If GA4 lists (not set) among your landing pages, part of your website reporting is missing an entry point. That matters when you are deciding which service page deserves more advertising, which product category needs work or which content brings useful enquiries.
Start by checking the measurement behind the row. Renaming it in a dashboard will not tell you where those visits began, and changing campaign links may solve a different problem entirely.
The practical aim is to identify affected journeys, confirm whether their page views reach GA4 and repair the responsible implementation. Keep business decisions proportionate to the evidence you still have.
What does (not set) mean for a GA4 landing page?
Google defines the landing page using the first page view in a session. It can be a product, service or article page; it does not have to be a dedicated advertising page.
For this particular dimension, Google’s explanation is specific: (not set) can appear when a session has no page_view event. A session containing other activity does not necessarily provide the page-view evidence needed to identify its landing page.
This does not establish that your website was unavailable, that the visitor was a bot or that the page performed badly. Those are separate questions. It also differs from an unknown campaign source: knowing where a visit came from and knowing its first recorded page are different reporting tasks.
Establish whether the gap affects a decision
Open the Landing page report and record the date range, affected session count and any enquiries or sales reported against the unknown row. Save an unchanged copy before experimenting with report filters.
Add Session source / medium as a secondary dimension, an option documented in Google’s landing-page report guidance. Then compare periods and investigate whether the problem is concentrated in a particular campaign, device group or recently changed website area.
Use these comparisons to decide where to test first. A rise after a booking-tool change is a useful lead, but timing alone does not prove the cause. Likewise, a mobile concentration tells you which device to reproduce the journey on; it does not prove that mobile browsers are responsible.
Prioritise investigation if the unknown row contains commercially important activity or prevents a decision about substantial spend. A small, stable remainder deserves a different response from a sudden gap across your main enquiry route. Set the priority around the decision at risk rather than an invented acceptable percentage.
Trace a fresh visit from its actual entry page
Choose one journey you can repeat: open a service page directly on a phone, continue to its enquiry form and complete a clearly identified test submission. Start at the URL customers actually use rather than navigating there from your homepage.
Have your developer or analytics specialist use Tag Assistant or Tag Manager preview with GA4 DebugView. Google recommends enabling debugging for your own device so that its activity is easier to identify. DebugView lets you open an event and inspect its parameters.
Ask them to record:
- The starting URL, device, browser, test time and consent choice.
- Whether the initial page sends a
page_viewto the intended GA4 property. - Whether its
page_locationmatches the page being tested. - What happens when the visitor moves to another page or booking screen.
- Whether the intended enquiry event and the actual enquiry both arrive.
Use a test identity and avoid including customer details in screenshots or shared debugging records. Keep the findings specific: “the mobile service page sends a form event but no initial page view” gives a developer something repeatable to investigate.
Google notes that privacy controls or an absence of analytics-cookie consent can prevent events appearing in DebugView. Record the test conditions before treating an empty debugging screen as proof of a broken tag. Use acquisition reports, rather than DebugView, for final attribution analysis.
Check the implementation that owns page views
Automatic page views were disabled without a working replacement
The Google tag sends a page view by default. A custom implementation can suppress that initial event using send_page_view: false. If that setting exists, ask which part of the website is supposed to send the replacement and whether it does so on the affected entry page.
Google’s current page-view guide also distinguishes this setting from enhanced measurement of browser-history changes. History-based page views can still be sent when the initial automatic page view is disabled.
That distinction matters during repair. Do not add another page-view tag everywhere simply because one route is missing. Google warns that combining manual and automatic page views without managing their overlap can produce duplicates. Establish which mechanism should handle the initial load and which should handle subsequent navigation.
Navigation changes the screen without loading a new document
Some websites and booking journeys replace content dynamically. Test a direct visit to a deep URL as well as movement between screens: success on one route does not demonstrate that both work.
Google’s single-page application guidance describes virtual page views triggered through browser-history changes or custom events. It also warns against overlapping automatic history tracking with a Tag Manager implementation.
Ask the developer to follow the method appropriate to the build and verify the page location and referrer as screens change. Missing later views reveal an incomplete journey, but they do not, by themselves, prove why a session’s landing page is unknown. Keep the initial-entry test separate.
The problem appears when consent changes
Repeat the journey using the consent choices your site offers, including a returning visit where a choice has already been saved. Ask the implementer to compare event behaviour before and after the choice.
Google’s guidance on (not set) and consent-mode implementation errors includes incorrectly using a default command to update consent, and sending later events with denied consent after consent was initially granted. These are investigation leads, not evidence that every unknown landing page has a consent fault.
Repair the sequencing while respecting the visitor’s choice. An analytics problem is not a reason to bypass the site’s consent controls.
Give the developer a focused repair brief
Provide the affected URLs, a working comparison journey, the first date you noticed the change and the results of your repeatable test. Include relevant release dates, such as a new page template or booking integration, as context rather than a verdict.
Request a clear explanation of the responsible component: website code, a tag container, an integration or the consent implementation. The handover should say what changed, why it resolves the observed gap and how overlapping page-view collection was checked.
Keep the repair small enough to review. If several unrelated tracking changes are released together, it becomes harder to understand which one helped or introduced a new problem.
Verify both collection and the processed report
Repeat the original tests after the repair, including the affected mobile route and a previously working page. Confirm that the correct entry page is recorded, subsequent navigation behaves as intended and the test enquiry reaches the team.
Check for duplicate page views as well as missing ones. Also verify that existing enquiry or purchase events still represent the intended customer action. A tidier landing-page table is poor compensation for damaged conversion measurement.
Then review a fresh reporting period. Google says data processing can take 24–48 hours, during which report figures may change. Immediate debugging evidence and a later processed report answer different questions; keep both in the handover.
Record the repair date and compare equivalent periods where possible. If the campaign mix or traffic level changed substantially, avoid attributing the entire reporting difference to the technical work.
Use the remaining evidence honestly
Do not redistribute historical unknown sessions across your best-known pages to make a report look complete. A future collection repair does not justify inventing earlier entry points. Mark the affected period and explain which page comparisons remain uncertain.
While the issue is investigated, use your actual enquiry and order records to support commercial decisions. Our guide to connecting enquiries with sales explains how to keep reporting focused on useful business outcomes.
If you need help tracing the gap, BuzzBoost’s analytics and reporting service can help connect website measurement with the decisions you need to make. Tell us which report is unclear and which customer journey matters, and we can identify the next useful check.
Featured image: AI-generated editorial artwork, not a photograph of a real BuzzBoost office, client or result.


