Adobe Analytics to GA4 Migration: A Different Beast (2026)

Why is Adobe Analytics → GA4 harder than UA → GA4?

Adobe Analytics and GA4 have fundamentally different data models. Adobe's props (page-level traffic variables) and eVars (persistent conversion variables) map awkwardly to GA4's flat event-parameter structure. Adobe's calculated metrics, segments, virtual report suites, and Workspace projects have no direct GA4 equivalents — they need re-implementation in BigQuery + Looker Studio.

Adobe's processing rules, marketing channels, and classification rules require manual replication. Typical migration scope: 80–200 engineering hours for an enterprise property, with stakeholder report rebuilds taking another 40–80 hours.

The four-phase pattern that works: (1) parallel deployment with both systems running, (2) historical data bridge via shared BigQuery warehouse, (3) report-by-report rebuild with stakeholder validation, (4) Adobe decommissioning only after 90 days of GA4 acceptance.

The conceptual model differences

Adobe Analytics has analytical primitives that GA4 doesn't:

Props vs eVars vs Custom Dimensions

Adobe Props: page-level traffic variables. Set per-page-view, expire at end of page. Maximum 75.

Adobe eVars: conversion variables that persist beyond a page. Configurable expiration (visit, 30 days, 90 days, until next reset). Maximum 250+. The persistence is what makes them powerful — set "campaign" as eVar, conversions on next page get attributed.

GA4 Custom Dimensions: event-scoped (per event, like Adobe props) OR user-scoped (persistent for the user, similar but not identical to long-expiry eVars) OR item-scoped. 50+25+10 limits.

The migration challenge: Adobe properties using 100+ eVars with various expiration windows can't all fit into GA4's User-scoped 25 dimensions. Audit which eVars are actually used in active reports. Most properties have 30-50% unused.

Calculated Metrics

Adobe lets you build custom metrics in the UI: (Revenue - Refunds) / Visits, conditional metrics, ratio metrics, time-based metrics. These are first-class objects, queryable in any Workspace project.

GA4 has Calculated Metrics (introduced 2024) but the feature is more limited: simple arithmetic on event-scoped metrics, no conditional logic, no advanced functions.

The fix: most calculated metrics need to be re-implemented in your reporting layer (Looker Studio formulas, BigQuery views) rather than within GA4 itself. Plan for this.

Segments

Adobe Segments are powerful — based on Visit, Visitor, or Hit container, with complex AND/OR/SEQUENCE logic. Reusable across Workspace projects.

GA4 has Audiences and Comparisons but with simpler logic. Sequence-based segments (visitor did A then B then C) require Path Exploration or BigQuery.

The migration pattern: identify the most-used segments in Adobe (probably 10-30 across the team). Re-implement the simple ones as GA4 Audiences. Re-implement the complex ones in BigQuery.

Virtual Report Suites

Adobe lets you create filtered views of your data — "EMEA only", "Brand X only", "Tablet visitors only". Each VRS appears as its own report suite with all metrics scoped to the filter.

GA4 has filtered Sub-properties (Analytics 360 only) and comparison filters in standard reports. Standard tier doesn't have anything quite like VRS.

For 360 customers, sub-properties are the closest equivalent. For standard tier, BigQuery + Looker Studio with shared filtered views is the workaround.

Marketing Channels and Classification Rules

Adobe lets you classify campaigns and traffic sources via lookup tables uploaded to Adobe. "Tracking code starts with 'em-' becomes Email channel."

GA4 has channel groups (default + custom) and Data Import for classifications. The mechanics differ — Adobe's Classification Rule Builder doesn't translate directly. Plan to rebuild classification logic in GA4 channel groups + custom dimensions or in BigQuery.

The four-phase migration

A clean Adobe → GA4 migration has these four phases. Don't skip any.

Phase 1 — Parallel deployment (4-8 weeks)

Both systems running. GA4 collecting full data; Adobe still primary for reporting. The discipline:

• All new tracking is implemented in BOTH systems for the duration of this phase
• No new Adobe-only tracking
• All existing Adobe tracking is mapped to GA4 equivalents (with documented gaps)
• Adobe data continues feeding stakeholder reports

This is the lowest-risk phase. If GA4 isn't capturing something, Adobe's still working. Use this phase to validate GA4's data quality.

Phase 2 — Historical data bridge (4-6 weeks)

Build the BigQuery warehouse that holds both Adobe historical data and GA4 ongoing data:

• Adobe Analytics → Adobe Customer Journey Analytics → BigQuery (or via Adobe's Data Warehouse export)
• GA4 → BigQuery via native export
• Joined reporting tables in BigQuery that normalise schema differences

This phase is engineering-heavy. Plan for 40-80 hours of data engineering work.

The output: a unified data warehouse where year-over-year reports work continuously across the migration cutover.

Phase 3 — Report-by-report rebuild (8-16 weeks)

Identify every Adobe Workspace project and report. For each:

• Determine if it's still needed (many aren't — graveyard reports)
• Re-implement in GA4 reports OR in Looker Studio (against GA4 + BigQuery)
• Validate with the report owner
• Migrate stakeholder access to the new report

Plan for 40-80 hours of report engineering work, plus stakeholder validation time.

The discipline: don't decommission an Adobe report until its GA4 replacement has been signed off by the report owner.

Phase 4 — Adobe decommissioning (90-day window)

Once Phase 3 is complete, give stakeholders a 90-day window where Adobe is read-only and GA4 is primary:

• Stakeholders use GA4 reports for daily work
• Adobe is available for historical reference only
• Issues with GA4 reports get escalated and fixed
• After 90 days, formally decommission Adobe (cancel contract, retain BigQuery historical export)

The 90-day window catches edge-case reports that didn't get migrated in Phase 3. It's cheap insurance.

What 80-200 engineering hours actually cover

The realistic time breakdown for an enterprise Adobe → GA4 migration:

A typical enterprise migration costs $30,000-$80,000 in agency or consulting time, plus internal team hours equivalent. Plan accordingly.

What gets lost (be honest)

Six things that don't survive the migration cleanly:

1. Some calculated metrics — complex ones can be re-implemented but the experience is worse in GA4/Looker Studio than in Adobe Workspace.
2. Sequence-based segmentation — GA4 lacks native sequence segmentation. Path Exploration is the workaround but isn't reusable.
3. Adobe-specific data integrations — Adobe Audience Manager, Adobe Target integrations need separate migration.
4. Custom-trained Adobe attribution — Adobe's flexible attribution modelling is more powerful than GA4's; some attribution analyses become harder.
5. Workspace project investments — years of stakeholder workflow built around Workspace UX. People miss it.
6. Some level of historical analysis — even with BigQuery bridge, schema differences mean some pre-migration reports can't be exactly reproduced post-migration.

Set expectations honestly. GA4 is not a strict superset of Adobe Analytics. Some capabilities are lost; others (like native Google Ads integration) are gained.