Most marketers open Google Analytics, see a wall of numbers, and close it again. The dashboard looks busy, but the actual signal - what is working, what is wasting budget, what needs fixing - is buried under terminology that was never designed to be intuitive.
This post cuts through that. You will learn what each major metric actually measures, how the different traffic reports connect to each other, and how to use the data to make real decisions instead of just staring at graphs.
The Difference Between Users, Sessions, and Pageviews
These three numbers sit at the top of almost every GA4 report, and they are not interchangeable.
A user is a person (technically: a browser or device identifier) who visited your site. GA4 tracks four user types:
- Total users: Everyone who triggered any event during the date range
- Active users: People who had an engaged session, meaning they stayed at least 10 seconds, viewed two or more pages, or converted
- New users: People visiting for the first time (tracked via the
first_visitevent) - Returning users: People who visited at least once before the current date range
The number GA4 shows in most standard reports is active users, not total users. This is important: if you have 10,000 total users but only 6,000 active users, that 4,000-person gap represents people who landed and immediately left without engaging at all.
A session is a single visit. One user can have multiple sessions. If someone visits your site on Monday morning, leaves, and returns Monday afternoon, that is two sessions from one user. By default, GA4 ends a session after 30 minutes of inactivity.
A pageview (called screen_page_views in GA4) counts every page load. A single user in a single session could generate 10 pageviews by clicking through your site. Pageviews are a measure of content consumption, not audience size.
Why this matters in practice: If your sessions are climbing but users are flat, your existing audience is visiting more often - which could be great (loyal readers, returning shoppers) or a warning sign (people returning because something is broken). If users climb but sessions stay flat, you are reaching more people but they are not coming back.
Traffic Channels: The Map of Where Visitors Come From
The Traffic Acquisition report in GA4 classifies your traffic into channels. These are not arbitrary buckets - each one tells you something specific about your marketing.
Organic Search
Visitors who found you through unpaid search results on Google, Bing, or other search engines. This is your SEO traffic. GA4 identifies it when the traffic source matches a known search engine and the medium is organic.
High organic traffic means your content is ranking. Low organic traffic for a site that has been publishing for months usually points to a technical SEO issue, thin content, or targeting keywords you cannot realistically compete for.
Paid Search
Traffic from search ads (Google Ads, Bing Ads). GA4 classifies it as Paid Search when the source is a known search engine and the medium is cpc or ppc. This traffic should be tagged with UTM parameters in your ad URLs to ensure clean attribution.
If you are running Google Ads but see very little Paid Search traffic in GA4, check: (1) your auto-tagging settings in Google Ads, (2) whether your UTM parameters are correctly formatted, and (3) whether your GA4 property is linked to your Google Ads account.
Direct
Direct traffic is often misunderstood. It does not mean people typed your URL into their browser. It means GA4 could not identify where the visit came from. Common causes include email newsletters without UTM tracking, PDF downloads with links, mobile apps, HTTPS-to-HTTP referral stripping, and dark social sharing (Slack, WhatsApp, private messaging).
If your Direct traffic is unusually high - say, above 20-25% of total sessions for a site with active paid and email channels - you almost certainly have a tracking gap. The fix is applying UTM parameters consistently to every link you control.
Organic Social
Unpaid clicks from social platforms like Facebook, Instagram, LinkedIn, and X. GA4 recognizes these when the source matches a known social site and there is no paid medium identifier.
For most B2B companies, organic social drives a small fraction of traffic but can drive very high-quality leads when the content matches what your audience is searching for. Compare engagement rate and conversion rate from organic social against other channels before writing it off as a vanity channel.
Paid Social
Clicks from paid social ads. Requires proper UTM tagging or native integrations (Facebook's Meta Pixel with GA4 integration, for example) to attribute correctly.
Referral
Traffic from other websites that link to you - blogs, directories, media coverage, partner sites. This is often undervalued. Referral traffic tends to come in pre-qualified because the linking site's audience already has context about who you are.
Traffic from email campaigns. GA4 does not automatically recognize email as a channel - you have to add utm_medium=email to every link in every email you send. Without this, email traffic disappears into Direct.
How to Read the Traffic Acquisition Report
Open GA4, go to Reports > Life cycle > Acquisition > Traffic acquisition.
The default primary dimension is Session default channel group, which is the broadest view. Change it to Session source / medium for a more specific breakdown - this shows exactly which platform + medium combination is driving traffic (e.g., google / organic, linkedin.com / referral, newsletter_march / email).
The columns you need to understand:
Sort by Engaged sessions rather than Sessions. This filters out bot traffic and accidental one-second clicks that add no value.
Engagement Rate vs. Bounce Rate
GA4 replaced the old bounce rate metric with engagement rate. They are inverse of each other - an engagement rate of 65% means a bounce rate of 35%.
An engaged session requires at least one of:
- The visitor stayed for 10+ seconds
- They viewed 2 or more pages
- A conversion event fired
A 10-second threshold is low. Someone who reads one paragraph and leaves after 12 seconds counts as engaged. This means engagement rate is an optimistic metric. Do not treat 70% engagement rate as proof that your content is resonating - look at average engagement time per session alongside it.
Benchmarks to work from: For blog content, average engagement time above 2 minutes per session is healthy. For landing pages, 45-90 seconds is typical for people who convert. If your top organic landing pages have engagement time under 30 seconds, the content is not matching what searchers expected to find.
User Acquisition vs. Traffic Acquisition
These are two different reports and they answer different questions.
Traffic Acquisition tracks sessions - it credits whichever source drove a specific visit. If someone clicks a Google ad on Tuesday and then returns via direct on Thursday and converts, the conversion gets credited to the direct session (or Google, depending on attribution settings).
User Acquisition tracks users - it credits whichever source brought the user to your site for the first time. That same conversion would be credited to Google Ads because that was the first-touch source.
Use Traffic Acquisition for day-to-day performance monitoring. Use User Acquisition to understand which channels are actually growing your audience and acquiring new customers over time.
Finding Which Sources Drive Conversions
The Key events column in the Traffic Acquisition report shows conversions by source. But this data is only useful if you have actually configured key events in GA4.
To check: go to Admin > Data display > Events. Any event with a star icon is marked as a key event. If you are a SaaS company and generate_lead or sign_up is not in that list, your conversion data is incomplete.
Once key events are set up, the analysis becomes straightforward:
- Sort the Traffic Acquisition report by Key events (descending)
- Compare the Session key event rate across sources - not just the raw count
- A source with 500 sessions and 25 conversions (5% rate) outperforms a source with 2,000 sessions and 40 conversions (2% rate) on a quality basis
A source with high traffic but a conversion rate significantly below your site average is worth investigating. Either the audience is wrong (bad targeting), the landing page is wrong (mismatch between ad promise and page), or the offer is wrong (what you are asking people to do does not match where they are in the buying process).
Sessions vs. Users: Which Number to Report
The right answer depends on what the stakeholder wants to know.
Report users when the question is about audience size: "How many people visited our site this month?" Report sessions when the question is about activity: "How much traffic did we generate?" Report new users when the question is about growth: "Are we reaching new people?"
Where this gets confusing: GA4's default "Users" metric in many reports is actually active users, which is lower than total users. If your number does not match what someone else sees exported from GA4, this discrepancy is usually the reason.
The Direct Traffic Problem
Direct is the black hole of analytics. Industry data from multiple sources consistently shows that 15-30% of web traffic cannot be properly attributed, and a large chunk of that ends up in Direct.
The biggest untracked channel for most B2B companies is email. If you send newsletters without UTM parameters, every click lands in Direct. For a company sending 50,000 emails monthly with a 2% click rate, that is 1,000 sessions per campaign that disappear from your channel reports.
Other common sources of inflated Direct traffic:
- Mobile apps that strip referrer data
- Links shared in Slack, WhatsApp, and Telegram (no referrer passed)
- PDFs and PowerPoint files with embedded links
- Redirects that strip UTM parameters before GA4 fires
The fix is systematic UTM tagging. Every link in every email, every link in every slide deck, every link on every document you distribute externally should carry UTM parameters.
How to Diagnose a Traffic Drop
When organic traffic drops, the diagnostic path in GA4 is:
- Check the Traffic Acquisition report filtered to Organic Search - confirm the drop is real and not a date range issue
- Look at which landing pages lost traffic (add Landing page as a secondary dimension)
- Cross-reference with Google Search Console to see if impressions dropped (ranking issue) or CTR dropped (title/description issue)
- Check if engagement metrics changed on the affected pages - a drop in engagement time alongside a traffic drop can indicate a quality signal to Google
When paid traffic drops, check:
- Budget exhaustion (campaigns paused by budget)
- Quality Score drops in Google Ads
- Bid competition changes (cost-per-click up, same budget = fewer clicks)
- Ad approval issues
Automating Your GA4 Reporting
Manually checking GA4 every day creates a reporting lag. Issues discovered on Friday about what happened on Monday cost you a week of wasted spend. Automated daily reporting solves this by surfacing anomalies as they happen.
With Toffu, you can connect your Google Analytics account and set up natural language queries that run on a schedule. Instead of logging into GA4 and navigating through reports, you ask: "Which channels had the biggest drop in engagement rate this week compared to last week?" and get a structured answer with the specific numbers.
This is particularly useful for marketing analytics across multiple channels - when you are tracking organic search, paid campaigns, and referral traffic simultaneously, automated alerts catch the drops before they become expensive problems.
If you are managing reporting across multiple campaigns or clients, audience performance reporting automation lets you set up recurring reports that flag underperforming segments automatically - removing the manual work of checking each channel individually every week.
Key Metrics Cheat Sheet
For quick reference, here is what each metric should prompt you to do:
High Direct traffic (20%+): Audit your UTM tagging across email, social, and documents. You have a tracking gap.
Low engagement rate on organic landing pages (under 50%): The page content does not match searcher intent. Revise the headline, opening paragraph, and content structure.
High traffic, low Key event rate from a specific source: Either the audience is wrong, the landing page does not convert, or the offer is misaligned with buyer stage.
Engagement time under 60 seconds on blog posts: Content is not holding attention. Consider shorter paragraphs, better formatting, or a stronger opening hook.
New users flat while sessions grow: Existing audience is returning more. Good for retention, bad if growth is the goal. Invest in top-of-funnel channels.
Session key event rate 3x higher from one source vs. others: Scale that source. The audience quality is better there.
Connecting GA4 to Actual Marketing Decisions
The purpose of traffic analysis is not to produce reports - it is to make better decisions about where to spend time and budget.
A practical weekly GA4 review takes about 15 minutes:
- Check Traffic Acquisition: Are sessions and engaged sessions up or down vs. last week? Which channels moved the most?
- Check Key events by source: Which channels are driving conversions? Has anything dropped?
- Check average engagement time on top landing pages: Is content quality holding up?
- Flag anything unusual for deeper investigation
That is it. You do not need to analyze every dimension and metric in GA4 every week. You need to know what changed and why, then act on it.
For teams that want a more automated version of this workflow, Toffu connects directly to Google Analytics and can run this analysis on a schedule, flagging anomalies and surfacing the specific numbers that need attention. See how it works on the pricing page or start with a free account to connect your GA4 property and run your first automated analysis.
Common GA4 Misconceptions
"More traffic is always better." Not true. Traffic from the wrong audience drives up your bounce rate, wastes ad spend, and can hurt Quality Scores in Google Ads. 500 highly-engaged sessions from qualified visitors outperform 5,000 unengaged sessions from broad targeting.
"Direct traffic means brand awareness is working." Sometimes. But often it means your tracking is broken. Do not celebrate high Direct traffic until you have confirmed your UTM coverage is comprehensive.
"Organic traffic is free." Organic traffic costs time - writing, editing, building links, managing technical SEO. The cost is just less visible than paid. For most companies, the cost per organic visit is measurable when you factor in content production and SEO tooling.
"GA4 numbers match Google Ads numbers." They will not match exactly. GA4 uses last-click attribution by default (configurable), while Google Ads uses data-driven attribution. The session counts differ because GA4 counts a session as starting on landing, while Google Ads counts a click at the ad level. Expect 5-15% discrepancies - anything larger warrants investigation.
"Engagement rate means users are happy." Engagement rate only measures minimum interaction thresholds. A user who spent 12 seconds arguing in the comments section counts as engaged. Use average engagement time and conversion rate together with engagement rate for a real picture of content quality.
Conclusion
Google Analytics traffic data is only valuable when you understand what each number actually represents. Sessions are not users. Direct is not branded intent. Engagement rate is not satisfaction. Once you know what you are looking at, GA4 becomes a powerful diagnostic tool - not just a collection of graphs that go up and down.
The analysis framework is consistent: look at which channels are driving traffic, check which ones are converting, diagnose the gaps, and fix them. Then automate the monitoring so you are not doing this manually every week.
If you want to skip the manual dashboard-checking entirely, Toffu connects to your Google Analytics property and lets you ask questions about your data in plain language - getting the same insights without the navigation overhead.