GA4 E-Commerce Enhanced Tracking in SPAs without DataLayer Race Conditions
A technical tutorial on implementing deterministic, race-condition-free GA4 e-commerce tracking in Single-Page Applications (React, Vue, Next.js).
59 articles in this category
Follow this category by RSSA technical tutorial on implementing deterministic, race-condition-free GA4 e-commerce tracking in Single-Page Applications (React, Vue, Next.js).
A technical step-by-step tutorial on building an automated event reconciliation and validation pipeline between GA4 and Meta Conversions API in BigQuery.
Since 8 June 2026 a Business Profile can be linked in the GA4 admin, and Analytics builds a report collection for it on its own. Six of the seven metrics describe actions that never reach the website, and three limits decide what can be done with them.
The Gemini assistant went to all English accounts in December 2025. Nine months on the help page still says Beta, calls it by a different name than the announcement did, and names one setting that decides whether it appears at all.
Since May 2026 the Data API returns conversions. That release makes visible in code what the interface has done quietly since 2024: two counters run beside each other, and only one of them exists in every property.
A + Create button at the top of the Reports navigation opens a grid canvas with six card types on it. Three things it does not do decide whether it replaces an existing report, and publishing is property-wide with no smaller option.
The export records the consent state that applied when each event was written, in a nested record beside the event. The field is a string rather than a boolean, it can be null, and a consent rate computed without allowing for that will be wrong in a direction that flatters.
Setup Assistant has been replaced. Its successor reads the property, sorts what is missing into six categories, and names three roles as sufficient for access. The two it leaves out are the two that spend their day in the reports.
Total, active, new and returning users look like four slices of one pie. They are one population and three different selections over it, and two of those selections can pick the same person inside the same date range.
Google’s lead-generation tutorial recommends holding form starts against form submits. Enhanced measurement caps form_start at one per session and leaves form_submit uncapped, so that comparison is not a completion rate and can pass 100 percent with nothing wrong.