Search Console vs Google Analytics 4: why your data doesn't match

Open Search Console vs Google Analytics for the same month and the scene repeats: you export a number in each tool and they do not match. Nothing is broken, they are two instruments measuring different things. Here are the official reasons and how to compare them without drawing false conclusions.

Search Console vs Google Analytics: two instruments, two questions

They are not two versions of the same number, they are two different meters. Search Console measures what happens inside Google: how often you appeared in the results and how often people clicked you. Google Analytics 4 measures what happens inside your site: sessions, behavior and conversions. The same person can generate a click in one tool and no session at all in the other.

What each tool measures, with the documentation in hand

Clicks. The number of times a user clicked your site from Google Search results. Google only, Search results only: ads do not occupy a search position.

Impressions. How many times your site appeared in Search results. An impression is recorded when the item appears on the current results page even if the user has not scrolled to it, but inside carousels or expandable blocks you do have to scroll or expand for it to count. CTR is clicks divided by impressions.

Sessions. A period of time during which a user interacts with your website or app. It starts with the session_start event and ends after 30 minutes of inactivity by default.

The example that makes it obvious

One click from Google can end in one session, in two (if the person comes back after 30 minutes of inactivity) or in none (blocked tag, denied consent, or the page closed before Analytics loaded). The reverse also happens: Analytics can record an organic session with no click counted in Search Console if the visit came from another search engine. It is the same pattern as field data and lab data: two legitimately different numbers for a single phenomenon.

The official reasons for the gap (one by one)

Google keeps a page dedicated to Performance report discrepancies. These are its causes, translated into practical decisions.

Cause 1: each report measures a different surface

In Analytics, the Organic Search channel groups anyone arriving through non-paid links in organic search results, including AI Overviews and Google's AI Mode. The channel group rule is that the medium matches organic exactly, so every search engine is included, not just Google.

Analytics also has its own default channel called AI Assistant (ChatGPT, Gemini, Deepseek, Copilot or Grok), which does not include AI Overviews or AI Mode, and that traffic does not appear in Search Console. And because the Performance report can be filtered by search type (Web, Image, Video or News), comparing one filtered view against all organic traffic means comparing two different sets, as we cover in multimodal search in Search Console.

Cause 2: clicks vs sessions vs users are not the same thing

Search Console counts clicks with a twist that surprises people: if a user clicks a link inside your property, goes back to the results and clicks another link of the same property, it counts as a single click.

Analytics counts sessions by estimating unique identifiers and relies on HyperLogLog++, a statistical estimate with small differences from an exact count: it may not even match itself. And the Search Console CTR is not the Analytics CTR, they are two divisions over different numerators.

Cause 3: session attribution, the least known and the biggest one

Analytics still attributes sessions with the non-direct last click model, with a 90-day lookback window. The official example is brutal: on day 1 the user arrives from organic Google and is assigned google/organic; on day 68 they come back by typing the address and that session is also attributed to google/organic. That is why Analytics can show more organic traffic than Search Console on a day when nobody clicked from Google.

In the other direction, clicks that never turn into a measured session vanish from Analytics: Search Console tends to mark the ceiling and Analytics the floor.

Cause 4: privacy, processing and estimation

For privacy reasons the Performance report omits queries searched a very limited number of times, and processing of the source data can make statistics differ, for example when duplicates are removed.

Inside Search Console itself there are three internal mismatches: the table shows a maximum of 1,000 rows, anonymized queries are omitted from the table but still counted in the chart totals, and the chart aggregates by property while the table aggregates by page. Hence the classic case of three URLs of your property on the same results page: one impression for the chart, three for the table.

In Analytics, when analytics storage is denied, cookieless pings are sent and Google fills the gaps with modeling: part of the number you see is an estimate, not an observation.

Cause 5: JavaScript, blockers and time

The biggest cause in practice: Analytics only records traffic from users who have JavaScript enabled, while Search Console does not depend on that because it reads what Google already served and already clicked.

Then there is time. Search Console data is usually available within 2 to 3 days, daily views are labeled in Pacific Time and the most recent data is preliminary (dotted line). In Analytics, processing can take 24 to 48 hours. Comparing the last few days of one source with the other is a guaranteed mismatch.

How to compare them properly: a 5-step routine

You do not remove the gap, you control it:

1. Close the window. Use closed ranges of 28 days ending 3 or more days ago, never the current day.
2. Match the time zone. Search Console labels in Pacific Time; if Analytics uses another zone, daily bars never line up.
3. Narrow the surface. Pick the search type in Search Console and filter to Google and organic in Analytics.
4. Reconcile the URLs. Search Console attributes clicks to the canonical URL of the link, so www and non-www variants may be counted under different hosts (see what www is and how it affects SEO).
5. Compare trends, not totals. The useful question is whether both curves move together: if they rise and fall in sync, your measurement is healthy.

No deviation figure is official: Google does not publish an acceptable percentage. And if the drop comes from a single source and is real, it is no longer a data gap but a traffic drop, which you diagnose in how to diagnose a traffic drop in Google.

Which to trust for each decision (and what not to ask either one)

Search Console rules before the click: real queries with their CTR and average position, pages by impressions and clicks, indexing coverage, structured data and inbound links. Its limit is clear: it does not know what happened after the click.

Analytics rules after it: sessions, conversions, revenue and retention, plus all the traffic that does not come from Google (social, email, direct, AI assistants). Its limit: it measures only what the tag managed to see.

The question each one answers

It helps to think in terms of two questions: Search Console answers whether people find you and click you; Analytics answers what they do once they are in. And there are two impossible requests: do not ask Search Console for conversions or revenue, and do not ask Analytics for keywords, because the search term never reaches the tag. If you need both figures in one view, join them without mixing them in an SEO dashboard with Data Studio.

Five mistakes that create the false alarm

• Comparing the current month. Search Console lags 2 or 3 days and its newest data is preliminary, so the total always looks short.
• Comparing a single day. A time zone difference is enough to break everything.
• Comparing apples with oranges. Total organic traffic in Analytics includes other search engines and AI assistants, while the Web view in Search Console leaves out image, video and news.
• Accepting the gap without checking the context. Look at annotations, incidents and whether the tag loaded that day.
• Reconfiguring the tool mid-period. It is the fastest way to lose the comparability of your history.

The honest closing note: the gap is not fixed, it is understood. Do not aim for the numbers to match, aim to always know what you are reading.

Checklist to reconcile your data today

1. Confirm the Analytics tag always loads, opening the page in a browser without extensions.
2. Fix a closed 28-day window ending 3 or more days ago.
3. Align the time zone of both tools.
4. Filter Search Console by search type and Analytics by Google and organic.
5. Check annotations and incidents for the period before interpreting.
6. Compare trends week over week, not raw totals.
7. Document the decision: queries to Search Console, conversions to Analytics, and never mix both figures without saying where each one comes from.