GA4 recorded 81 sessions per 100 Google Ads clicks, 19% fewer sessions than clicks, on the median of 53 GA4 properties connected to ConvRadar that ran Google Ads from 4 October to 2 November 2026. The count covers Search, Performance Max, Shopping and Demand Gen campaigns. The middle half of the properties sat between 55 and 95 sessions per 100 clicks; 12 recorded fewer than 50, and 9 recorded more sessions than clicks.
Analytics Playbook calls a 10% to 20% gap between Google Ads clicks and GA4 sessions normal, and Clixtell says no percentage is normal for every account. On the 53 GA4 properties connected to ConvRadar, 6 sat inside that band of 80 to 90 sessions per 100 clicks, 21 recorded 90 or more, and 26 recorded more than 20% fewer sessions than clicks. The clicks here are GA4's Ads clicks metric, the click count GA4 imports from the linked Google Ads accounts.
| GA4 properties | Count | GA4 sessions per 100 Google Ads clicks, median |
|---|---|---|
| All properties | 53 | 81 |
| Online stores | 29 | 84 |
| Lead generation sites | 16 | 73 |
| European reporting time zone | 13 | 60 |
| Any other time zone | 40 | 86 |
| Search campaigns only, European time zone | 10 | 61 |
| Search campaigns only, other time zones | 33 | 92 |
The Search rows cover the 43 properties with 100 or more Search clicks.
Why GA4 records fewer sessions than Google Ads clicks
A Google Ads click becomes a GA4 session only when a page with the GA4 tag loads and the visitor's consent lets the tag send. Google Ads counts the click either way.
On the median property 1.8% of google / cpc sessions had no Google Ads campaign type, so almost every session GA4 credits to google / cpc keeps its campaign. That count misses clicks whose gclid a redirect drops: those sessions leave google / cpc and arrive under another source, the way a sign-in round trip hands a paid visit to a login page in why GA4 attribution goes wrong. And 8 of the 53 properties had at least one Search, Display or Performance Max campaign with spend and no GA4 session in the 30 days; on 5 of them those campaigns held more than 10% of spend. Their clicks went to pages without this property's tag, or the linked account advertises another site.
The rest of the gap does not show in GA4 data at all. A visitor who declines analytics cookies, or leaves before the tag fires, produces a click and no session. A tap on a call asset is billed as a click and opens no page.
GA4 can also count more sessions than clicks. Google's help page on campaigns and traffic sources says "A direct-traffic visit that follows a referred visit will never override an existing referrer." A visitor who clicks an ad on Monday and types the address on Thursday adds a second Google Ads session for one click, and a shared link that still carries the gclid does the same. That credit sits inside every ratio here, so the share of clicks that never arrive is larger than the gap between clicks and sessions.
Do European sites lose more Google Ads clicks in GA4?
On 13 GA4 properties connected to ConvRadar with a European reporting time zone, GA4 recorded 60 sessions per 100 Google Ads clicks on the median property from 4 October to 2 November 2026, against 86 on the 40 properties in other time zones. On Search campaigns alone the medians were 61 on 10 European properties and 92 on 33 others. 9 of the 13 European properties sat below 70, against 12 of the 40. Shuffling the 53 properties between the two groups at random gives a gap this wide 2 times in 100 across all campaign types and 7 times in 100 for Search alone. The 13 belong to 13 different clients and 2 of them are outside the EU; the time zone is the property's setting, not the visitor's country.
Consent was not measured. A banner that holds the analytics tag until the visitor accepts loses every click from a visitor who declines or never answers, which makes it the first thing to check on a European site, and GA4 adds part of it back as estimated sessions through behavioral modeling on properties that qualify. For a European site, 60 is the median to compare against, not 81.
Do Performance Max clicks turn into fewer GA4 sessions than Search clicks?
Performance Max recorded fewer GA4 sessions per 100 clicks than Search on 16 of 22 GA4 properties connected to ConvRadar that ran both with 100 or more clicks each, from 4 October to 2 November 2026. The median gap was 16 sessions per 100 clicks across all 22, and 20 on the 16 where Performance Max ran lower. On 2 of the 22 the gap was 50 or more: one property recorded 62 sessions per 100 Search clicks and 4 per 100 Performance Max clicks.
Performance Max serves on Search, Shopping, Display, YouTube, Discover, Gmail and Maps, and GA4 reports all of it as one campaign type. Google Ads splits a Performance Max campaign's clicks by channel in its channel performance report, for dates from 6 June 2025, and that is where a Performance Max campaign far below Search in the same property gets read. Outside Performance Max, one property ran a standalone Display campaign at 1.1 sessions per 100 clicks while its Search ran at 38.
Google Ads traffic with no click or cost data in GA4
15 of 72 GA4 properties connected to ConvRadar with 50 or more google / cpc sessions between 4 October and 2 November 2026 showed no Google Ads clicks or cost at all. 10 of the 15 had no Google Ads account linked, and 5 had a link that imported no clicks in the 30 days. Google documents that auto-tagged clicks from an unlinked Google Ads account still arrive as google / cpc, with no campaign data behind them. The two largest recorded 76,624 and 61,706 google / cpc sessions with no cost beside them, so GA4 gives no cost per session or per conversion for any of their campaigns. The fix starts in GA4 under Admin → Product links → Google Ads links: link the account that pays for the clicks, and keep auto-tagging on in it.
Two properties: sessions above clicks, and clicks with no sessions
An online store that spent $25,434 on Google Ads in the 30 days recorded 135 GA4 sessions per 100 Search clicks: 28,189 sessions on 20,810 clicks. Performance Max ran at 85 and Shopping at 84, both above the panel median. On Search, GA4 credited more sessions than Google Ads counted clicks, from return visits that keep the paid source or shared links that carry the gclid. Cost per Search session there reads a quarter lower than cost per click, so judge Search spend by clicks or by conversions, not by sessions. Published paid search conversion rates split on the same base: WordStream's 7.5% divides by clicks and Ruler's 5.4% by sessions, as laid out in conversion rate by traffic source. The 50-point gap between its Search and Performance Max comes from the Search surplus, not from a Performance Max loss.
A lead generation site that spent $19,720 in the same 30 days recorded 165 GA4 sessions on 75,799 Google Ads clicks: 146 on 72,794 Display clicks, 19 on 2,995 Search clicks and none on 10 Video clicks. 54 of its 62 campaigns with spend had no GA4 session, and they held $17,001 of the $19,720. Search clicks went missing as completely as Display clicks, 0.6 and 0.2 sessions per 100, so consent, which loses a share of clicks, does not explain it. Either the ads send visitors to pages without this property's tag, or the linked Google Ads accounts advertise another site. The property recorded 191,592 sessions in the same 30 days and no key event. The check is each campaign's final URL: a page on this site without the tag gets the tag, and a page on another site belongs to that site's GA4 property and its own Google Ads link.
How to check your own
Open Explore → Free form with the last 30 days. Put Session Google Ads campaign type in rows and Sessions, Ads clicks and Ads cost in values, then add Session Google Ads campaign as a second row dimension and set Show rows to 100. Divide sessions by clicks and multiply by 100 on each row. A second tab with Session source / medium in rows and Sessions and Ads clicks in values shows google / cpc traffic with no clicks beside it.
Outside Europe, under 50 sessions per 100 Search clicks puts a property in the bottom quarter of this panel, with 8 of the 33 that had 100 or more Search clicks. Add ?gclid=test to the campaign's final URL, open it, and check that the parameter is still in the address bar of the page that loads: a redirect that drops it moves the session out of google / cpc. A campaign with spend and zero sessions sends its clicks to pages without the tag or to another site. google / cpc sessions with no clicks beside them mean GA4 has no working link to the account that pays for them.
Connect GA4 to ConvRadar and ask Claude or ChatGPT for sessions, Ads clicks and Ads cost by Google Ads campaign for the last 30 days, sorted by cost. That is one question instead of the exploration: the assistant divides sessions by clicks on each row, and a campaign that spent and got no visit comes back with 0 sessions beside its clicks and cost. With no Google Ads link there are no clicks to compare, so link first.
How the 53 properties were measured
GA4 Data API reports were read on 3 November 2026 for every GA4 property connected to ConvRadar by an outside client with an active connection, 179 properties, for 4 October to 2 November 2026. 61 of them imported Google Ads clicks, and the 53 in the panel, belonging to 52 clients, had 200 or more clicks on Search, Performance Max, Shopping and Demand Gen campaigns, 3,105 on the median property. ConvRadar's own and demo properties are excluded. Clicks are GA4's Ads clicks metric; sessions are GA4 sessions whose Session Google Ads campaign type is one of the four. Ratios are computed per property, and figures are medians across properties. GA4 estimates its session count with a 95% interval of ±3.25%, as covered in is GA4 data accurate, small next to the gaps here. Business type is the one the owner stated when connecting, or detected from the property's events where the owner left it blank. Region is the property's reporting time zone. The 2 and 7 in 100 for the European gap come from 20,000 random reassignments of the properties between the two groups. The Search and Performance Max comparison uses properties with 100 or more clicks on each. Campaigns with spend and no session are campaign rows with cost above zero and zero sessions, Video excluded; campaign rows were read up to 100 per property and one property had 109, so 8 is a minimum. The 72 properties in the no-data count are those with 50 or more google / cpc sessions in the window.