTU-based pricing punishes traffic spikes. Took over Statsig brand from OpenAI in May 2026, ownership transition uncertain for Statsig customers.

Wish List: Public Growth tier pricing. Soft caps or burst protection for viral weeks.

Value for Money: 7/10. Safe choice if product analytics is your job. Budget for renewal sticker shock.

Pricing: Free Starter, Plus $49/mo, Growth/Enterprise custom.

---

13. PostHog

The Good: Generous free tier (1M events, 5K replays, 1M flag requests, 100K errors, 1.5K surveys/mo). All-in-one platform (analytics, replays, flags, experiments, surveys, errors) at one usage-based bill vs four vendors. Open source. $1.4B unicorn.

Frustrations: Steep learning curve cited across G2/Reddit. HogQL needs SQL. Usage-based pricing causes bill shock when modules turn on without guardrails. Dashboard overwhelming for early-stage users.

Wish List: Predictable spend caps and budget alerts. A 'simple mode' UI.

Value for Money: 8/10. Best for technical teams that want every product-data tool in one place. Overkill for non-technical SMBs.

Pricing: Free generous tier, then usage-based.

---

14. Heap

The Good: Auto-capture is the headline. Drop a snippet, retroactively track every click, form, pageview. Real-usable free tier (10K sessions, 6 months history). Strong session replay paired with autocapture.

Frustrations: Pricing opaque and quote-based above free tier. Reddit users: 'gets very expensive, very quickly'. Steep learning curve, advanced queries feel SQL-like. Now part of Contentsquare via Heap acquisition (2023).

Wish List: Publish Growth/Pro tier prices. Easier mobile-app instrumentation.

Value for Money: 6.5/10. Powerful auto-capture if you have budget and patience. Contentsquare merger pushes it more enterprise.

Pricing: Free (10K sessions), Growth/Pro sales-quoted.

---

15. Statsig

The Good: Generous Developer free tier (2M events/mo, 50K replays, unlimited flags, 1-year retention). Strong experimentation engine used by OpenAI, Atlassian, Notion. Pro tier $150/mo for 5M events.

Frustrations: OpenAI acquired Statsig $1.1B September 2025. May 2026: Amplitude took over the brand and customers while OpenAI kept the engineers. Optimizely's CEO publicly warned customers to be worried. 'Race car without a driver'.

Wish List: Clear roadmap commitments under Amplitude ownership. Better mid-market pricing.

Value for Money: 6.5/10. Best-in-class experimentation tech, but the 2025-2026 split put existing customers in limbo.

Pricing: Free Developer, Pro $150/mo.

---

16. Amplitude Product (alt slug)

The Good: Same engine as Amplitude. Same free Starter (50K MTUs, 12-month retention). Same Plus self-serve $49/mo.

Frustrations: Duplicate listing. There is no separate 'Amplitude Product' SKU, it is just Amplitude. Same Growth/Enterprise opacity. 8% annual auto-hikes.

Wish List: Clarify naming. 'Amplitude Product' confuses buyers comparing tools.

Value for Money: 7/10. Same as Amplitude.

Pricing: Same as Amplitude.

---

Tier 4: UX and session replay

17. FullStory

The Good: Best-in-class session replay. Autocapture means every click, scroll, keystroke recorded retroactively without prior instrumentation. Unusually generous free tier (30K sessions/mo, 10 seats). StoryAI powered by Vertex AI / Gemini.

Frustrations: Pricing fully opaque. Lowest reported paid tier ~$247/mo for 75K sessions, 2-month retention. Mid-market commonly $20K to $60K/yr. Aggressive renewal pricing.

Wish List: Published mid-market SKU. Cap on renewal price hikes.

Value for Money: 7/10. Excellent product, opaque sales motion. Free tier is a genuine gift.

Pricing: Free 30K sessions, paid sales-quoted.

---

18. Hotjar

The Good: Heatmaps + recordings + on-site surveys in one. De-facto starter heatmap product. Free Basic (35 daily sessions). 20% multi-product bundle discount with Observe + Ask + Engage.

Frustrations: Heavy data sampling. Users complain about the 'blind spot' on organic search traffic. Trustpilot ~2.5/5 with more 1-star than 5-star. Pricing escalates fast.

Wish List: Ditch sampling on paid tiers, especially for organic search. Real human support.

Value for Money: 6/10. Solid entry-level qualitative tool. You will outgrow the sampling caps.

Pricing: Free Basic, paid from $32/mo.

---

19. Mouseflow

The Good: Captures 100% of sessions on paid plans (no Hotjar-style sampling) with friction scoring. Free 500 sessions/mo and unlimited heatmaps. Paid from ~$31/mo. Strong funnel + form analytics.

Frustrations: Session-credit model burns through quotas fast on high-traffic sites. Tier jumps feel steep. Recording load and data search slow. 'Friction Score' opaque.

Wish List: Pay-as-you-go session top-ups. Faster replay loading.

Value for Money: 6.5/10. Better-than-Hotjar capture rate at similar price. Session-credit ceiling is the friction.

Pricing: Free 500 sessions, paid from $31/mo.

---

20. Contentsquare

The Good: Genuinely all-in-one experience analytics post Hotjar (2021) + Heap (2023) acquisitions. Session replay + heatmaps + product analytics + zone-based UX in one platform. Zoning analysis is unique (auto clickmaps tied to revenue per zone).

Frustrations: Pricing fully opaque. Mid-market deals (1-3M monthly sessions) typically $50K to $150K/yr per Vendr. Heap + Hotjar + Contentsquare merge means three legacy products stitched together. Layoffs.

Wish List: Real unified product instead of three legacy stacks. Public mid-market pricing.

Value for Money: 6.5/10. If you need session replay + heatmaps + product analytics in one enterprise contract, works. Watch the layoff trajectory.

Pricing: Sales-gated, $50K-$150K/yr mid-market.

---

Tier 5: Onboarding and product growth

21. Userpilot

The Good: Strong combo of product analytics + onboarding flows + in-app surveys. Useful for PLG SaaS. No-code flow builder. Resource Center, NPS, segmentation in higher tiers. Integrates with Mixpanel, Amplitude, Segment.

Frustrations: Starter $299/mo (annual) but excludes onboarding checklists, resource centers, A/B testing (those need Growth at $799/mo+). Pricing scales steeply with MAUs. Steep learning curve.

Wish List: Genuine self-serve cancellation. Cheaper entry tier with the basics.

Value for Money: 6/10. Powerful suite for funded PLG SaaS. Tough sell for early-stage.

Pricing: Starter $299/mo, Growth $799/mo+.

---

22. Pendo

The Good: Combines product analytics with in-app guides, NPS, feedback. Strong B2B SaaS fit. Acquired Forwrd.ai (2025) for predictive analytics and Chisel Labs (Feb 2026). Free tier up to 500 MAU.

Frustrations: Pricing famously opaque. Capterra/Vendr median customer pays $48,500/yr; range $7K to $133K+. MAU-based pricing punishes growth. Auto-renewing 1-year minimum contracts requiring Director-level approval to exit.

Wish List: Publish real prices. Flexible MAU bands.

Value for Money: 6/10. If you actually need product analytics + in-app guides + feedback in one stack, leader. If you just want analytics, overpaying 5-10x.

Pricing: Free 500 MAU, paid sales-quoted.

---

Tier 6: Enterprise

23. Adobe Analytics

The Good: Deep, surgical segmentation and calculated metrics. Workspace builder genuinely powerful for analysts. Customer Journey Analytics stitches cross-channel journeys in ways GA4 cannot.

Frustrations: Pricing opaque and brutal. No public list. Server-call/SKU-based quotes commonly $50K to $200K+/yr. First-year cost with implementation services often hits $200K to $500K. Steep learning curve.

Wish List: Transparent published mid-market pricing. Faster CJA migration with native UA-style reports.

Value for Money: 6.5/10. If deep in Adobe Experience Cloud with analyst headcount, still the most powerful. For everyone else, overkill at five-figure prices.

Pricing: $50K-$200K+/yr.

---

24. Kissmetrics

The Good: Person-based behavioral analytics. Tracks individuals across devices/sessions, not pageviews. Strong funnel + cohort with built-in A/B test analysis for SaaS/ecommerce. Cheaper entry than Mixpanel/Amplitude (~$25.99/mo for 10K events).

Frustrations: Brand turbulent. Domain handed to Neil Patel for SEO content in 2018. Bounced through ownership again with the SandStorm acquisition April 2025. Small team (~40 employees). Higher tiers escalate quickly (Gold reportedly steep).

Wish List: Transparent pricing. Modern UI refresh.

Value for Money: 5.5/10. Niche behavioral analytics. Cheaper than the big names. The company history makes it riskier long-term.

Pricing: From ~$25.99/mo.

---

25. Woopra

The Good: Customer journey analytics is core. People-profile views with action-by-action timelines beat session-blob analytics for product/marketing teams. Free Startup tier still exists.

Frustrations: Maintenance/rebrand limbo. G2 lists as 'Appier AIRIS (formerly Woopra)'. Standalone Woopra brand gone quiet. Pro plan ~$1,200/yr feels steep vs Mixpanel Free/Growth. Tracxn lists ~7 employees mid-2024.

Wish List: Clear product direction. Self-serve modern pricing.

Value for Money: 5/10. Once-loved tool now living inside Appier AIRIS. Fine if you already use it. Hard to recommend new in 2026.

Pricing: Pro ~$1,200/yr.

---

Where DataCops fits (the layer underneath)

DataCops is not a Plausible, Fathom, or Mixpanel replacement. It is the trust-infrastructure layer that sits underneath whatever analytics dashboard you already use.

What it adds:

• First-party CNAME tracking on datacops.yourdomain.com. JS served from your own subdomain. Survives uBlock, Brave Shields, Pi-hole, iOS Safari ITP. Recovers 15-25% of lost session data that even Plausible misses.
• Server-side CAPI to Meta, Google, TikTok, LinkedIn. Your privacy-friendly dashboard does not handle conversion fan-out. DataCops does.
• Bot/fraud filtering on 361B+ tracked IPs (146.4B datacenter, 11.9B VPN). Filters bots before they pollute your dashboard.
• TCF 2.2 first-party CMP. Consent state stored on your subdomain.

The honest framing: keep the dashboard you like, plug DataCops in for the parts those tools do not do. Bundles four vendor categories into one. Free tier real (2K sessions, no card). $7.99/mo Growth, $49/mo Business with HubSpot, $299/mo Organization, Enterprise talk-to-sales.

Not for: shops that already have a four-vendor enterprise stack and do not want to consolidate.

Value for Money: 9/10 for the trust-infrastructure layer. N/A as a Plausible/Mixpanel swap.

---

So what should you actually use?

A lot of tools. No one-size-fits-all. The real question is what you actually need.

• Indie blog or landing page? Try Plausible, Fathom, Umami, or Rybbit.
• Free heatmaps and session replay? Microsoft Clarity is unbeatable.
• Free traffic dashboard, no setup? Cloudflare Web Analytics.
• B2C product team needs funnels and retention? Mixpanel (read the breach disclosure first) or Amplitude.
• Technical team wants every product tool in one bill? PostHog.
• Strict EU residency, compliance-driven? Matomo, Piwik PRO, or Friendly Captcha-style + Umami self-host.
• Need session replay with auto-capture? FullStory free tier or Heap free tier.
• Need product analytics + in-app guides + feedback? Pendo.
• Already deep in Adobe Experience Cloud? Adobe Analytics.
• Want CNAME tracking + CAPI + bot filter + first-party consent underneath your dashboard? DataCops.

---

The mistake I see people make

Replacing GA4 with Plausible and calling it done. Plausible is great but it is a dashboard. It does not push server-side conversions to Meta or Google. It does not filter bots before they hit your numbers. It does not manage consent. The bot that hits your site still hits your CAPI, still pollutes your ad algorithm, still triggers your Stripe checkout cookies which still need a CMP. Cookieless analytics solves the cookie banner question for the dashboard layer. It does not solve the trust-infrastructure question for the rest of your stack.

Related reading:

• DataCops vs PostHog
• DataCops vs Fathom
• DataCops vs Mixpanel
• Best GA4 alternative 2026
• Best Google Analytics alternative 2026

---

Now your turn

What is your analytics stack in 2026? Plausible + GA4? Mixpanel post-breach? PostHog all-in-one? Drop your setup (or your horror story) below.