The short answer
Tools disagree because they define terms differently, measure at different points, and see different visitors. Cookie consent, bots, internal traffic, time zones, attribution windows and estimation all move the totals. A gap is normal if it stays roughly steady. Worry about sudden changes, a tool showing zero, or gaps larger than the traffic could explain.
Different tools count different things
The word "visit" is not a standard. Each tool defines its own unit and measures it at its own point.
A click in Search Console is not a session in GA4. Someone can click and leave before the page loads, so there is a click and no session. In Search Console a click is also not counted when it stays within Google, such as a query refinement. A user in GA4 is an estimate based on devices, and the same person on two devices can be two users.
| Tool | What it counts | Where it measures |
|---|---|---|
| Google Search Console | Impressions (a link was seen or could be seen) and clicks (a click from Google to a page outside Google) | On the Google results page |
| Google Analytics 4 | Sessions, users and events | In the visitor browser, after the page loads and the tag runs |
| Ad platform (Google Ads, Meta) | Clicks on ads and conversions it attributes to ads | On the ad, and on its own conversion tag |
| CRM or inbox | Leads, deals, calls actually received | After a person is recorded by you or your team |
Consent mode and cookie banners
If your site shows a cookie banner and a visitor declines analytics cookies, GA4 may not record them normally. With Google consent mode, when analytics storage is denied, no analytics cookies are set or read. In an advanced setup the tag still sends cookieless pings, which Google uses for modeling. In a basic setup, tags are blocked until consent, so there are no pings and no modeling for those visitors.
This means GA4 may undercount visitors who decline, or fill the gap with modeled estimates. Modeled data supplements observed data, so figures can differ from what the raw counts would show. Search Console is not affected by your cookie banner, because it measures on Google's side. A steady gap between Search Console clicks and GA4 sessions is often partly a consent effect.
Bots, spam and internal traffic
- Bots: some crawlers do not run the analytics tag and so never appear in GA4, while some do appear. Tools that count server requests will show much higher numbers than GA4.
- Internal traffic: your own team browsing the site inflates numbers. GA4 can exclude it by IP address using a data filter, which can take 24 to 36 hours to take effect and can be tested before it is switched on. Be aware that once applied, the effect on the data is permanent.
- Spam form submissions can inflate conversions in GA4 and in the ad platform while the CRM, which a person curates, shows fewer.
Time zones, windows and attribution
Each tool has a reporting time zone. If Search Console, GA4 and your ad account are set to different zones, a late-evening visit can land on different days in each. Over a month the effect is small, but for a single day it can be visible.
Attribution differs even more. An ad platform generally credits itself for a conversion if the person clicked or viewed an ad within its attribution window. GA4 uses its own models, such as data-driven attribution or last-click-style models, and its reports exclude direct visits from credit unless the whole path was direct. The same lead can therefore be credited to Google Ads in one tool, to Organic Search in GA4 and to "website" in your CRM. Add them together and the leads total far more than you really received.
Never add conversions from different platforms together. Each tool is claiming credit for the same people.
Estimation, thresholds and sampling
Google Analytics says session and user metrics are calculated through estimation, and some metrics use approximate distinct counts, so small differences between GA4 reports are not unusual. GA4 can also withhold rows through data thresholds when counts are low or demographic data is involved. Widening the date range reduces many of these effects.
Diagnosis table
| What you see | Likely cause | First check |
|---|---|---|
| Search Console clicks higher than GA4 organic sessions, steadily | Consent declines, quick exits, bots, different definitions | Is the gap about the same each month? |
| The gap suddenly widened | Tag removed or broken, consent banner change, redesign | Ask when the site or banner last changed |
| GA4 shows zero for a day or a page | Tag not firing on that template, or an outage | Open the page and check Realtime in GA4 |
| Ad platform leads higher than CRM | Spam, duplicates, attribution windows | Compare against your inbox for the same dates |
| GA4 leads higher than ad platform | GA4 counts all sources, ad platform only its own | Compare by channel, not in total |
| Direct traffic spiked | Lost source data, untagged links, app traffic | Check campaign links for tracking parameters |
| Traffic spike on one odd day | Bot, newsletter, media mention | Look at the source and country for that day |
When a gap is a problem
A gap is probably fine if it is stable, if direction and trend agree across tools, and if the size is explainable by consent and bots. It is likely a problem if one tool shows a sudden step change on a known date, if a number is zero where it should not be, if conversions in one tool are more than double those in your own records, or if the gap changes after a site update.
Example: Illustrative example only: the numbers are made up
A tradesperson sees 1,000 Search Console clicks and 750 GA4 organic sessions in April, and 1,100 clicks and 830 sessions in May. The ratio is about 75 percent both months. That is a stable, explainable gap, consistent with consent declines and quick exits. In June, clicks are 1,150 and sessions 400. The ratio has fallen to 35 percent, right after the site was redesigned. That is a real problem: the analytics tag was probably left off the new templates. It was only found because someone compared the tools.
When to ask us
Ask us when a gap changes suddenly, when a number is zero that should not be, or when the totals contradict your own records. The fastest way to get to the answer is to give us the two numbers and the exact date range for each. Use the support form and choose Report or analytics question. For the basic terms, see GA4 basics and Search Console basics.
- The two numbers that disagree, which tool each came from, and the exact date range and time zone if you know it.
- Screenshots of both reports.
- The date the site, cookie banner or phone number last changed.
- Files as Google Drive, Dropbox or Loom links. Never send passwords.
