GA4 is only powerful if the schema is right.
Most brands ported UA to GA4 in a hurry and never rebuilt the event schema. That is why nothing matches Google Ads, cohort reports look wrong, and BigQuery data is unusable. I fix all three, in the same engagement.
If your GA4 does not match Google Ads, everything downstream is guesswork.
Universal Analytics forgave sloppy implementation. GA4 does not. Event-based schema, consent-mode signals, enhanced conversions, and BigQuery export are all interdependent. Get one wrong and every dashboard, every attribution report, every retargeting audience gets slightly off — and slightly off compounds.
As a GA4 / GTM / attribution consultant, I rebuild the entire measurement stack: event taxonomy, custom dimensions, user-ID unification, consent mode v2, server-side GTM, Enhanced Conversions, BigQuery streaming export, and a Looker Studio (or Sigma / Mode) layer that finance can trust.
The goal is not "more data". The goal is reliable, actionable, executive-grade reporting where paid-media claims, product usage, and finance revenue reconcile within 2% every week.
Six layers, one integrated stack
Event Taxonomy
A documented event schema (naming, parameters, business definitions) versioned in Notion — the single source of truth for engineering + marketing.
GTM Web + Server
Web GTM restructured. Server-side GTM (GCP or Stape) deployed. All ad platform tags migrated for signal recovery.
Enhanced Conversions
User-provided data hashed (SHA-256) and forwarded to Google Ads and Meta CAPI for match-rate restoration.
GA4 Configuration
Custom dimensions, custom metrics, conversion events, audiences, and exploration templates — all mapped to your growth model.
BigQuery Export
Streaming export activated. Warehouse tables partitioned. Base queries and views built for cohort / LTV / attribution analysis.
Attribution & Reporting
Data-driven attribution enabled, model comparison built, custom Looker Studio suite handed over.
What is in the engagement
Six-week rebuild, 100% reconciled by week 8
Audit + schema
Current-state audit. Draft the event schema together with your product and marketing teams.
Build
GTM web + server rebuild. Enhanced Conversions and CAPI wired. Consent mode deployed.
BigQuery + reporting
Warehouse export live. Base queries + Looker Studio dashboards.
Reconcile + train
Numbers reconciled across GA4, Google Ads, Meta, revenue. Handoff.
GA4 & attribution questions I get every week
Do I need server-side GTM?
If you spend >$20k/month on paid and depend on Meta or Google conversion signals — yes. iOS ITP and browser extensions are eating 15–35% of your signal. sGTM is the only structural fix.
Can you migrate from Universal Analytics if I never did?
Yes. Even in 2026 I see brands still on shadow UA workflows. Full migration is part of the engagement.
Is BigQuery worth it for a small brand?
If you have >100 transactions a day or a multi-touch B2B funnel, absolutely. BigQuery costs a few dollars a month and unlocks LTV, cohort, and attribution analysis GA4 UI cannot do.
What is data-driven attribution?
A machine-learning model that credits conversions across touchpoints based on their actual incremental contribution — not last-click. It requires ~600 conversions and 30 days of data in GA4.
Do you support Piwik Pro, Matomo, Segment, or Rudderstack?
Yes — I have implemented all four. Analytics platform choice is negotiable; the schema discipline is not.
Continue reading
Ready to see the leaks in your growth stack?
Book a free 30-minute diagnostic. If there is no leverage to unlock, I will tell you on the call.
Book a Diagnostic