← Back to desktop
Tracking

GA4 Reports That Actually Matter for Marketers

NUVIX · 16 July 2026 · 9 min read
TLDR: GA4 ships with dozens of reports and most marketers need about six of them: traffic acquisition, key events, landing pages, user acquisition, monetisation (if you sell online), and attribution paths. Add one custom Exploration crossing landing page, channel, and key events, and you can answer almost every question leadership actually asks. Ignore engagement rate obsession and real-time dashboards, run a 20-minute monthly routine, and stop panicking when GA4 doesn’t match your ad platform numbers. It never will, and that’s fine.

Why GA4 Feels So Overwhelming

Universal Analytics was built around sessions. Someone visited, looked at pages, maybe converted. The reports matched how marketers think: traffic in, conversions out.

GA4 is built around events. Every page view, scroll, click, video play, and form submission is an event with parameters attached. It’s a more flexible model and a far worse default experience. The interface gives you Reports, Explore, and Advertising sections, each with its own logic, and the standard reports bury the useful numbers under engagement metrics nobody asked for.

The result: marketing managers open GA4, click around for ten minutes, find nothing they can put in a board report, and close the tab. We see this constantly with SMEs and charities. The data is there. The navigation is the problem.

The fix isn’t learning all of GA4. It’s knowing which six reports answer real business questions and ignoring everything else.

The Six Reports Worth Your Time

1. Traffic acquisition: which channels drive results

Found under Reports > Acquisition > Traffic acquisition. This is the single most useful default report in GA4. It shows sessions, key events, and revenue broken down by channel: organic search, paid search, paid social, email, direct, referral.

The question it answers: where is my traffic coming from, and which sources actually convert?

Two things to do every time you open it. First, change the date range to something meaningful: last 30 or 90 days, not the default 28. Second, sort by key events, not sessions. A channel sending 10,000 sessions and 12 conversions is worth less than one sending 800 sessions and 40 conversions. Sessions flatter; key events tell the truth.

Use the dropdown above the table to switch between ‘Session default channel group’ and ‘Session source / medium’ when you need detail. Channel group tells you ‘Paid Search works’. Source / medium tells you ‘google / cpc works but bing / cpc doesn’t’.

2. Key events: what conversions are happening

Found under Reports > Engagement > Key events (Google renamed ‘conversions’ to ‘key events’ in 2024, same thing). This shows each conversion action you’ve marked as a key event: form submissions, purchases, donations, sign-ups.

The question it answers: how many conversions did we get, and is that number going up or down?

One warning: this report is only as good as your setup. If you haven’t defined key events properly, the report is empty or, worse, full of junk like scroll marked as a conversion. If that’s you, fix the foundation first. We’ve written a full guide on conversion tracking setup.

3. Landing pages: where converting visitors arrive

Found under Reports > Engagement > Landing page. If you don’t see it, your property predates Google adding it as a default. Go to Reports > Library, edit the Engagement collection, and add the Landing page report. Takes two minutes.

The question it answers: which pages are pulling their weight?

Add the key events metric to the table and divide by sessions per page. A landing page with 2,000 sessions and 10 key events (0.5%) has a problem. A page with 300 sessions and 24 key events (8%) deserves more traffic. This single comparison drives better budget decisions than any engagement metric in the product.

For charities, run this on your donation and appeal pages specifically. We regularly find appeal pages receiving 60% of paid traffic while converting at half the rate of the main donation page.

4. User acquisition vs traffic acquisition: first touch vs every touch

GA4 has two acquisition reports and they answer different questions. Traffic acquisition is session-scoped: it credits the channel that drove each visit. User acquisition (Reports > Acquisition > User acquisition) is user-scoped: it credits the channel that brought each person to your site for the first time.

Why it matters: someone discovers you through a Meta ad, comes back three days later via Google search, and converts. Traffic acquisition credits organic search with the converting session. User acquisition credits Meta with introducing them. Both are true.

Use user acquisition when you’re asking ‘which channels bring us new people?’ Use traffic acquisition when you’re asking ‘which channels drive converting visits?’ If paid social looks weak in traffic acquisition but strong in user acquisition, it’s doing its job at the top of the funnel. Don’t cut it based on the wrong report.

5. Ecommerce purchases: if you sell online

Found under Reports > Monetisation > Ecommerce purchases. Items viewed, added to basket, and purchased, with revenue per item. The question it answers: what’s actually selling, and where does the funnel leak?

The ratio to watch is add-to-basket rate against purchase rate. If 8% of viewers add an item to basket but only 1% buy, your problem is checkout, not product pages. Skip this report entirely if you’re lead generation or a charity without an online shop. For a non-ecommerce site, monetisation reports are just empty tables.

6. Attribution paths: how channels work together

Found under Advertising > Attribution paths. This shows the sequence of channels people touch before converting: early touchpoints, mid touchpoints, late touchpoints.

The question it answers: is any channel quietly assisting conversions that another channel gets credit for?

Typical finding: paid social appears in the early touchpoints of 40% of converting paths but rarely gets last-click credit. That’s evidence it’s feeding the funnel. Check this quarterly, not weekly. Path data needs volume to mean anything. For the full picture of how GA4 assigns credit, see our guide to GA4 attribution models.

The One Exploration Worth Building

Default reports show one dimension at a time. The question they can’t answer: which landing page converts best for each channel? Your Google Ads traffic might convert brilliantly on one page and die on another. That’s an Explore job, and it takes about five minutes to build:

  1. Go to Explore and choose Blank, then select the Free form technique.
  2. Under Dimensions, click + and import Landing page + query string and Session default channel group.
  3. Under Metrics, import Sessions, Key events, and Session key event rate.
  4. Drag Landing page into Rows, Session default channel group into Columns, and all three metrics into Values.
  5. Add a filter: Sessions greater than 50, so low-volume noise doesn’t clutter the table.

Name it something like ‘Landing page × Channel’ and it saves automatically. You now have a single grid showing every landing page’s conversion rate per channel. This one table has reshaped media plans for us more than any default report. When a page converts paid search at 6% and paid social at 1%, you stop sending social traffic there and build it a dedicated page instead.

Link Google Ads for the Paid View

If you run Google Ads and haven’t linked it to GA4, you’re running blind in both tools. Go to Admin > Product links > Google Ads links, click Link, choose your account, and keep personalised advertising and auto-tagging enabled.

What you get: campaign, ad group, and keyword dimensions inside GA4’s Advertising section, the ability to import GA4 key events into Google Ads as conversion actions, and GA4 audiences available for remarketing. The link takes five minutes and costs nothing. There is no good reason not to do it.

Do the equivalent for Search Console too (Admin > Product links > Search Console). It adds organic query data to the same interface.

What to Ignore

Engagement rate. GA4’s replacement for bounce rate counts a session as ‘engaged’ if it lasts 10 seconds, has two page views, or fires a key event. It’s a soft metric with a generous definition, and it tells you nothing about revenue. A page can have a 75% engagement rate and convert nobody. Glance at it for diagnosis; never report on it as an outcome.

Real-time reports. Useful for exactly one thing: checking that tracking fires when you test it. As a monitoring habit, it’s a slot machine: entertaining, addictive, and decision-free. Nobody has ever made a better budget call because they watched 43 users be on the site right now.

Average engagement time, scrolls, and most of the Engagement overview. Context at best. If a number can go up while your revenue goes down, it doesn’t belong in your monthly report.

Not sure which of your GA4 reports are telling you the truth? We’ll do a no-obligation free GA4 review for charities and growing teams, and tell you plainly what’s worth reporting and what to ignore.

Get a free GA4 review →

The 20-Minute Monthly Routine

You don’t need to live in GA4. Once tracking is solid, this routine covers a typical SME or charity:

Quarterly, add ten minutes on attribution paths and user acquisition to sanity-check the channel mix. That’s the entire analytics workload for most organisations under £20k/month in spend.

When GA4 Won’t Match Your Ad Platforms

GA4 says Google Ads drove 60 conversions. Google Ads says 85. Meta claims 110. None of them are lying. They’re measuring differently:

Expect a 20–40% gap and don’t waste hours reconciling it. Use each tool for what it’s good at: ad platforms for optimising within the platform, GA4 as the neutral referee comparing channels on the same rules, and your CRM or bank account as the final word. If all three trend in the same direction, your marketing is working. If they diverge wildly, that’s when to dig in.

GA4 is a worse interface than its predecessor but a perfectly good measurement tool once you stop trying to use all of it. Six reports, one exploration, twenty minutes a month. Everything else is noise.