Plausible Sells Subscriptions, Not Surveillance

Analytics vendor marketing still sells attribution theater, "free" tiers funded by adtech, and dashboards that require a certification course. Plausible Analytics sells those surfaces too—but the durable bet is quieter. Plausible is privacy-first web analytics funded entirely by subscriptions: no cookies, no personal-data collection, no ad network, no investor board pushing surveillance pivots. Judge it on whether your custody contract matches GDPR reality—and whether AGPL code plus export paths give you an exit ramp—not on whether "Plausible" clears a generic brand-naming worksheet.

The insight was GA integration disgust, not a category map

Uku Täht is an Estonian developer who wrote the first line of Plausible code in December 2018 after a mundane assignment: the marketing head at his employer asked him to integrate Google Analytics into a landing page. His first thought, he told Marko Saric's $1M ARR retrospective, was blunt: "Ugh. Can we just use something other than Google Analytics?"

That reaction is the product thesis in one sentence. Web analytics had drifted from "help site owners understand traffic" into surveillance infrastructure—cookies, cross-site profiles, data brokerage, and a business model where the vendor's incentives conflict with the customer's privacy promises.

Uku published "The analytics tool I want" in early 2019, spent two months on a prototype, and launched a public beta in February 2019 on Indie Hackers—"build in public" before it was a LinkedIn cliché. Paid subscriptions followed in May 2019 at deliberately low price points ($6–$36/mo by traffic tier) because GA is free and switching requires a reason (launch post).

Marko Saric joined as co-founder in March 2020 after Uku cold-emailed him. Saric had published a guide on de-Googling a website; Uku had built the analytics piece Saric's readers kept asking for. On Saric's first day, Plausible showed 49 total visitors and zero from Google search (same retrospective). His first goal was ten organic Google visitors per day—not vanity billboards.

The March 2020 relaunch fixed early inconsistencies (the product had briefly appeared as "Plausible Insights" in some places), removed the first-party cookie the beta used, and sharpened positioning: simple, lightweight, open source, privacy-first Google Analytics alternative. The viral wedge arrived April 8, 2020: "Why you should stop using Google Analytics on your website"—top of Hacker News, 25,000+ visitors in one day, and the beginning of compounding word-of-mouth Plausible still relies on.

Bootstrapped economics: subscribers are the only shareholder

Plausible's business model is intentionally boring—which is the point.

| Revenue source | What Plausible does | What Plausible refuses | | --- | --- | --- | | Hosted SaaS | Traffic-tier subscriptions from $9/mo (10k pageviews) on current plans | Freemium tiers that fund support debt for solo founders | | Self-hosting | AGPL Community Edition you can run on your infra | Pretending self-host is the same SLA as managed EU cloud | | Giveback | 5% of gross revenue to environmental and open-source causes (since 2021) | "We are a nonprofit" cosplay—this is a for-profit company with values | | Growth | Content, SEO, community recommendations—zero paid ads per about page | Affiliate spam, surveillance-funded user acquisition |

The company is incorporated in Estonia, team and infrastructure EU-based, with no outside investors, no acquisition plans, and no secondary revenue from visitor data (about). Subscription fees fund salaries, ClickHouse clusters, and documentation—not ad targeting R&D.

Public milestones worth keeping straight:

• June 2022: $1M ARR; 7,000+ paying subscribers; 50,000+ sites; 1B+ monthly pageviews counted (milestone post).
• 2026 (homepage): 20,000+ paying subscribers; 260 billion pageviews tracked lifetime; 99.99% uptime (last 90 days); team of 10.

Growth from $400 MRR to $10,000 MRR took nine months after paid launch; $500k ARR to $1M ARR took eight more—classic bootstrapped compounding, not blitz-scaling. Uku and Saric burned >$50,000 in personal savings before $10k MRR made the company sustainable (journey post). Donations as monetization failed too: six months of tip jars yielded six $5 donations while cloud MRR climbed past $8,500—a lesson they published so other OSS founders would not repeat it.

Open-core model: AGPL inspection, cloud convenience

Plausible's enduring strategy is not "be a cheaper GA." It is two compounding layers with a license boundary:

| Layer | What Plausible open-sources | What Plausible sells | | --- | --- | --- | | Core analytics | Full SaaS codebase on GitHub under AGPLv3 (LICENSE) | Trust, inspection, fork rights | | Managed cloud | Nothing hidden in "enterprise edition"—same repo powers plausible.io | EU hosting, backups, ClickHouse ops, support, Stats API at Business tier | | Script delivery | <1KB gzip/brotli variants via combinatorial build (script docs) | Managed proxy options, WordPress plugin, NPM @plausible-analytics/tracker | | Compliance pack | Public data policy + independent GDPR legal assessment | DPA, subprocessors list, EU-only residency story |

September 2019: Plausible went open source under MIT—full transparency, self-host path. October 2020: switched to AGPLv3 after permissive-license risk became visible—corporations can strip MIT products into proprietary SaaS without contributing back (AGPL post). Saric learned about AGPL days before the switch; the boundary has held since.

June 2020: migrated analytics storage from PostgreSQL to ClickHouse when trial volume spiked—now counting 1B+ pageviews/month with sub-second dashboards (retrospective). That infra bet is part of the product: privacy-first analytics fails if the dashboard is slow enough to tempt teams back to GA.

Self-hosting is real—but ops are yours. The cloud product is the default buyer path for teams that want GDPR paperwork and EU residency without running ClickHouse on weekends.

Privacy architecture: aggregate stats without a surveillance contract

Plausible's compliance story is structural, not a marketing badge:

• No cookies and no personal data collection by default; no cross-site or cross-device tracking (data policy).
• Visitor data processed only in aggregate; IP addresses discarded or hashed—never stored as identifiers.
• All visitor data stays in the EU on European-owned infrastructure—no US transfers, no US parent company, no US investor board control (EU hosting page).
• Independent GDPR legal assessment published; DPA available for organizations that need processor paperwork.

The 2022 regulatory tailwind was external: Austria's Data Protection Authority found Google Analytics GDPR-noncompliant; French, Italian, Danish, and other DPAs followed similar reasoning under Schrems II—US surveillance law breaks EU transfer adequacy even with Data Privacy Framework paperwork. Plausible did not cause those rulings, but EU-only custody became a procurement checkbox overnight (retrospective).

That does not mean "install Plausible and forget lawyers." You still disclose analytics in your privacy policy; you still verify subprocessors; you still confirm Plausible fits your processing purpose. Plausible publishes "when Plausible is not a fit"—a rare honesty move in vendor marketing.

Integration is one deferred script tag—or a first-party proxy mapping /js/script.js and /api/event to Plausible to reduce adblock loss (6–26% of visitors block third-party analytics scripts; tech audiences can hit 60% per Plausible's adblocker study).

<script defer data-domain="yourdomain.com" src="https://plausible.io/js/script.js"></script>

Business-tier additions—Stats API, funnels, ecommerce revenue, Search Console import, AI referrer tracking—expand usefulness without reintroducing adtech identifiers. The script stays modular: Handlebars-built variants keep compressed payloads tiny even as features multiply.

Product arc: beta → AGPL → GDPR moment → scale

| Era | What shipped | Strategic move | | --- | --- | --- | | Feb 2019 | Public beta on Indie Hackers | Build-in-public community | | May 2019 | Paid tiers ($64 MRR day one) | Prove willingness to pay vs free GA | | Sep 2019 | Full OSS (MIT) on GitHub | Trust in privacy market | | Mar 2020 | Marko joins; cookie removed | Marketing + positioning split with Uku on product | | Apr 2020 | Anti-GA HN post | Category polemic as distribution | | Jun 2020 | ClickHouse migration | Scale without dashboard shame | | Oct 2020 | MIT → AGPLv3 | Protect OSS sustainability | | Jun 2022 | $1M ARR | Bootstrapped OSS SaaS proof point | | 2023–2026 | GA import, funnels, AI traffic, SSO (Enterprise) | Land on compliance; expand on product teams |

Customers cited publicly include Basecamp, Ghost, Hugging Face, MongoDB, the Python Software Foundation, and Harvard—not logos to borrow for your DPIA, but signal that engineering-led orgs accept the tradeoffs.

Competitors: same brochure, different custody and business model

Privacy analytics checklists converge: cookieless, simple dashboard, UTM campaigns. The custody and monetization contracts do not.

| | Plausible | Google Analytics 4 | Fathom | Simple Analytics | Umami | | --- | --- | --- | --- | --- | --- | | Business model | Subscriptions only | Surveillance/ad ecosystem | Subscriptions | Subscriptions | Cloud + MIT self-host | | Open source | AGPL full SaaS repo | Closed | Closed | Closed | MIT | | EU-only hosting | Yes (Estonia; EU infra) | US Google Cloud | EU + US CDN paths | Netherlands EU | Where you deploy | | Self-host | Community Edition (AGPL) | No | No | No | First-class | | Sweet spot | EU compliance + OSS credibility + content SEO | Free + ad product integration | Agency simplicity | Plain-English privacy UX | Dev cost control | | Weak spot | Not ad-attribution deep; script blocked unless proxied | GDPR transfer risk; complexity | No OSS fork; fewer infra features | Price at scale; closed code | You own ops | | Entry hosted price | $9/mo @ 10k pageviews | Free | ~$15/mo @ 100k pageviews (verify) | ~$20/mo @ 100k (verify) | Free self-host; cloud tiers vary |

Choose Plausible when you need EU-only data custody, want AGPL inspectability, and value content-driven migration guides plus GA import—not when you need Google's ad network feedback loops.

Choose GA4 when marketing lives inside Google Ads/Search 360 and legal accepts US transfer mechanisms—accept surveillance economics.

Choose Fathom when you want a closed, simple SaaS with flat multi-site pricing and minimal ops—fork rights matter less.

Choose Simple Analytics when Netherlands EU residency and minimalist UX beat OSS—and you accept proprietary custody.

Choose Umami when MIT self-host on your Postgres beats managed EU SaaS—and you have engineers to patch.

Matomo (not in table to save space) still wins regulated self-host deployments that need GA-like depth with on-prem custody—different ops tax.

Model total cost at your actual pageviews, not entry-tier marketing. A self-hosted AGPL fork saves subscription fees but adds infra, backup, and security patch labor.

China: cross-border custody vs mainland analytics gravity

Plausible's China story is not "Plausible 中国." It is cross-border teams needing EU-grade analytics custody—a different problem from owning mainland ISP and regulatory relationships.

What Plausible does well (出海 / GDPR-facing sites):

• Chinese SaaS, AI tools, and indie products targeting EU users increasingly need Schrems II–safe analytics—Plausible's no US transfer story is the product.
• Cookieless, aggregate measurement aligns with tightening global privacy expectations; under GDPR, properly configured Plausible often avoids consent-banner friction for basic traffic stats—still document it in your privacy policy.
• AGPL appeals to engineering teams that want inspectability without sending visitor data to US ad platforms.
• English-first docs, GA migration importers, and HN-era credibility match global dev stack buyers (Vercel/Next.js sites, static docs, product marketing pages).

What Plausible does not replace (境内站点):

• Mainland-facing sites typically run 百度统计, 友盟+, GrowingIO, CNZZ, or vendor suites inside 微信/字节/阿里 ad ecosystems—not plausible.io scripts blocked or slowed by network paths.
• 《个人信息保护法》（PIPL） and 网络安全法 impose localization and consent rules that differ from GDPR framing; EU-only hosting does not automatically satisfy CII operator or cross-border transfer security assessment（数据出境安全评估） requirements for Chinese user data.
• ICP备案 and domestic commercial analytics often require local entity relationships Plausible does not provide.
• Ad attribution and funnel depth for domestic performance marketing still lives in closed Chinese ad platforms—Plausible will not replace 巨量引擎 or腾讯广告 feedback loops.

Practical split-stack for Chinese teams:

| Audience | Typical stack | Why | | --- | --- | --- | | EU/UK users (GDPR) | Plausible cloud or EU self-host | No US transfer; cookieless aggregate | | Mainland users (中文产品) | 百度统计 / 友盟 / GrowingIO | Local compliance, ad network integration | | Global product + China entity | Plausible for .com marketing site; domestic tool for .cn app | Two privacy policies—two processors | | OSS purist, any region | Self-host Umami or Plausible CE | You operate Postgres/ClickHouse + backups |

The old startup-showcase template implied one analytics vendor "works everywhere." Reality: Plausible wins EU custody and OSS trust; mainland daily measurement still rewards domestic analytics gravity.

The name and domain (briefly—facts beat folklore)

The verified story is shorter than a naming-pillar worksheet:

• "Plausible" is plain English—believable, credible, reasonable—a deliberate contrast to analytics numbers inflated by bot traffic and dark patterns. It signals trustworthy metrics, not surveillance depth.
• Early copy inconsistently used "Plausible Insights"; Saric's 2020 relaunch unified Plausible Analytics everywhere (retrospective).
• plausible.io is product, docs, and dashboard on a clean .io dev-tool domain—honest about audience without .com gymnastics. Script default lives at plausible.io/js/script.js; managed proxy maps it to your first-party path.

That is category naming aligned with privacy positioning—not a five-act "domain scorecard."

Prove the exit ramp: parallel run + self-host drill

Privacy trust is not a homepage adjective. Run this before you treat "AGPL means we're free" as full exit freedom:

1. Parallel run against GA4 for 30 days. Install Plausible alongside existing analytics; compare pageview deltas after bot filtering. Expect differences—define which tool is source-of-truth for board metrics.

2. Export and API inventory. Business tier includes Stats API and GA import—not full historical export of every custom event. List goals, funnels, and segments that live only in Plausible's UI.

3. Self-host smoke test. Bring up Community Edition with ClickHouse/PostgreSQL per docs. One tracked pageview end-to-end is the bar—you are proving AGPL custody, not auditioning to run analytics infra forever.

4. Proxy/adblock rehearsal. Measure traffic loss on plausible.io third-party script vs first-party proxy; update CSP script-src and connect-src accordingly (CSP docs thread).

5. Document the delta. Date the drill. If compliance bought Plausible for DPA + EU residency but marketing still runs GA4 for ads, you have split custody—not a migration.

What to verify before your team standardizes on it

1. Legal fit, not logo fit. Read Plausible's not-fit page and your DPA requirements—confirm aggregate analytics matches your processing register.

2. Traffic tier math. Price at peak monthly pageviews across all domains on consolidated plans—Starter ($9), Growth ($14), Business ($19) scale with volume (pricing).

3. Feature depth vs GA4. Funnels and ecommerce exist; advanced ad attribution and BigQuery export do not—confirm marketing will not secretly keep GA4 "just for ads."

4. China audience split. Mainland user analytics likely needs domestic vendors regardless of how clean Plausible's EU story reads in English.

5. Script custody. Third-party loading loses data to adblockers; proxy setup adds eng work but recovers signal—budget both.

Plausible's competitive edge is not the loudest slide in a "cookieless future" webinar. It is subscription-funded, AGPL-inspectable, EU-only analytics built by a team of ten that crossed $1M ARR without VC, grew through anti-GA content and word of mouth, and still funds itself only from customers who want measurement without surveillance. Price the custody contract—and run the parallel drill before your privacy policy outlives your analytics stack.