What Is PostHog? How the Product Analytics Platform Works?

Product teams love data. However, they hate juggling tools. For example, one app tracks clicks. Another records screens. A third controls feature launches. Soon, the bills pile up, and your data lives in five places. In short, that is the problem PostHog set out to solve.

So, what is PostHog? Simply put, it is an open-source product analytics platform. It also bundles session replay, feature flags, experiments, surveys, and more. In this listicle, you’ll learn how PostHog works, what tools it offers, and what it costs.

You will encounter actual figures, straightforward examples, and actions you can take today as you progress. By the end, you will know if this platform fits your team.

What Is PostHog? The All-in-One Product Platform Explained

The Short Answer

Put simply, PostHog shows you how people use your app or website. Then it helps you act on what you learn. However, many analytics tools do only one job. PostHog, in contrast, offers a full toolkit in one place.

In fact, the company calls it a “Product OS.” In plain words, you add one code snippet and get analytics, replays, flags, and surveys together. Furthermore, it was built by engineers for engineers.

For that reason, the pricing is open, and the docs are deep. It is also fully self-serve. As a result, you can sign up and start the same day without talking to sales.

Who Built It and Why It Matters

First, James Hawkins and Tim Glaser founded the company in January 2020. Then they joined Y Combinator’s Winter 2020 batch. Since then, growth has been rapid.

For example, in June 2025, PostHog raised a $70 million Series D led by Stripe at a $920 million valuation, according to Crunchbase News. Later in 2025, a $75 million Series E led by Peak XV Partners lifted the value to $1.4 billion.

Meanwhile, Sacra estimates that annual recurring revenue hit about $57.5 million in February 2026. That number is almost twice as high as it was the previous year. However, keep in mind that this figure is an outside estimate, not an official number.

Why does this figure matter to you? In short, funding and growth signal staying power. You want to build your data stack on a tool that will still be available next year.

The Insight Most Guides Miss

Key Insight:

However, most listicles stop at the feature list. Instead, the bigger idea is consolidation. When analytics, replays, and flags share one home, you can click from a chart straight to the real users behind it.

By contrast, separate tools make that challenging. In fact, the connection between numerical data and actual user sessions is the true value of the product. It is also why a product analytics platform like PostHog can save hours each week.

How PostHog Works: From Tracking Code to Real Insight

Step 1: Capture Events

So, everything starts with an event. In other words, an event is a record of something a user did, such as a pageview, a click, or a signup. Specifically, PostHog collects events through SDKs, which are small code libraries.

For the web, it supports JavaScript, React, Next.js, and Vue. Likewise, for mobile, it offers libraries for Android, iOS, React Native, and Flutter.

Here is a simple way to begin.

• First, create a free account on PostHog Cloud.
• Next, copy your project key and add the snippet to your site or app.
• Then turn on autocapture, which logs clicks and page views on its own.
• Finally, add a few custom events for key moments, such as a plan upgrade or a finished checkout.

Step 2: Analyze with Insights and HogQL

Thereafter, your data lands in ClickHouse, a highly performant database made for huge event tables. On top of it, PostHog adds HogQL, a query layer that works like SQL. So you can build charts with a few clicks or write your own queries if you prefer.

In addition, popular views include trends, funnels, retention, and user paths. For example, a funnel shows how many visitors move from your landing page to sign up and then to a first purchase.

Step 3: Connect Numbers to People

Here is where the platform shines. Each event carries a session ID, so related actions stay linked. Therefore, when a funnel shows a big drop, you can open the replays of users who left.

For instance, you may see a stuck button, a confusing form, or an error message. In short, the chart tells you what happened, and the replay tells you why.

A Tip to Keep Costs Low

Autocapture is handy, but your bill depends on event volume. In fact, the cheapest event is the one you never send. So, start with ten to fifteen events tied to your main user journey. Then add more only when a real question needs them.

Key PostHog Features and Products You Can Use

The Core Toolkit

PostHog lists more than ten products today. However, you do not need all of them on day one. Instead, concentrate on the few that most effectively address your biggest question.

Product analytics tracks trends, funnels, retention, and user paths. Furthermore, web analytics gives a simple dashboard for visitors, page views, and conversions.

Meanwhile, session replay allows you to view actual sessions, complete with console logs and network requests. Plus, error tracking captures errors and stack traces, enabling developers to resolve issues faster.

Next come the release tools.

• First, feature flags let you turn a feature on for some users and roll it back at once.
• Then, experiments run A/B tests with statistical checks.
• Surveys ultimately gather feedback through no-code forms on both mobile and web.

The Data and AI Layer

Beyond that, PostHog also works as a light data stack. For example, its data warehouse joins product data with outside sources such as Stripe or Postgres. Likewise, its pipelines move data between more than 145 tools.

In addition, LLM analytics helps teams that ship AI features. Specifically, it tracks traces, costs, and errors. Meanwhile, PostHog AI answers questions and writes SQL for you.

Real Results from Real Teams

Customer stories also provide a feel for the impact. Keep in mind that PostHog publishes these, so treat the numbers as vendor-reported. Still, they are useful. For example, Brainboard reports a 350% rise in conversion rates.

Likewise, Contra says session replays lifted registrations by 30%. In addition, Hasura saw conversion gains of 10 to 20%. Meanwhile, Webshare replaced Mixpanel, Hotjar, and FullStory with one tool.

Try This Combo First

So, pair a feature flag with session replay.

• First, release a new feature to 10% of users.
• Then watch their recordings for confusion or errors.
• Finally, if all looks good, raise the rollout to 100%.

As a result, this simple habit catches problems before most users ever see them.

PostHog Pricing: What the Free Tier Really Covers

What You Get for $0

First, PostHog uses pay-as-you-go pricing with no seat fee. Furthermore, every product has its own free monthly allowance. For example, on the free plan, you get 1 million analytics events, 5,000 session recordings, and 1 million feature flag requests each month.

Third-party reviews also list 1,500 survey responses and 100,000 error exceptions. Plus, nothing needs a credit card, and the limits reset every month. In fact, PostHog says more than 90% of its customers use it for free.

How far does 1 million events go? In other words, that equals about 33,000 events per day. In practice, that often covers 10,000 to 50,000 monthly active users for a young startup, as long as you track only core actions. Thus, many teams run for months at zero cost.

Free monthly allowance	Included on the free plan
Product analytics events	1,000,000
Session recordings	5,000
Feature flag requests	1,000,000
Survey responses (third-party reviews)	1,500
Error exceptions (third-party reviews)	100,000

What You Pay After the Free Tier

Past the free tier, you pay per event, and the rate falls as volume grows. For instance, events from 1 to 2 million cost $0.00005 each. Then, at 250 million and above, the price drops to $0.000009 each.

Here is a quick example. Say you send 10 million anonymous events in a month. The first million is free. The next million costs $50. Then the remaining 8 million cost about $274 at $0.0000343 each. In total, that comes to roughly $324, which is still modest for that volume.

10 million anonymous events	Events	Rate per event	Cost
First tier	1,000,000	Free	$0
Next tier	1,000,000	$0.00005	$50
Remaining events	8,000,000	$0.0000343	about $274
Estimated monthly total	10,000,000		about $324

Keep two details in mind.

• First, identified events, which tie to a known user, cost more than anonymous ones.
• Second, each product bills on its own meter.

So replays, flags, and surveys add their own charges once you pass their free limits.

Smart Ways to Control Spend

Money-Saving Tips:

• First, plan before you scale.
• Next, pick only the events that answer real questions.
• Sample your session recordings instead of saving every one.
• Please review your billing page each week during the first month.
• Finally, remember that prices change.

Therefore, confirm the current rates on the official PostHog pricing page before you commit.

Is PostHog Right for You? Self-Hosting, Alternatives, and Fit

Cloud or Self-Hosted?

PostHog is open source, so you can run it on your servers. This option appeals to teams with strict data rules, since the data stays inside your systems. However, PostHog itself says Cloud is the better choice for most teams.

The open-source version can run with Docker Compose. However, self-hosted setups miss new features that arrive first on the cloud. They also need your people to keep them running. So, unless you have a clear compliance need, start with Cloud.

How It Compares with Other Tools

• First, Google Analytics focuses on marketing traffic.
• Next, Mixpanel and Amplitude focus on pure product analytics.
• PostHog, by contrast, spans analytics, replays, flags, and experiments in one place.

The wide range of features is a significant advantage. On the downside, reviewers say the interface and setup can feel complex for new users. Likewise, G2 reviewers note a steeper learning curve for non-technical or marketing-only teams.

Privacy and Data Control

PostHog lists privacy controls in its docs for both analytics and session replay. Therefore, review them before you launch.

• First, hide sensitive screens, such as payment forms, from recordings.
• Next, update your privacy policy so users know what you track.
• Furthermore, collect only the data you need.

This habit lowers risk, and it keeps your event count and your bill under control.

Who Should Use It, and Who Should Skip It

In general, PostHog fits startups, software-as-a-service teams, and AI product builders that have at least one engineer. However, it is less ideal for marketers who only want simple traffic reports.

Here is a quick test you can run.

• First, pick one question, such as “Where do new users drop off?”
• Next, set up PostHog Cloud in a single day.
• Then track just a handful of events for two weeks.
• Finally, review your funnel and replays.

Overall, if you identify a clear fix, the tool has earned a place in your stack.

Conclusion: Is PostHog Worth Trying?

So, what is PostHog in the end? In short, it is an open-source product analytics platform that puts analytics, session replay, feature flags, experiments, surveys, and data tools under one roof. Because everything connects, you can move from a chart to the real user behind it in a few clicks.

Let’s recap the key points.

• First, PostHog works by capturing events, storing them in a fast database, and letting you explore them with charts or SQL.
• Second, its toolkit covers the full loop of measuring, diagnosing, and releasing changes.
• The free tier is generous, allowing for 1 million events per month, so most small teams can begin without incurring any costs.
• Finally, the platform suits technical teams best, while non-technical users may face a learning curve.

Overall, the main lesson is simple. A connected platform can save both time and money, as tool sprawl is costly. At the same time, discipline matters. Track fewer events, watch your usage, and test one clear question first.

Ready to act? First, sign up for the free plan on PostHog Cloud, add the tracking snippet today, and build one funnel this week. Then watch five session replays. You will likely spot a quick win by the end of the week.

Meanwhile, bookmark the official pricing page so you can check costs as you grow. Start small, learn fast, and let real user data guide your next product decision.

Frequently Asked Questions About PostHog

1. Is PostHog free to use, and what does the free plan actually include?

Yes. PostHog Cloud has a free plan with 1 million analytics events, 5,000 session recordings, and 1 million feature flag requests each month. Furthermore, no credit card is needed, and limits reset monthly. Thereafter, you pay per use.

2. How does PostHog collect user data from a website or a mobile app?

First, PostHog collects data through SDKs and a JavaScript snippet. Then you add the code, turn on auto-capture, and log custom events. Events then flow into its database, ready for funnels, retention charts, and session replay.

3. Can you self-host PostHog, or is the cloud version the better pick?

Yes, the open-source version can run on your servers with Docker. However, PostHog recommends the cloud for most teams because self-hosted setups miss new features and need your own staff for upkeep.

4. How is PostHog different from Google Analytics or Mixpanel tools?

Google Analytics centers on marketing traffic, and Mixpanel centers on product analytics. In contrast, PostHog adds session replay, feature flags, and experiments in one open-source platform, which suits engineering-led teams that want fewer separate tools.

5. Who should use PostHog for product analytics and feature flag testing?

PostHog suits startups, SaaS companies, and AI product teams that have at least one engineer. Furthermore, it works well for funnel analysis, safe feature releases, and A/B testing. Marketing-only teams, on the other hand, might prefer simpler tools.