You built something with AI. You deployed it. You shared the link. Now people are visiting your app — and you have no idea what they are doing when they get there. Are they signing up? Are they leaving after three seconds? Is anyone actually using the feature you spent all weekend prompting into existence? Without product analytics, you are guessing. PostHog fixes that, and you do not need to be a data analyst to use it.
This tutorial is specifically for non-technical founders and vibe coders who have shipped an MVP and need to understand what happens next. We are going to walk through installing PostHog, understanding what it captures automatically, building your first funnel, and knowing what to look for when your first users arrive.
Why PostHog Over Google Analytics?
You probably already have Google Analytics on your site (we recommend setting it up as part of shipping your project). So why add PostHog?
Google Analytics tells you how many people visited and where they came from. PostHog tells you what they did when they arrived. Google Analytics is built for marketing websites. PostHog is built for products — apps where users sign up, click buttons, create things, and (hopefully) come back.
The difference matters when you have an interactive application rather than a static website. GA4 can tell you that 200 people visited your landing page. PostHog can tell you that 40 of them clicked "Sign Up," 25 completed the signup flow, 12 created their first project, and 3 came back the next day. That level of detail is what helps you decide what to build next.
PostHog also has a generous free tier: 1 million events per month, which is more than enough for any MVP. No credit card required to start.
Installing PostHog: The Two-Minute Version
PostHog installation is one of the simplest things you will do. You add a snippet of JavaScript to your app, and it starts tracking automatically. Here is how.
Step 1: Create an account. Go to posthog.com and sign up. Choose the cloud option (PostHog Cloud US or EU). You will get a project API key immediately.
Step 2: Add the snippet. PostHog gives you a JavaScript snippet that looks like this:
<script>
!function(t,e){var o,n,p,r;e.__SV||(window.posthog=e,
e._i=[],e.init=function(i,s,a){...},
e.init('phc_YOUR_PROJECT_KEY', {api_host: 'https://us.i.posthog.com'}))
}(document,window.posthog||[]);
</script>You paste this into the <head> of your HTML, or if you are using a framework like Next.js or React, you add it to your layout component. If you are using Cursor or another AI coding assistant, you can literally tell it: "Add the PostHog tracking snippet to my app" and paste in the code PostHog gives you. The AI will know where to put it.
Step 3: Verify. Visit your app in a browser, then check the PostHog dashboard. You should see your own visit appear within a few seconds. If it shows up, you are done with installation.
That is genuinely it. Three steps, two minutes. PostHog starts capturing data immediately.
Autocapture: What PostHog Tracks Without You Doing Anything
The most important PostHog feature for vibe coders is autocapture. Once installed, PostHog automatically records:
- Page views — every page a user visits, with timestamps
- Clicks — every button, link, and interactive element a user clicks
- Form submissions — when users submit forms
- Page leave events — when users navigate away
- Session recordings — optional, but PostHog can record actual user sessions as video replays
This means you do not need to write any tracking code yourself. PostHog captures the data first, and you decide what to do with it later. This is the opposite of how analytics traditionally worked, where you had to plan your tracking in advance. For vibe coders who are iterating quickly, autocapture is perfect.
One thing to know: autocapture identifies elements by their CSS selectors and text content. If your AI-generated code uses generic class names (which it often does), the event names in PostHog might look like "Clicked button.btn.btn-primary" rather than "Clicked Sign Up." You can clean this up later with PostHog's event labeling, but it is worth being aware of.
Your First Dashboard: What to Look At
When you log into PostHog after a few days of data collection, the interface can feel overwhelming. There are a lot of features. Ignore most of them for now. Here is where to focus.
Activity tab: This shows a real-time feed of events. Watch it for a few minutes while someone uses your app. You will see every click, every page view, every action in chronological order. This alone gives you more insight than most analytics tools.
Web analytics: PostHog has a built-in web analytics dashboard that shows page views, unique visitors, bounce rate, and traffic sources. This overlaps with Google Analytics but is useful to have in one place. Check this weekly to understand your traffic trends.
Session recordings: Turn this on. Then watch three to five user sessions. You will learn more about your app in 15 minutes of watching real users than in a week of looking at charts. You will see where they hesitate, where they get confused, and where they give up. This is the single most valuable PostHog feature for early-stage products.
Building Your First Funnel
A funnel shows you the steps users take toward a goal and where they drop off. For a typical vibe-coded MVP, your funnel might be:
- Visited landing page
- Clicked "Sign Up"
- Completed registration
- Performed the core action (created a project, sent a message, uploaded a file — whatever your app does)
To build this in PostHog:
- Go to the Funnels section (under Product Analytics)
- Click "New funnel"
- Add your first step: select the "$pageview" event and filter by URL containing your landing page path
- Add subsequent steps using the autocaptured click events. PostHog will show you a dropdown of events it has already captured.
- Set the time window (how long users have to complete the funnel — start with 24 hours)
- Click "Calculate"
PostHog will show you a visualization of how many users completed each step and what percentage dropped off at each point. This is where the real insights live.
If 100 people visit your landing page and only 5 click "Sign Up," your landing page messaging needs work. If 50 click "Sign Up" but only 10 complete registration, your signup form is the problem. If 10 complete registration but none perform the core action, your onboarding is broken. The funnel tells you exactly where to focus.
Custom Events: When Autocapture Is Not Enough
Autocapture handles most tracking, but sometimes you need to track something specific that autocapture does not catch — like when a user successfully generates an AI response, completes a multi-step process, or reaches a specific state in your app.
For these cases, you add custom events. The code is simple:
posthog.capture('project_created', {
project_type: 'landing-page',
template_used: true
});Tell your AI coding assistant: "Add a PostHog custom event called 'project_created' that fires when a user successfully creates a new project. Include the project type as a property." The AI will place the tracking call in the right location in your code. You do not need to understand the code structure yourself.
Do not go overboard with custom events early on. Start with autocapture, identify the gaps after a week of data, and then add custom events for the specific moments that matter most. Three to five custom events is plenty for an MVP.
What to Look for When Your First Users Arrive
The first 100 users of your vibe-coded app will teach you more than any amount of planning. Here is what to watch for in PostHog during that critical period.
Bounce rate on your landing page. If more than 70% of visitors leave without clicking anything, your value proposition is unclear or your page loads too slowly. Check the prototype to production guide for performance tips.
Signup completion rate. Industry average for SaaS signup forms is around 20-40%. If yours is below 15%, simplify the form. Every field you remove increases completion. Consider using social login (Google, GitHub) instead of email/password.
Time to first value. This is the time between when a user signs up and when they first do the thing your app is designed for. If this is longer than 5 minutes for a simple MVP, you need better onboarding. Watch the session recordings to see exactly where people get stuck.
Retention. PostHog has a retention table that shows what percentage of users come back on day 1, day 3, day 7, and so on. For an MVP, do not panic about low retention yet — but if literally zero users come back after day 1, that tells you something important about your product's stickiness.
Error patterns. Watch for users who click the same button multiple times in a row or navigate back and forth between the same pages. This usually indicates something is broken or confusing. Session recordings will confirm it.
PostHog's Free Tier: What You Get
PostHog Cloud's free tier includes:
- 1 million events per month
- 5,000 session recordings per month
- 1 million feature flag requests per month
- Unlimited team members
- No time limit — the free tier does not expire
For an MVP with fewer than a few thousand users, you will not come close to these limits. PostHog will be free for months, possibly years, depending on your growth rate. If you do outgrow the free tier, pricing starts at usage-based rates that are competitive with alternatives. Compare it against other options in our analytics tools roundup.
Common Mistakes to Avoid
Do not look at analytics before you have enough data. Checking PostHog when you have had 12 visitors is a recipe for anxiety, not insight. Wait until you have at least 100 sessions before drawing conclusions from funnels or retention data.
Do not track everything from day one. Autocapture is sufficient for the first few weeks. Adding 30 custom events before you have users creates noise, not signal. Start simple, add tracking as specific questions arise.
Do not ignore session recordings. Charts and funnels are useful, but watching five real user sessions will give you insights that no dashboard can. Make it a weekly habit: watch five recordings every Monday. You will always find something you did not expect.
Do not confuse page views with value. A user who visits 20 pages but never signs up is not a success. Focus on the events that indicate real engagement with your product, not vanity metrics like total page views.
PostHog vs Alternatives: Quick Comparison
| Feature | PostHog | Google Analytics | Plausible | Umami |
|---|---|---|---|---|
| Best for | Product analytics | Marketing analytics | Privacy-first web stats | Self-hosted web stats |
| Session recordings | Yes (free tier) | No | No | No |
| Funnels | Yes | Yes (limited) | Basic | No |
| Autocapture | Yes | Limited | No | No |
| Free tier events | 1M/month | Unlimited | No free tier | Unlimited (self-host) |
| Feature flags | Yes | No | No | No |
| Ideal user | App builders | Content sites | Privacy-conscious sites | Developers who self-host |
For most vibe coders building interactive apps, the recommendation is straightforward: use Google Analytics for your marketing pages and PostHog for your product. They complement each other and both have free tiers.
Next Steps
Once you have PostHog installed and have collected a week of data, here is what to do:
- Watch 10 session recordings and write down every point where users seem confused
- Build your primary funnel (landing page to core action) and identify the biggest drop-off
- Fix the biggest drop-off point and watch the funnel numbers change over the following week
- Repeat
This simple loop — observe, identify the problem, fix, observe again — is the foundation of product-led growth. PostHog gives you the observation layer. Your AI coding assistant gives you the ability to fix things quickly. Together, they let you iterate on your product faster than traditional teams can schedule a meeting about it.
For a broader view of monitoring and analytics options, explore our analytics and monitoring tools page. And if you are still building your MVP, start with our build your first app guide — it covers setting up the foundation that PostHog will help you measure.