- Is this guide worth your time, or is it just for tech people?
- What did Google publish, and why does it matter now?
- Why do Search Console and Analytics show different numbers?
- Discover, Search and Images: how do I tell where my traffic comes from?
- What is "dark traffic," and how much organic traffic are you losing without knowing it?
- Redirects and gclid parameters: what's the risk if you don't manage them well?
- Three real-world scenarios: how does the picture change for different Italian companies?
- How do you link Search Console to GA4 for more reliable data?
- Why everyone's talking about it now: the organic traffic landscape in 2026
- How do you get started: the steps to audit your tracking
- What mistakes are most common when analyzing organic traffic?
- How do I know if my tracking is working properly?
- So, is a tracking audit worth it for your company?
- How we see it at SHM
- Frequently Asked Questions
- On November 12, 2026, Google published an official page explaining how to identify search traffic through HTTP Referer headers: you now have an official reference for running audits.
- Search Console counts clicks at the moment of the click; Analytics counts visits only after the page has loaded and the tracking script has run. This is the main cause of discrepancies.
- Discover, Search and Images use the same referrer (google.com): in GA4 you can't tell them apart without extra configuration, and this inflates or skews your organic data.
- 20-40% of the traffic classified as Direct in GA4 is actually organic traffic that lost its referrer because of privacy settings or redirects.
Your Analytics and Search Console data never match? It’s not a bug: it’s a structural difference that Google has just documented systematically for the first time, and it can make your organic traffic look much lower than it really is.
Is this guide worth your time, or is it just for tech people?
If you make marketing decisions based on Analytics data, this guide is for you: understanding why the numbers don’t match keeps you from cutting an SEO budget that’s actually working.
Every week, in dozens of marketing meetings, someone says: “Search Console says X clicks, Analytics says Y visits: which one is right?”. The answer: both, but they measure different things. Google has finally put it all in writing on an official Search Central page, published on November 12, 2026.
You don’t need to be a developer to understand it. But you do need to know what to look for, because the practical consequences — budgets cut by mistake, overestimated campaigns, undervalued SEO — are very real.
| Your situation | Is it worth a closer look? | why |
|---|---|---|
| Analytics shows fewer visits than Search Console | Yes | You're probably losing sessions to ad blockers, redirects or privacy mode |
| You have a lot of unexplained "Direct" traffic | Yes | Part of it is dark traffic: organic traffic that lost its referrer |
| You can't distinguish Discover from Search in GA4 | Yes | Discover can account for more than 60% of Google traffic for some sites |
| You use redirects on campaign landing pages | Yes | You risk losing the gclid parameter and not tracking paid conversions |
| You've already linked GA4 with Search Console | Check anyway | The standard integration doesn't separate AI Overviews, Discover and classic Search |
What did Google publish, and why does it matter now?
Google has created an official page listing all the HTTP Referer values passed when a user clicks from Google Search, Discover or Images: for the first time, marketing teams have an official reference for audits.
The HTTP Referer is a piece of technical information that the browser sends to the destination site when a user arrives by clicking from another page. In practice, it says: “this visitor comes from here.” Google Search uses a policy called origin referrer policy: it sends only the originating domain (e.g. https://www.google.com/), without the page path, the search query or other details.
This means Analytics sees traffic coming from google.com, but it doesn’t know whether that person searched for something on Google Search, saw an article on Google Discover or found an image on Google Images. Three different channels, same referrer.
The official Google documentation also clarifies the cases where the referrer disappears entirely: private browsing mode, opening in a new tab (depends on the browser), switching from a mobile search widget to an external browser. In all these cases, Analytics records the visit as Direct — even though it actually came from Google.
Why do Search Console and Analytics show different numbers?
The two tools measure different moments of the same journey: Search Console counts the click the moment it happens, while Analytics counts the visit only after the page has loaded and the tracking script has run.
Imagine a user clicking your link in Google results. Search Console immediately logs +1 click. But if the user has an ad blocker (a program that blocks tracking scripts), has rejected cookies, or the page doesn’t fully load, Analytics records nothing. As far as Analytics is concerned, that visit never happened.
On top of that, Search Console counts only organic clicks, while Analytics may also include traffic from Google Ads — depending on how the site is configured. The result is that the two numbers almost never match, and that’s not an error: it’s a difference in design.
- 1 · UserClicks the link on GoogleSearch Console records +1 click at the exact moment of the click.
- 2 · BrowserLoads the destination pageIf there are redirects, the referrer can change or disappear.
- 3 · Analytics scriptRuns on the pageIf the user has an ad blocker, declined cookies or the page doesn't load, the script doesn't run.
- 4 · GA4Records (or doesn't record) the sessionOnly if the script has run does Analytics count the visit as organic from Google.
- 5 · Marketing teamCompares the two toolsSees a discrepancy and often can't explain it, but now has an official reference.
Discover, Search and Images: how do I tell where my traffic comes from?
In GA4 you can’t tell them apart by referrer alone: all three use google.com as the source. The only way to separate Discover is to use the Discover report in Search Console.
This is the point that surprises marketing managers the most. Google Discover — the feed of news and content that appears in the Google app and on Chrome mobile — uses exactly the same HTTP referrer as Google Search. So in GA4 everything ends up in the Organic Search channel, all mixed together.
For publishing and content sites, the problem is huge. According to Keywords Everywhere data based on Chartbeat (August 2025), 68% of Google traffic to almost 2,000 news and media sites comes from Discover, not traditional Search. If you don’t know this, you’re optimizing for the wrong channel.
For an e-commerce or B2B site, the Discover share is generally lower, but it can still distort your keyword and landing page analysis. The right way to separate the data is to open the Discover report in Search Console — not to rely on the referrer in Analytics.
| Google channel | Referrer in Analytics | Where to find separate data |
|---|---|---|
| Google Search (web) | https://www.google.com/ or local domain | Search Console → Performance → Search |
| Google Discover | Same as Search (google.com) | Search Console → Performance → Discover |
| Google Images | Same as Search (google.com) | Search Console → Performance → search type filter |
| Google App (iOS) | https://www.google.com/ | Not distinguishable from web Search in Analytics |
| Google App (Android) | google.com or android-app://… | The android-app referrer can be identified in server logs |
| AI Overviews / AI Mode | Included in GA4 Organic Search | Not yet separable in Search Console (Nov. 2026) |
What is "dark traffic" and how much organic traffic am I losing without knowing it?
Dark traffic is real traffic that Analytics classifies as Direct because it lost its referrer: according to Measureboard, it can account for 20 to 40% of all direct sessions.
The term dark traffic refers to visits that reach your site from an identifiable source — like Google Search in private mode, a link shared on WhatsApp, or a privacy-focused browser like Brave — but that Analytics records as Direct because the referrer was removed.
According to Measureboard (April 2025), for most websites, Direct traffic in GA4 accounts for between 20% and 40% of all sessions. A significant part of this is dark traffic: real organic traffic that looks direct.
This has immediate practical consequences. If your monthly report shows a drop in organic traffic, part of it may have moved to private mode or to privacy-first browsers — rather than your SEO getting worse. Before cutting budget, an audit is worth doing.

Redirects and gclid parameters: what's the risk if you don't manage them well?
A poorly configured redirect can wipe out the Google referrer and the gclid parameter — the code that links a Google Ads click to a conversion — making it impossible to measure the return on paid campaigns.
The gclid (Google Click ID) is a parameter that Google Ads automatically adds to the URL when someone clicks an ad. It links that click to a later conversion — a purchase, a form submission, a phone call. If your landing page redirect strips the query string (the part of the URL after the question mark), the gclid disappears and that conversion becomes invisible to Google Ads.
The same goes for the referrer: a redirect with an overly restrictive referrer policy can make a visit look like it came from nowhere. The result is paid traffic showing up as Direct, and campaigns that seem not to convert when they actually do.
This problem is especially common in e-commerce stores that use redirects to manage product variants, or in companies that migrated their site without checking that the redirects pass parameters. If you have doubts about how your SEO is doing after a migration, you can dig deeper in our article on declining organic traffic: real problem or illusion?
| Redirect type | What happens to the referrer | What happens to the gclid | Risk |
|---|---|---|---|
| Standard 301/302 | Passes normally | Passes if the URL is complete | Low |
| Redirect that strips the query string | Passes | Disappears | High for paid campaigns |
| Redirect with strict Referrer-Policy | Disappears | Disappears | Very high |
| Client-side redirect (JavaScript) | Replaced with the site's URL | May disappear | Medium-high |
| HTTP to HTTPS redirect | Passes normally | Passes | Low if configured correctly |
Three real-world scenarios: how does the picture change for different Italian companies?
An e-commerce store, a professional firm and a manufacturing company have different tracking problems, and different solutions.
Illustrative examples: numbers are meant to show the reasoning, they are not real customer data.
| Company type | Typical problem | Likely cause | Priority action |
|---|---|---|---|
| Fashion e-commerce (50k visits/month) | Analytics shows -15% organic, Search Console stable | Growing ad blocker use and cookie rejection | Cookie consent audit + comparison with Search Console data |
| Accounting firm (5k visits/month) | A lot of unexplained Direct traffic | Dark traffic from links shared on WhatsApp and email | Segment Direct by landing page and time of day |
| B2B manufacturing (20k visits/month) | Google Ads campaigns with no tracked conversions | Landing page redirect that strips the gclid | Make sure all redirects pass the full query string |
How do you link Search Console to GA4 for more reliable data?
Google has released a preconfigured Looker Studio template that puts GA4 and Search Console data side by side. It's the quickest starting point for reconciling the two tools.
Looker Studio (formerly Google Data Studio) is Google's free tool for building data dashboards. In February 2025, Google published an updated guide on integrating GA4 and Search Console, with a preconfigured template that shows data from both tools side by side. The GA4 data is already filtered by Session source = google and Session medium = organic.
There are two important limitations to know about, though. First: Search Console data appears in Analytics with a 48-hour delay. Second: available history is limited to 16 months, and the reports don't support time-series charts, as confirmed by SEOCrawl's guide to the integration.
There's one limitation this integration doesn't solve: GA4 includes clicks from AI Overviews and AI Mode in its definition of the Organic Search channel. You can't tell traditional SEO traffic apart from traffic generated by AI answers without additional configuration, as Lovarank (October 2026) explains. To learn more about how to read Google's AI reports, check out our article on how to turn Google's AI report into useful data.

Why everyone's talking about it now: the organic traffic landscape in 2026
Google's guide arrives at a time when organic traffic is under pressure from several directions: AI Overviews, a growing Discover and browser privacy features that hide more and more data.
Data reconciliation isn't a new topic, but it has become urgent for three converging reasons. First: Google's AI Overviews and AI Mode are changing how users interact with search results, and their clicks land in GA4 mixed in with traditional organic traffic, with no distinction. We covered this in detail in our article on Google Zero and the disappearance of organic traffic.
Second: Discover is growing as a channel, but it's invisible in Analytics without dedicated configuration. Third: privacy-first browsers and consent regulations are increasing the share of dark traffic, making organic data less and less reliable without a periodic audit.
If you don't reconcile your data, you risk making the wrong decisions: cutting an SEO budget that works, or failing to notice that a paid campaign isn't tracking conversions. To understand how to tell a real traffic drop from a statistical illusion, also read our deep dive on lost organic traffic and Google's human review.
How do you get started? The steps to audit your tracking
Five steps are enough to see where you're losing data and how to fix the most critical setups.
- 1 · Marketing managerCompare Search Console and GA4Open both for the same period and compare Search Console organic clicks with organic sessions from Google in GA4. A gap of more than 20% deserves a closer look.
- 2 · Tech teamCheck your landing page redirectsMake sure all redirects pass the full query string, including the gclid parameter. Use a tool like Screaming Frog or your server logs.
- 3 · Marketing managerAnalyze Direct traffic by pageIn GA4, segment Direct traffic by landing page. If the pages with the most Direct traffic are the ones you optimize for SEO, you have dark traffic to investigate.
- 4 · Tech teamSet up the GA4 + Search Console integrationUse Google's Looker Studio template to view the data side by side. Remember the 48-hour delay and the 16-month history limit.
- 5 · Marketing managerUse the dedicated reports in Search ConsoleTo separate Discover from Search, use the Discover report in Search Console. For images, use the search type filter in the Performance report.
What mistakes are most common when analyzing organic traffic?
The three most common mistakes are comparing the wrong numbers, ignoring dark traffic and not using Search Console to separate Google channels.
| Common mistake | Consequence | How to avoid it |
|---|---|---|
| Comparing Search Console clicks with GA4 sessions 1:1 | You conclude that Analytics "loses" data, without understanding why | Understand that they measure different moments and use both to triangulate |
| Ignoring dark traffic in the Direct channel | You underestimate your real organic traffic | Segment Direct by page and time of day, and compare with Search Console |
| Not checking redirects before a paid campaign | gclids get lost and conversions aren't tracked | Test redirects with URL analysis tools before launch |
| Using only GA4 to evaluate Discover | Discover traffic gets attributed to Search | Always open the Discover report in Search Console |
| Not updating referral exclusions in GA4 | Google traffic gets reclassified as Referral or Direct | Check your custom channel settings in GA4 |
How do I know if my tracking is working properly?
Five indicators signal reliable tracking: monitoring them every month avoids surprises in quarterly analyses.
| Indicator | Benchmark | Warning sign |
|---|---|---|
| Gap between Search Console clicks and GA4 sessions | Below 20% | Above 30%: look into ad blockers, redirects, consent |
| Direct traffic share of total | Below 25% | Above 40%: likely high dark traffic |
| Paid conversions with tracked gclid | Above 90% | Below 80%: check landing page redirects |
| Discover sessions in Search Console vs GA4 Organic | Monthly comparison | Growing gap: Discover is increasing without being tracked |
| Search Console data available in GA4 | Within 48-72 hours | No data after 72 hours: integration issue |
So, is a tracking audit worth it for your company?
It's worth it if you have active Google Ads campaigns with redirects on landing pages, you notice discrepancies above 20% between Search Console and GA4, or you have a lot of unexplained Direct traffic: every week without an audit is a week of decisions based on bad data.
Somewhere in between: your site has stable organic traffic and you don't run paid campaigns. An audit is still useful, but you can plan it calmly over the next 30 days, with no urgency.
Not yet if you have a brochure site with very little traffic and no active campaigns: the issue is real but not a priority compared to other foundational work.
How we see it at SHM
To us, the guide published by Google isn't technical news for developers: it's a reminder that the data behind marketing decisions needs maintenance. We believe simplifying digital is the highest form of expertise, and that applies to tracking too: you don't need a complicated dashboard, you need to understand what you're really measuring and what you're losing.
In our experience with sites that handle large volumes of traffic across multiple markets, a discrepancy between Search Console and Analytics is almost always there. The problem isn't the tool: it's that nobody has ever run a structured audit. Often two hours of analysis are enough to find the redirect that wipes out the gclid or the custom channel that reclassifies organic traffic as Referral.
If you want to understand how your SEO is really doing, separating noise from real signals, we can help with a tracking audit, a custom GA4 + Search Console integration, or an SEO for AI strategy that also accounts for new channels like AI Overviews and Discover. You can also explore how we use data processing and business intelligence to turn scattered data into clear decisions, or how we build the strategic integration of AI into marketing processes. To dig deeper into organic visibility today, the starting point is our SEO and AI visibility hub.
Frequently Asked Questions
Search Console counts every click the moment it happens, even if the page doesn’t fully load. GA4 records a visit only after the page has loaded and the tracking script has run. Ad blockers, cookie rejection and loading issues reduce the number of sessions recorded in GA4, creating a natural discrepancy that can reach 20-30% or more.
You can't do it directly in GA4: both use the same HTTP referrer (google.com). The only correct way is to use the Discover report in Google Search Console, which shows Discover clicks and impressions separately. In GA4, everything ends up in the Organic Search channel with no distinction.
Dark traffic is real traffic that Analytics classifies as Direct because it lost its referrer. It happens with private browsing mode, with privacy-first browsers like Brave, or with links shared in messaging apps like WhatsApp. To spot it, check the landing pages of your Direct traffic in GA4: if they’re the same pages you optimize for SEO, you’re probably looking at organic dark traffic.
Yes. If a redirect strips the URL's query string, the gclid parameter, which links a Google Ads click to the conversion, disappears. Google Ads can no longer attribute the conversion to that campaign. Before launching paid campaigns, always check that all landing page redirects pass the full URL, including parameters.
No. According to Lovarank (October 2026), GA4 includes clicks from AI Overviews and AI Mode in the Organic Search channel definition, together with traditional SEO traffic. You can’t separate them without additional configuration. Search Console’s standard Performance report also counts clicks from both categories together.
The native GA4 and Search Console integration shows a maximum of 16 months of history, and Search Console data appears in GA4 with a delay of about 48 hours. The integrated reports don't support time-series charts. For analyses over longer periods, you need to export the data from Search Console separately.
Your organic data isn’t wrong: it’s incomplete — and there’s a huge difference between the two.
Related articles
Discover more articles exploring similar topics, selected to offer you a more complete and stimulating perspective. Each piece of content is carefully chosen to enrich your experience.