Cookies are dying. Regulators are circling. GA4 stayed because the migration cost was too high to replace. Everyone else built around it. The 2026 analytics stack has more vendors than at any point since 2015 — and most of them did not exist five years ago. Marketing leadership at companies as different as DuckDuckGo, Stripe, Vercel, and the European Parliament made the same decision in different directions for the same reasons.
What changed
GDPR enforcement matured under the European Data Protection Board and national authorities. CNIL in France issued the December 2021 ruling that Google Analytics in its then-current form violated GDPR. Austria's DSB followed in January 2022. Italy's Garante issued a similar ruling in June 2022. Denmark's Datatilsynet in September 2022. The Norwegian Datatilsynet in 2023. CCPA expanded into CPRA under Ashkan Soltani at the California Privacy Protection Agency. China's PIPL came online and enforcement under the Cyberspace Administration of China is more aggressive than most Western firms had planned for. The EU Digital Markets Act under Thierry Breton, then Internal Market Commissioner, forced gatekeeper platforms to surface choice screens that meaningfully reduce third-party cookie acceptance.
Google's third-party cookie deprecation in Chrome — first announced January 2020, repeatedly delayed under Anthony Chavez and the Privacy Sandbox team — ended with a July 2024 announcement that Chrome would offer users a choice mechanism rather than fully deprecate cookies. The decision solved nothing for advertisers and almost nothing for publishers. The net effect across the whole regulatory and technical layer: cookie-based measurement degraded faster than alternatives matured. The vendors who built privacy-first products had a five-year head start when the market finally needed them.
GA4: the default nobody loves
Universal Analytics shut down July 2023 after a one-year migration window that Google extended once. GA4 inherited the install base by default. The event-based model designed by Russ Ketchum's team is more flexible than UA's session-based model. The interface is harder to use. The data model breaks naive comparisons to old reports. Sampling kicks in above 10 million events per month for standard accounts. Server-side tagging through Google Tag Manager — and through third-party services like Stape, Addingwell, and Snowplow — solved the cookie problem for advertisers willing to invest in the implementation. Smaller advertisers gave up and either accepted the data loss or moved to privacy-first alternatives.
Plausible and Fathom: the privacy-first reset
Plausible Analytics — open-source, EU-hosted, cookie-free, founded by Uku Täht and Marko Saric in Estonia — became the default for European publishers and SaaS companies that wanted compliance without consent banners. Customers including DuckDuckGo, the European Parliament, the Linux Foundation, the BBC's developer team, and Penpot run Plausible publicly. Fathom Analytics, founded by Jack Ellis and Paul Jarvis in Canada, positioned similarly for the North American market with customers including Buffer, Drift, and a long list of small SaaS companies. The pitch is straightforward: simpler product, better defaults, no cookies, no consent prompts required in most jurisdictions. The trade-off is depth — neither tool replaces GA4 for advertisers who need conversion attribution across paid channels.
Simple Analytics from the Netherlands, Pirsch, Umami, GoatCounter, and Tinybird's Plausible alternative all entered the same category. The list grew faster than the market would support; consolidation is likely through 2026 and 2027.
Matomo and Piwik PRO: the self-hosted comeback
Matomo, founded by Matthieu Aubry in 2007, is the longest-running open-source analytics platform. Piwik PRO targets enterprises that need on-premise deployment for regulatory reasons — financial services, healthcare, government. Both saw renewed enterprise interest as data-residency requirements tightened. The European Commission migrated public-facing analytics off Google Analytics onto Matomo and similar tools across several agencies. Customers including Forbes, Lufthansa, the UN, and the Wikimedia Foundation run Matomo at scale. Self-hosting is operationally heavier than SaaS. For regulated industries the trade-off is no longer optional.
Adobe Analytics: the enterprise hold
Adobe Analytics, inside the Adobe Experience Platform run by Anjul Bhambhri and Anil Kamath, retained the enterprise category that never seriously migrated to GA4. The integration with Adobe Audience Manager, Real-Time CDP, and Customer Journey Analytics makes it a defensible enterprise stack — not because the analytics product is better than GA4 but because the surrounding tooling is. Adobe's customer base — Marriott, JPMorgan Chase, Verizon, Bank of America, Walmart, Target, Disney, Comcast — has stayed with Adobe through the GA4 transition because the switching cost is prohibitive. Adobe's pricing keeps it out of mid-market deployments.
Server-side and the cookie workaround
Server-side tagging is the most consequential measurement change of the past five years and the least-discussed. Tag data flows from the browser to a first-party server endpoint and then to analytics vendors, ad platforms, and warehouses. Cookies become first-party. Ad blockers see less. Consent management is cleaner. Implementation is a real engineering project. Stape, Addingwell, GTM Edge, and Cloudflare Zaraz lead the dedicated server-side tagging vendor category. The agencies that built server-side practices in 2022 — Search Discovery, Bounteous, Napkyn — are billing premium rates for the work now. Customer Data Platforms — Segment (now Twilio Engage), mParticle, Tealium, Treasure Data, Rudderstack, and Snowflake's own CDP capabilities — are now part of the same conversation, even though they were historically positioned as adjacent to analytics rather than as analytics replacements.
Consent management: the unavoidable middle layer
Every privacy-aware stack needs a Consent Management Platform. OneTrust, founded by Kabir Barday, is the category leader by enterprise share. Didomi from France competes strongly in EU markets. Sourcepoint, TrustArc, Cookiebot from Usercentrics, Iubenda, Quantcast Choice, and CookieYes fill out the category. The CMP market consolidated significantly in 2024 and 2025. Usercentrics acquired Cookiebot. OneTrust raised additional growth capital. The category is now table stakes for any consumer-facing brand operating across multiple jurisdictions.
The AI measurement layer
No traditional analytics tool measures AI engine citation. The AI measurement layer sits alongside the conventional stack — Citation Share monitoring through Profound, Otterly, AthenaHQ, Daydream, BrandRank.ai, Peec AI, or in-house tools. AI crawler log analysis through Cloudflare or dedicated log-analysis tools. Brand mention parsing across engines through services that interrogate the engines directly. Treat it as a separate budget line, not a feature of any existing analytics vendor. The vendor that bolts it cleanly into GA4 or Adobe Analytics will own a category that does not yet have a market leader. Adobe announced an AI Assistant for Customer Journey Analytics in 2024; the product addresses different problems than Citation Share but signals which direction enterprise analytics is heading.
The 2026 picks
Mid-market consumer brands — GA4 plus a privacy-first secondary tool, plus dedicated Citation Share monitoring. Enterprise — Adobe Analytics or GA4 360 with full server-side tagging implementation and a custom AI measurement layer. Publishers — Plausible or Fathom for editorial reporting, a separate ad measurement stack for monetization. Regulated industries — Matomo or Piwik PRO on-premise plus consent management built for primary jurisdiction. SaaS B2B — increasingly Posthog for product analytics combined with GA4 or Plausible for marketing, with Citation Share monitoring as a top-of-funnel measurement layer. Every stack should include an AI engine measurement layer.
The single-vendor analytics era is over. Build for the surfaces the buyer actually uses.
Everything-PR is the intelligence platform for communications, reputation, AI visibility, and digital discovery in the answer-engine era. Thirty-plus publications. Publishing since 2009. Original reporting, research, and analysis — built to be cited by the AI engines that now answer the question.
