PostHog

James Hawkins and Tim Glaser met at Arachnys, a London-based RegTech startup. Hawkins had arrived there as a former competitive cyclist who'd taught himself to code to fund race fees. Glaser had been writing code since age 11 and taking money for it by 13. When they left to start a company together in January 2020, they had drive, complementary skills, and absolutely no idea what they were building.

What followed was nine months of controlled chaos: six products, six pivots, and one abandoned idea they killed after five days. The playbook most founders write about persistence usually glosses over how many wrong turns come first.

Pivot six came from a real frustration: companies wanted to understand their users but couldn't share behavioral data with third-party analytics tools because of privacy restrictions, compliance departments, and general wariness about where data was going. The answer was to build analytics that companies could run themselves - on their own infrastructure, with their own data, under their own control.

They shipped the MVP four weeks after writing the first line of code. Posted it on Hacker News. Within days, 300 teams had deployed it. Within two weeks, 1,500+ GitHub stars. According to the team, it was the most successful B2B software launch on Hacker News since 2012.

In June 2024, PostHog's CEO James Hawkins sent a tweet. Stripe read it, reached out, and led a $70 million Series D at a $920 million valuation. No deck required. No pitch meeting scheduled. No introductions brokered through a warm connection.

The story spread fast - partly because it was remarkable, and partly because it fit exactly the kind of company PostHog presents itself as: transparent, direct, and a little allergic to the usual theatrics of startup fundraising.

That round valued PostHog at $920 million. A year later, the Series E - $75 million led by Peak XV Partners (formerly Sequoia India) - pushed the valuation to $1.4 billion. Total funding now stands at approximately $182 million across five rounds, with YC, GV, Sofina, Lux Capital, and notable angels including Jason Warner (former CTO at GitHub) and Solomon Hykes (founder of Docker) on the cap table.

A typical product engineering team in 2024 was paying for Amplitude or Mixpanel (product analytics), Hotjar or FullStory (session replay), LaunchDarkly (feature flags), Optimizely (A/B testing), and Segment (data pipelines) - all separately, all with different pricing models, all with data that lived in different places.

PostHog replaces all of it. Not by doing everything badly, but by building each product to a level where it is competitive on its own. Each tool in the platform is run by a small, autonomous team with its own roadmap, treated as a startup within a startup.

The pricing reflects the model. The free tier covers 1 million events per month, 5,000 session recordings, 1 million feature flag requests, and 100,000 exceptions tracked. Above that, you pay per use. 98% of PostHog's 190,000+ teams never pay a cent. The remaining 2% - mostly larger companies with serious data volumes - generate the revenue that funds everything else.

The analytics engine runs on ClickHouse, an open-source columnar database that delivers sub-second query performance on billions of events. PostHog built a custom SQL dialect called HogQL on top of it, letting engineers write their own queries directly against their data. No dashboards. No black boxes. Just SQL.

Notable angels include Jason Warner (former CTO, GitHub), Solomon Hykes (Founder, Docker), and Stripe itself, which stepped into a lead role after that now-famous tweet. The involvement of builders-turned-investors who've operated at developer-scale is not coincidental - PostHog's thesis resonates most clearly with people who've lived the problem it's solving.

PostHog publishes everything: compensation bands, equity structure, how they hire, how they fire, what they email investors, what their strategy is, and why. The company handbook lives at posthog.com/handbook and is updated regularly. There's no PR filter between the reader and the actual thinking behind the company.

It's a notable choice in an industry where most companies treat internal processes as competitive secrets. PostHog's bet is that transparency attracts better people, builds trust with users, and - frankly - keeps the company honest.

With the Series E in September 2025, PostHog announced something they're calling "Act 2." The direction is a logical extension of the original thesis: if engineers should be closer to user data, the next step is giving them tools that close the loop automatically.

The vision: you log into PostHog and see a list of pull requests already written, based on support tickets, user feedback, and product events since you last checked in. For smaller issues, the workflow gets the developer straight to reviewing, merging, or editing. Routine fixes that shouldn't require a planning meeting do not require one.

This is AI applied to the part of software development that's mostly friction: the gap between knowing something is wrong and doing something about it. PostHog's platform already holds the analytics, the feedback, and the session data. Connecting it to code generation is the next logical move.