Google Analytics delivers solid trend insights, yet it can miss 10‑30% of visits and conversions because it relies on client‑side code, consent models, and sampling.
Why GA4 numbers differ from backend logs
GA4 collects data through a JavaScript snippet that runs in the visitor’s browser. If the script fails to fire, if an ad blocker blocks the request, or if a user declines consent, the event never reaches Google. Back‑end systems like payment gateways record every transaction, so GA4 often undercounts traffic and conversions.
Impact of privacy regulations and consent banners
GDPR, CCPA and similar laws require cookie consent. When users opt out, GA4’s Consent Mode stops the tag from firing and instead uses machine‑learning to estimate the missing data. These estimates have no published confidence interval, so the figures are not raw observations.
Sampling and data thresholding in GA4
- Sampling occurs when a report exceeds 10 million events; GA4 analyses a subset and extrapolates, losing granularity.
- Thresholding hides rows that fall below a minimum user count to protect privacy, which can make traffic appear to vanish.
- Other row groups low‑frequency values into a catch‑all bucket, distorting detailed breakdowns.
Technical blockers that hide traffic
Ad blockers, privacy browsers, corporate firewalls and VPNs can prevent the GA4 script from loading or sending data. Even server‑side tagging cannot bypass user opt‑outs, so some visitors remain invisible to GA4.
Reconciling GA4 with other sources
Because GA4 is a statistical tool, compare its metrics with server logs, CRM data or e‑commerce back‑ends for absolute counts. Use GA4 to spot trends, channel performance and anomalies, then verify critical figures with a primary data source.
When GA4 is still useful
GA4 excels at ranking marketing channels, tracking user engagement over time, and detecting sudden spikes or drops. For budgeting or financial reconciliation, pair GA4 with a reliable backend source.