How We Use PostHog to Track Visitor Behaviour on Our HubSpot App Site
For a while I was tracking the basics: session recordings, daily visitors, UTM sources, countries. I could see traffic was coming in, but I could not answer the questions that actually mattered.
So I set up PostHog to answer three things.
1. Who Is Actually Visiting?
Knowing you had 500 visitors yesterday is nice. Knowing whether those visitors are RevOps managers at mid-market companies or students researching for a school project - that changes everything.
I set up a two-question popover survey with a 30-second delay. It asks: “What is your role?” and “What is your company size?” That data writes straight to visitor profiles in PostHog.
Now I can see whether we are attracting the right people before they ever fill out a contact form. If the mix shifts - say, we start getting more developers and fewer ops managers - I know our content strategy needs adjusting before it shows up in pipeline numbers weeks later.
2. Where Do They Drop Off?
I built conversion funnels for every key journey on the site. From first visit through to form submission, CTA click, or app install.
The first thing I spotted was which steps were losing the most people. One funnel showed that visitors were landing on a product page, scrolling to the pricing section, and then leaving without clicking anything. That is a very different problem from “not enough traffic” - it is a conversion problem, and it told me exactly where to focus.
Session recordings can get you there eventually, but watching 50 recordings to find a pattern takes hours. A funnel shows it in 30 seconds.
3. Which Traffic Sources Actually Drive Action?
This was the biggest insight. There is a real difference between a channel that sends traffic and a channel that sends people who actually do something.
I started breaking down CTA clicks and form submissions by source. Some channels that looked great on a visitor count basis had almost zero conversion. Others that seemed small were consistently producing engaged visitors who clicked through to install or contact.
That kind of breakdown changes where you spend your time. Instead of optimising for impressions, you optimise for the sources that produce real outcomes.
The Setup
Everything lives in a single PostHog dashboard I can check in 30 seconds:
- Visitor profile breakdown by role and company size (from the survey)
- Conversion funnels for each key journey
- CTA and form submission breakdown by traffic source
- Session recordings linked to specific funnel drop-off points
PostHog’s free tier covers everything I described here. The survey, the funnels, the source breakdown - all available without a paid plan.
What I Wish I Had Done Sooner
Honestly, all of it. I spent months looking at surface-level analytics (pageviews, bounce rate, country distribution) when the real questions were about visitor quality, journey friction, and source effectiveness.
If you are running a SaaS site or an app landing page and you are only tracking pageviews, the three questions above are a good place to start. They are simple to set up and immediately useful.