Most teams split their stack in two: web analytics for the marketing site, PostHog product analytics for the app. It feels logical. It’s actually costing you attribution, audience intelligence, and conversion context you’ll never get back.
Here’s the argument: product analytics should replace web analytics entirely — for most brands, in 2026, without exception.
This isn’t a theoretical take. It’s built from analyzing how ChatGPT, Figma, Salesforce, and yes, PostHog itself handle tracking — and from running live PostHog product analytics data on the VisionLabs website.
Is the line between web analytics and product analytics actually real?
Think about what most websites actually contain: a homepage, a pricing page, a blog, comparison pages, and an app with feature usage baked in.
Teams look at that list and draw a line. Homepage, pricing, blog = web analytics. App = product analytics.
That line is artificial. And it’s getting more artificial every year.
Apps now include pricing pages. Blogs live inside logged-in dashboards. Homepages render conditionally depending on whether you’re signed in. The separation you’re managing in your toolstack doesn’t reflect how users actually move.
There’s only one distinction that actually matters: logged in vs. logged out.
And even that’s not the whole picture. Because here’s what’s really happening the moment someone lands on your site.
You’re Building a User Profile Before Anyone Signs Up
The second someone hits your homepage, you can start tracking them. Anonymous user ID. Page views. UTM source. Geographic region. Content consumed. Time on page.
That is PostHog product analytics. Not web analytics. Product analytics.
You’re creating a user profile before that person has typed a single character into a sign-up form. The question is whether you’re capturing it or throwing it away.
Most teams throw it away. They track it in Google Analytics, segment it in a separate tool, and then start “real” tracking when someone creates an account. By that point, you’ve lost the full story.
The user journey doesn’t start at sign-up. It continues from sign-up.
This is the core insight. PostHog’s person profiles documentation shows exactly how anonymous users get identified and linked to accounts. Our PostHog Complete Course 2026 walks through the full setup if you want to see how this works in practice.
Five Real Brands, One Framework
Should ChatGPT’s UTM tracking follow users into the product?
Search “ChatGPT” on Google. You’ll likely hit a paid ad. Click it and paste the URL into a parameter parser — JJ used one live during the analysis to break down every parameter — and you’ll see: utm_source=google, utm_medium=paid_search, utm_campaign=chatgpt.
You’re in the product immediately. There’s no separate marketing site experience to speak of. That UTM data needs to follow you into the product, attach to your account, and persist through your usage history.
A few things worth noting in their UTM setup: the medium should be cpc per Google’s conventions, not paid_search. They’re using a non-standard custom parameter for campaign ID instead of utm_id. And they’re correctly passing gbraid and gclid for Google’s click tracking. PostHog’s UTM tracking documentation covers the standard parameters and how to capture them correctly. Not perfect, but instructive.
The question is simple: do you want your product to know someone came from a Google ad when they sign up? Yes. That’s PostHog product analytics. Not web analytics.
For teams connecting ad attribution to downstream product behavior, our guide on connecting your CRM to PostHog for ads covers how to carry that data through.
What does product analytics look like for a services company with no app?
VisionLabs is a services company. There’s no app to sign into. The website itself is the product experience. Content builds authority. The goal is a contact form fill.
So is the VisionLabs website “web analytics” territory? No.
Every piece of content someone reads, every page they visit, every form they abandon — that’s behavioral data you can use to understand your pipeline. We’ll come back to the live PostHog data from this site in a moment.
How should a PLG company like Figma track the full user journey in PostHog?
Figma lets you sign up and start using the product without talking to anyone. Trust comes from using the tool, not from a sales call. The journey is: Google ad → homepage → pricing → sign up → feature usage.
Every step of that journey should be tracked in one system. Cohort the users who came from a specific campaign, see how many reached a key feature milestone, correlate with retention. That’s PostHog product analytics covering what most teams would call “web” and “product” separately.
Our funnels guide shows exactly how to build that kind of multi-step journey analysis in PostHog.
Why do enterprise companies like Salesforce need product analytics on their marketing site?
You know something is really enterprise when they put a phone number on the website. Go to Salesforce. No pricing. No self-serve. Just a phone number, case studies, and a lot of content. Nobody self-serves a $200K contract.
But that content and brand experience is doing real selling work. In 2026, the website is the product for enterprise companies just as much as a SaaS app is the product for PLG companies. Every interaction a prospect has with Salesforce’s content is data. Tracking it as “just web analytics” misses the point entirely.
Does PostHog itself track its own marketing site as product analytics?
PostHog’s own website is treated internally as a web analytics property. But look at how their site actually works: the pricing page is accessible while you’re logged in, inside the product. Documentation, comparisons, and feature pages blur into the app experience constantly.
JJ’s argument: PostHog’s own marketing site should be tracked with PostHog product analytics. Because users move between the logged-in product and the marketing site without ever feeling a hard boundary. Your tracking stack shouldn’t create one either.
PostHog even documents web analytics as a feature — but the case here is that product analytics covers everything web analytics does, and more.
What does product analytics look like on a real website?
This isn’t theoretical. Here’s what PostHog showed on the VisionLabs website over 30 days:
- 50 contact form applications tracked in 30 days
- Real-time activity feed flowing with anonymous events as they happen — page views, UTM sources, geographic data, content interactions, all streaming live
- Anonymous user example: A user viewed the PostHog cohorts guide, came in through a direct source, located in North America
That real-time feed is what makes the argument concrete. You’re not waiting for a report. You’re watching your pipeline build in real time — anonymous users reading your content, moving between pages, dropping signals before they ever fill out a form.
That anonymous user hasn’t filled out a form. They might tomorrow. Or next week. With PostHog product analytics, you can build a cohort around everyone who viewed your PostHog content and then filter by who eventually filled out a contact form. That’s the answer to “how well is our PostHog content driving pipeline?”
Without product analytics covering the full pre-signup journey, you can’t answer that question. You just see form fills with no behavioral context attached.
This is why we use PostHog across the entire VisionLabs site — not just for a subset of tracked pages. To explore how to build the dashboards that surface these insights clearly, the PostHog dashboards guide walks through the setup step by step.
How do you set up PostHog to track the full user journey?
Start with anonymous user tracking
PostHog assigns an anonymous ID the moment someone lands on your site. Don’t wait for sign-up to start capturing behavior. Every pageview, every content interaction, every UTM source from day one. PostHog’s identify documentation explains how this identification process works under the hood — and how anonymous events get merged into the user record at sign-up.
How does PostHog connect anonymous session history to a signed-in user?
When someone signs up, PostHog links the anonymous session history to the new account. The full pre-signup journey becomes part of the user profile. You can now see that a user who converted to a paying customer spent three sessions reading your comparison content before signing up.
This is PostHog functioning as a customer data platform, not just a product analytics tool.
How do you build cohorts around content consumption in PostHog?
Group users by what content they consumed before converting. Did users who read your pricing page convert at a higher rate than users who came through the blog? Did a specific campaign drive users with better 30-day retention? Cohort analysis answers these questions. See the full cohorts and funnels guide to build these segments in PostHog.
When should you use session recordings instead of funnel data?
A funnel tells you where users drop off. Session recordings tell you why. Both are product analytics. Both apply to your marketing site, not just your app.
How do you set up PostHog alerts to monitor product metrics in real time?
If contact form fills drop below a threshold, you want to know immediately, not at the end of the month. PostHog alerts let you monitor product metrics in real time so you can act on problems before they compound.
Can you automate follow-up in PostHog based on what users do before they sign up?
A user reads three pieces of content about a specific feature. That’s a signal. PostHog workflows let you trigger actions based on that kind of behavioral data, so your follow-up is timed to intent rather than arbitrary drip sequences.
Which companies should replace web analytics with PostHog product analytics?
PLG companies (Figma, ChatGPT, PostHog): You’re already doing product analytics post-signup. The gap is pre-signup. Capture the full journey from first ad click to first feature activation so you can improve both acquisition and activation in one system.
Sales-led companies (VisionLabs, Salesforce): Your website is your product. Every piece of content is a touchpoint that either builds or loses trust. Track it with PostHog product analytics so you can connect content consumption to pipeline, not just traffic.
Hybrid companies: You have a marketing site and a product. The transition between the two is where data goes to die. Stitch the journey together with PostHog so no conversion context gets lost at the handoff.
Why is PostHog better than traditional web analytics tools?
Traditional web analytics tools answer one question: how much traffic did this page get?
That’s not a useless question. It’s just not a useful one. Not when you’re trying to understand what actually drives revenue.
PostHog product analytics answers the questions that matter:
- Which content did users consume before they contacted us?
- Do users who came from Google Ads retain better than users who came from organic?
- At what point in the journey do high-value users typically convert?
- What does the session look like for users who dropped off at pricing?
You get this because PostHog lets you track behavior, build cohorts, run funnels, replay sessions, and automate responses — all in one system, across the full pre-signup and post-signup journey.
To get started with the tool itself, the PostHog demo for 2026 is the fastest way to see it in action. For teams sending conversion events back to paid channels, the guide on sending conversion events to Meta shows how to close the loop on attribution.
Stop Drawing a Line That Doesn’t Exist
Web analytics and product analytics aren’t two categories. They’re one continuous tracking problem.
The user who clicked your Google ad, read your comparison page, came back two days later through organic search, spent 12 minutes on your pricing page, and then filled out a contact form — that person left a complete trail. The question is whether your stack is built to capture it.
If you’re still running Google Analytics for the marketing site and PostHog product analytics only inside the app, you’re seeing half the journey and making decisions with half the data.
Product analytics covers the whole thing. Run it that way.
Want help implementing PostHog product analytics across your full user journey?
VisionLabs implements PostHog for teams who want to track every touchpoint from first click to closed deal — anonymous users, signed-in users, ad attribution, cohorts, and everything in between.
See our PostHog consulting and implementation services →
If you want to build this yourself, start with the PostHog Complete Course 2026 — it covers the full setup from install to advanced analytics.