When you open Google Analytics 4 and look at your client's dashboard, two numbers jump out: sessions and users. One is always greater than the other. Your client asks which to look at. And sometimes, you're not entirely sure yourself.
Here's a clear explanation, with no jargon and concrete examples.
What a user is in GA4
An active user in GA4 is a person who visited the site and had at least one engagement (click, scroll, duration > 10 seconds, conversion event).
GA4 distinguishes two sub-types:
- New users: visit the site for the first time (cookie absent or Google ID absent)
- Returning users: have already visited, recognised via the Google Analytics ID
The same user can generate multiple sessions. If Mark visits the site on Monday and comes back on Thursday, GA4 counts 1 user but 2 sessions.
What a session is in GA4
A session is a period of continuous activity by a user on the site.
It starts at the first hit and ends after 30 minutes of inactivity (or at midnight, or if the traffic source changes).
Example: Marie arrives on your site from Google, reads an article for 8 minutes, closes the tab. She reopens the site 35 minutes later. GA4 counts 2 sessions for 1 user.
Sessions in GA4 vs Universal Analytics
Universal Analytics counted sessions differently: a session expired automatically at midnight and on a source change. GA4 simplified this behaviour, but the default 30-minute inactivity rule remains.
Why sessions are always greater than users
By definition, a user can have multiple sessions: but a session belongs to a single user only.
So: sessions ≥ users, always.
The sessions/users ratio gives you a loyalty indicator. If a site has 1,000 users and 3,200 sessions, the ratio is 3.2: on average, each visitor returns 3 times. Excellent for a blog or an app. Normal for an e-commerce site.
Which one to look at depending on the context
To measure real reach
Look at active users. It's the number of distinct people the site reached over the period. The metric to present to a client who wants to know "how many people saw my site".
To measure activity and engagement
Look at sessions. They reflect the intensity of site usage. A spike in sessions without an increase in users means existing visitors are coming back more: a sign of engaging content or a successful email campaign.
To measure SEO performance
Look at new users from the "Organic Search" channel. It's the only number that measures the real impact of SEO: how many unknown people found the site via Google.
To measure conversions
Neither in isolation. A conversion can come after multiple sessions from the same user. GA4 lets you see the conversion path (attribution) to understand which sessions contributed to the purchase or contact request.
What GA4 changed: engagement rate replaces bounce rate
Universal Analytics had bounce rate: the percentage of sessions with a single page view and no interaction. It was often misleading: a user reading a 2,000-word article for 8 minutes and leaving was counted as a "bounce".
GA4 replaced that with engagement rate: the percentage of engaged sessions (duration > 10s, or 2 page views, or a conversion event). A much fairer measure of real traffic quality.
What you should present to your client
For the vast majority of clients, here's the simplification to apply in a report:
| You say | You show |
|---|---|
| "How many people visited your site" | Active users |
| "How many times your site was visited" | Sessions |
| "Are visitors engaged?" | Engagement rate (not bounce rate) |
| "Is SEO working?" | New users / Organic Search |
Avoid presenting sessions and users side by side without explanation: the client will inevitably ask why the two numbers are different, and the answer takes 5 minutes to explain in a meeting.
How cookie consent impacts your numbers
In the EU, the GDPR requires a consent banner. If a visitor declines cookies, GA4 cannot identify them: they are absent from your reports.
Concretely: a site with 30% consent refusal sees its GA4 data underestimate 30% of its real traffic. Users are more affected than sessions: GA4 can record a partial anonymous session via Consent Mode v2, but cannot identify the user.
What this changes in a client report:
- GA4 numbers are always conservative, never to be overestimated
- If a client compares GA4 to their CMS or Hotjar, the gaps often come from consent, not a tracking issue
- You can add a footer to the report: "GA4 data: visitors who accepted analytics tracking"
Cross-device tracking: why 1 person ≠ always 1 user
Without advanced configuration, GA4 recognises users by cookie. If Mark visits the site from his iPhone in the morning and his MacBook in the evening without logging in: 2 users in GA4, but 1 real person.
GA4 can fix this via two mechanisms:
- Google Signals: if the user is signed in to a Google account with ad personalisation enabled, GA4 can recognise the same person across devices. Enabled in GA4 settings.
- User-ID: if your site has a login system, you pass an identifier directly to GA4. The most reliable and accurate method.
For most freelance clients' showcase websites, neither configuration is in place. So your user counts underestimate the cross-device reality. It's not a bug: it's a limit inherent to cookie-based measurement that you should mention if the client asks.
Summary
- Users = how many distinct people visited
- Sessions = how many times the site was viewed
- Sessions > Users, always
- The sessions/users ratio measures loyalty
- For SEO: look at new users from Organic Search
- Forget bounce rate, use engagement rate
Going further
Sessions and users are the basics. To understand the whole GA4 report, here are the complementary metrics to master:
- GA4 session duration: the most misinterpreted quality metric, with benchmarks by industry.
- GA4 engagement rate: the new metric that replaced the bounce rate. What it measures exactly and how to improve it.
- GA4 page views: the content-side metric that complements sessions.
- GA4 landing pages: where visitors really enter (and why it's not always the homepage).
- Mobile vs Desktop GA4: the device split, essential in any client report.
If you generate GA4 reports for multiple clients, these definitions are the foundation of any readable report. NarratIQ automates that summary and presents the metrics that matter in a PDF your client can read with no training.