Skip to content
Faisal Mahmud TusherDigital marketing, SEO & analytics
with Faisal Tusher

Collection

Analytics Troubleshooting

Each page in this section starts from a symptom seen in Tag Manager, Google Analytics or a dashboard, lists the plausible causes in a diagnosis table, and gives a check, a fix, a verification step and a prevention habit for each. The setting or mechanism behind every cause is explained in an analytics guide elsewhere on the site, and every page links to it rather than repeating it.

Looking for how to use this library? Read the guide below the list.

All 19 pages in this collection

How these pages work

Every page follows the same order. It begins by asking whether the problem is genuine or a reading error, because many "broken" numbers turn out to be a partial day, a filter left on, a preview session mistaken for the live container, or two reports using different definitions. Next comes a diagnosis table: the symptom as it appears in Tag Manager (GTM), GA4 or a dashboard, the check that confirms or rules out a cause, the cause, and the fix. Each cause then gets its own short section with the fix, a way to verify it worked, and the habit that stops it coming back, followed by a routine that catches the whole family of faults early.

Pick the page from the symptom you can see, not the cause you suspect.

What you are seeingStart withThen, if it is not that
A Tag Manager tag never fires, or fires only in previewGTM tag not firingData layer variable undefined
A tag fires but a variable it sends is empty or undefinedData layer variable undefinedGTM tag not firing
A property shows nothing at all, not even in RealtimeGA4 not showing any dataGTM tag not firing
Numbers fell sharply on the day a cookie banner went liveData drop after a cookie bannerGA4 not showing any data
The Unassigned channel holds a large share of sessionsGA4 unassigned trafficDirect traffic too high
Direct traffic is far above what the business expectsDirect traffic too highSudden traffic spike
Sessions jumped overnight with no campaign behind itSudden traffic spikeCommon GA4 mistakes for duplicate tags
A Looker Studio chart and the GA4 interface disagreeLooker Studio not matching GA4BigQuery export not matching GA4
Queries on the raw export disagree with the interfaceBigQuery export not matching GA4Looker Studio not matching GA4
A Looker Studio chart, page or report shows an error or no data at allLooker Studio chart showing an error or no dataLooker Studio not matching GA4 if it shows numbers that disagree
A Looker Studio report shows the same figures day after day, or its latest days never appearLooker Studio report not updatingLooker Studio not matching GA4 if closed days disagree rather than stand still
An audience stays empty, or is too small to use in Google AdsGA4 audience not populatingGA4 not showing any data
A server-side container receives no requests or forwards nothingServer-side GTM not receiving dataGTM tag not firing
The shop takes orders but GA4 shows no purchases or zero revenueGA4 purchase event not trackingData layer variable undefined
GA4 records purchases but its revenue or order count differs from the order systemGA4 revenue not matching ordersGA4 purchase event not tracking if purchases are missing rather than different
Leads arrive by email or in the CRM but GA4 shows fewer form submissions, or noneGA4 not tracking form submissionsGTM tag not firing
Cookies or vendor requests appear before the visitor answers the bannerGTM tags firing before consentData drop after a cookie banner for the opposite problem
Conversions claimed by every channel add up to more than the orders takenChannel conversions exceed salesRetargeting taking all the credit if the issue is where credit lands
Retargeting and brand search claim most sales at returns nothing else matchesRetargeting and brand search take all the creditChannel conversions exceed sales if the total is inflated too

The pages beneath this introduction are listed automatically, so a new diagnosis appears here once it is published.

Broad GA4 data problems live in Common GA4 Mistakes

Some questions are wider than one symptom: "our numbers look wrong" with no single report to point at. Double counting from two installations, "(not set)" values, self-referrals from a payment provider, the "(other)" row and data thresholds are all covered, with their own diagnosis table, in common GA4 mistakes. The pages in this section never repeat that material; where one of those causes is part of a narrower symptom, the page links to the relevant section instead. When the doubt is about the container rather than the property, the preview and debugging tutorial is the tool every page here relies on, and the GA4 hub explains the event model the reports are built from.

Platform-versus-analytics gaps are reconciliation, not troubleshooting

A difference between an ad platform's conversions and the analytics property is usually not a fault. The two systems count different units, credit different touchpoints, apply different windows and see different consent states, so they are expected to disagree. That question belongs to the guide on why platforms and analytics disagree, which gives a reconciliation method rather than a fix. The pages here cover disagreements inside one property: the interface against a dashboard built on it, or against its own raw export. Campaign tagging problems sit in between; when UTM values are the cause, the pages diagnose the symptom and hand the convention itself to the naming guide. For a standing routine that catches all of these before anyone searches for them, the marketing data quality guide sets out the weekly checks.