Episode 2: How to Prepare for the GA4 Migration with Brie Anderson

Key Takeaways

• Google was forced to scrap Universal Analytics because of global data privacy legislation
• GA4 uses a completely different event-based data structure — numbers will not match Universal Analytics one-to-one
• Go through every option in the GA4 property settings, including hidden features like Google Signals for demographics and IP filtering buried deep in tag configuration
• For multi-location brands, GA4 replaces views with audiences — create audience segments for each region or location and build custom reports around them
• GA4 is new for everyone, and spending 10 minutes a day exploring the platform will put you ahead of most marketers

The migration from Universal Analytics to Google Analytics 4 was one of the most disruptive changes in digital marketing infrastructure in a decade. For location marketers managing multi-location brands, franchises, and local landing pages, the shift meant rethinking how they track, report, and attribute performance — all while starting from a blank data slate.

In this episode of the Local Marketing Beat podcast, host Christian Hustle is joined by Brie Anderson, Founder of BEAST Analytics, to break down why Google forced the migration, what is fundamentally different in GA4, how to preserve historical data, the hidden settings every marketer needs to configure, and what multi-location brands specifically need to know about tracking across subdomains and locations.

Timestamps

00:00 Introduction to the GA4 migration and why everyone is talking about it

01:30 Brie’s last local search: drive-in movie theaters in Illinois

02:50 Why Google is forcing this change: GDPR, CCPA, and global data legislation

05:00 What’s different in GA4: event-based data structure and why numbers won’t match

06:40 Setup tips: go through every property setting, including hidden options

09:15 Preserving historical data: CSV exports, BigQuery dumps, and Looker Studio

11:00 Good surprises: predictive analytics and anomaly detection

13:00 Bad surprises: developer-style naming conventions on the front end

14:30 New features: landing page reports, funnel reports, and UI improvements

15:45 Multi-location tracking: audiences replace views, cross-domain setup simplified

18:00 UTM tracking for Google Business Profile locations and on-site search tracking

20:00 Final advice: get comfortable being uncomfortable

Why Google Was Forced to Scrap Universal Analytics

“Almost 70 percent of countries at this point have some form of data legislation. The reality was if Google wanted people to keep using their analytics products, they had to scrap what they had, get rid of that data, and start again.” — Brie Anderson

Brie explains that the GA4 migration was not a product upgrade — it was a legal necessity.

Universal Analytics, built in 2012, collected and stored data in ways that violated GDPR, CCPA, and similar legislation now active in over 70% of countries worldwide. Several countries had already declared Universal Analytics illegal due to how data was collected, stored, and retained.

For location marketers, this context matters because it explains why the migration feels so abrupt and why there is no backward compatibility. Google could not simply update the old platform — the entire data collection model had to change to comply with global privacy law. This also means GA4 is designed from the ground up with privacy constraints, which affects what data is available and how it is accessed.

Everything Is Event-Based — and Numbers Will Not Match

“It’s a completely different data structure. Everything is event-based in GA4. You’re going to see events like session start and first visit. If you’re trying to compare Universal to GA4, just know that data is not going to look one-to-one because the back end is completely different.” — Brie Anderson

Brie addresses the single biggest source of confusion in the GA4 migration: The data will not match. Universal Analytics used a session-based model; GA4 uses an event-based model where every interaction — page views, clicks, scrolls, form submissions — is tracked as an event. The same dimensions and metrics exist, but they are calculated differently and appear in different places.

For brands using analytics to track location performance, this means any year-over-year comparisons that bridge the UA-to-GA4 transition will show discrepancies. Brie’s advice: Accept this, do not waste time trying to reconcile the numbers, and instead focus on establishing new baselines within GA4 that you can measure against going forward.

‍

The Foundation Has to Be Right — Check Every Hidden Setting

“When you’re in the admin section, go through and touch every single one of those options in the property column. There are a lot of things that are kind of hidden — like demographic data, you have to go in and physically turn on Google Signals. IP filtering is buried under data stream, configure tag settings, see all.” — Brie Anderson

Brie highlights that GA4 hides critical configuration options deep within nested menus.

Google Signals — which provides demographic data like age, gender, and interests — is turned off by default and must be manually enabled. IP address filtering, which most marketers use to exclude their own office traffic, is buried several layers deep under data streams and tag configuration settings.

Her recommendation is practical: Open the GA4 admin panel, go through every single option in the property column, expand every subsection, and configure each setting according to your business needs. Even if you set up GA4 years ago, go back and do this again — many features were added after the initial rollout and may not be configured in older properties.

Multi-Location Brands: Audiences Replace Views

“You don’t have views in GA4. What I encourage people to do is create audiences. You can essentially create the same sort of filters that you were using for views in audiences, and then create custom reports that only bring in those specific audiences.” — Brie Anderson

For multi-location brands and franchises, one of the biggest structural changes in GA4 is the elimination of views. In Universal Analytics, marketers commonly created separate views for each region, city, or location to isolate performance data. That functionality no longer exists.

Brie’s workaround: Create audiences in GA4 that replicate the filters you previously used for views, then build custom reports that pull data only for those specific audiences. This achieves the same segmentation but requires a different workflow.

She also notes that cross-domain tracking is significantly easier in GA4 — as long as the same measurement tag is installed on all subdomains, GA4 automatically recognizes them as part of the same property. For brands using local pages with subdomains, this simplifies what was previously a complex configuration process.

Get Comfortable Being Uncomfortable — GA4 Is New for Everyone

“Be ready to be uncomfortable and get comfortable being messy. It’s completely new for everyone. Even if it’s just 10 minutes a day — ask yourself a question, go into GA4 and find it. If you do that enough, you’re going to be light years ahead of everybody else who’s just avoiding it.” — Brie Anderson

Brie closes with advice that applies to any major platform migration: The discomfort is universal. GA4 looks different, feels different, and uses naming conventions that resemble a developer tool more than a marketing platform. But the same data is available — it is just organized differently and requires learning new navigation patterns.

Spend 10 minutes a day in GA4 asking yourself one question and finding the answer. Over time, this builds the muscle memory and familiarity that turns GA4 from intimidating to routine. She also highlights two features worth exploring: predictive analytics (which uses machine learning to identify users likely to purchase in the next 28 days) and anomaly detection (which automatically flags statistically significant changes in your data). Both are available for free in GA4 and represent capabilities that did not exist in Universal Analytics.