Your Search Console shows organic clicks climbing for three straight weeks. Sales are flat. You open Google Analytics 4 to see which searches are converting, type “organic keywords” into your mental checklist, and find nothing that answers the question. The queries you saw in Search Console are not here. The landing pages are, but the numbers do not line up with the clicks you were just looking at. So which tool is lying?
Neither is. They measure two different halves of the same visit, and most confusion about “SEO analytics” comes from expecting one tool to answer a question the other one owns. Search Console describes how a person reached your site from Google: the query, the impression, the click, the position. Google Analytics 4 describes what happened after they arrived: which page they landed on, how long they engaged, whether they did anything you care about. One is pre-click. The other is post-click. Neither reports your rankings the way a rank tracker does, and neither is a scoreboard Google uses to rank you.
What each tool actually measures #
The cleanest way to keep the two straight is to split the visit at the moment of the click. Everything before it lives in Search Console. Everything after it lives in GA4. That single line resolves most of the “why don’t these match” questions before you open either report.
| Question | Search Console | Google Analytics 4 |
|---|---|---|
| What query did they search? | Yes (the Performance report) | No |
| Impressions, average position, CTR | Yes | No |
| Which landing page got the visit? | Yes (Pages tab) | Yes, in more detail |
| What did the visitor do after landing? | No | Yes |
| Did organic traffic convert or generate revenue? | No | Yes |
| How organic compares to other channels | No (Google organic only) | Yes (all channels side by side) |
Notice the two tools only truly overlap on one row: the landing page. That overlap is why linking them is useful, and also why their click and session counts will never be identical, which is covered further down.
What GA4 shows that Search Console cannot #
Search Console stops at the click. GA4 is where you learn whether that click was worth anything, and it answers four questions Search Console structurally cannot.
How organic behaves as a channel. GA4 sorts every visit into a default channel grouping, and “Organic Search” is one of those channels. You can look at organic on its own, then place it next to Direct, Referral, Paid Search, and the rest in the same report. Search Console only ever sees Google organic, so it can tell you organic is growing but never whether it is growing faster or slower than your other channels, or quietly stealing from them.
What visitors do after they land. This is GA4’s core job. For organic sessions you can read engaged sessions and engagement rate, pages per session, and the next pages people move to. GA4 defines an engaged session as one that lasts longer than 10 seconds, includes a key event, or has at least two page or screen views. Bounce rate exists too, but in GA4 it is simply the inverse of engagement rate, not the single-page-session metric it was in the old Universal Analytics.
Whether organic drives outcomes. You mark the events that matter to your business as key events (a form submit, a purchase, a signup), and GA4 attributes them back to the channel that earned the visit. That is how you learn a page pulling heavy organic clicks converts almost no one, while a quieter page carries the revenue. Search Console cannot see a conversion at all, because the conversion happens on a page Google never watches.
Paths and assisted value. Because GA4 is event-based, it can follow a visitor across a session and, through its attribution reports, show when organic started a journey that a different channel finished. A visit that looks like a dead end on a last-click view often turns out to be the introduction.
Google renamed “conversions” to “key events” across most GA4 reports on March 21, 2024, reserving the word “conversion” mainly for the Google Ads context. Only the labels and report names changed. The underlying measurement and your historical data stayed intact, so older tutorials that still say “conversions” are describing the same mechanism.
What GA4 cannot tell you #
Reaching for GA4 to answer a Search Console question is the most common mistake, and it goes both ways. Five limits are worth naming before you trust a number.
It has no organic keyword. GA4 does not record the search someone typed to find you. There is no keyword dimension for organic traffic. That data lives only in Search Console, and even there it is the query, sampled and filtered for privacy, not a perfect log. If a report claims to show you “organic keywords” inside GA4, it is pulling them from a linked Search Console, not from GA4 itself.
It is not ranking data. GA4 measures sessions and behavior, never position. A drop in organic sessions can mean a ranking slide, but it can equally mean seasonality, a tracking break, or a SERP feature eating your clicks while your rank holds. Treat a GA4 dip as a symptom to diagnose against Search Console and a rank tracker, not a verdict on its own.
The “Organic Search” channel is only as accurate as your tagging. GA4 assigns channels using rules, and a misconfigured setup routes real organic visits into Direct or Unassigned. Redirects that strip referrer data, missing campaign parameters, and consent gaps all leak organic sessions elsewhere. The number is a classification, not ground truth, and it is worth auditing before you build a report on it.
Thresholding and sampling round the edges off. When Google signals are on, GA4 can withhold rows for small segments to protect user privacy, so low-volume organic pages may show blanks rather than tiny counts. In explorations, very large date ranges can trigger data sampling, where GA4 estimates from a subset. Both mean the figure on screen is often close rather than exact, which matters most on smaller sites where a handful of sessions moves a percentage.
The attribution model changes the credit. GA4 defaults to data-driven attribution, which spreads conversion credit across touchpoints. Switch the report to last-click and organic’s contribution can rise or fall without a single real visit changing. When you quote a conversion number, know which model produced it.
Setting it up to answer SEO questions #
Out of the box GA4 reports on all traffic mixed together, which buries the organic story. A little setup makes it a genuine SEO instrument.
Start by isolating organic. In any report or exploration, apply a filter or comparison on the default channel group set to “Organic Search,” so every metric you read is organic only. Then link Search Console to the GA4 property and publish the two Search Console reports it unlocks, which puts queries and Google organic landing pages inside GA4 next to your behavior and key-event data. The link is the one bridge between the pre-click and post-click halves, and without it you keep flipping between two tabs.
Next, define your key events honestly. Mark the two or three actions that actually represent business value and leave the vanity events unmarked, so organic’s contribution reflects outcomes rather than clicks. For anything beyond the standard reports, build an exploration: landing page as the row, organic as the filter, and engaged sessions plus key events as the columns will surface which entry pages earn attention and which waste the click.
Scale changes the workflow. On a small brochure site with a few dozen pages you can read the organic landing-page report row by row each week. On a large publisher or store with tens of thousands of URLs, that is impossible; there you segment by page template or directory, watch cohorts and trends instead of single pages, and set the exploration to compare periods so a shift stands out against the noise.
Verifying the numbers and watching for breaks #
Before any GA4 figure drives a decision, confirm the tracking is sound. Use DebugView or the Realtime report to check that your key events actually fire when you complete the action yourself; a conversion count of zero is far more often a broken tag than a dead page. Then cross-check magnitudes: GA4 organic sessions and Search Console clicks should move in the same direction over time, though never match exactly, because a click is a tap in Google while a session is a tracked visit that can be blocked, consented away, or split across time.
When something breaks, it usually shows first in one specific place. A sudden collapse in organic sessions with Search Console clicks holding steady points at a broken or removed GA4 tag, visible in Realtime within minutes. Organic visits draining into Direct or Unassigned points at a channel or referrer problem in your tagging. Key events dropping to zero while sessions continue points at the conversion setup, not the traffic. Knowing which report lights up first turns a vague “analytics looks wrong” into a specific fix.
Set it up in this order this week: apply the Organic Search filter to your reports, link Search Console and publish its two reports, mark your real key events, then confirm each one fires in DebugView. Watch the organic landing-page and key-event trend for a few weeks against Search Console before you draw conclusions. Do that, and GA4 stops being the tool that hides your keywords and becomes the one that tells you whether your SEO is actually paying for itself.