Google Analytics 4 replaced Universal Analytics over two years ago. If you are a B2B marketer who inherited a GA4 property and never fully configured it, you are not alone. Most teams migrated in a rush, accepted the default settings, and moved on. The result is a GA4 setup that technically collects data but does not surface the insights B2B marketing teams actually need. This post covers what changed, what is broken by default, and the specific settings you need to fix today.
What actually changed from Universal Analytics to GA4
The fundamental shift is from sessions to events. Universal Analytics organized everything around sessions and pageviews. GA4 treats every user interaction as an event, including pageviews, scrolls, clicks, and form submissions. This is more flexible but also means your old metrics do not map cleanly to GA4. Bounce rate in GA4 is calculated differently. Sessions are defined differently. And many of the standard reports you relied on in Universal Analytics simply do not exist in GA4 unless you build them yourself.
For B2B teams, the biggest practical impact is that the default GA4 reports are designed for e-commerce and consumer apps. There is no out-of-the-box report showing which blog posts drive demo requests, which landing pages convert best for your ICP, or how organic search performance correlates with pipeline. You have to build those views using Explorations or connect to an analytics tool that does it for you.
What is broken by default in GA4
Several critical settings ship with defaults that hurt B2B analytics. Data retention defaults to 2 months, meaning you lose the ability to analyze trends beyond 60 days. Google Signals is off by default, so cross-device tracking does not work and a user who visits your site on mobile and then converts on desktop shows as two separate users. Internal traffic filtering is not configured, so your own team inflates session counts. Enhanced measurement auto-tracks certain events like outbound clicks and scrolls, but does not distinguish between a scroll on a pricing page and a scroll on a blog post, making the data noisy without additional configuration.
The 5 settings to fix right now
First, extend data retention to 14 months under Admin, Data Settings, Data Retention. Second, enable Google Signals under Admin, Data Settings, Data Collection so cross-device tracking works. Third, define your key conversions. GA4 calls them key events. Mark demo request form submissions, pricing page visits, and signup completions as key events so they appear in your conversion reports. Fourth, set up internal traffic filters under Admin, Data Streams, Configure Tag Settings, Define Internal Traffic. Add your office IP ranges and any VPN ranges your team uses. Fifth, configure content groupings by creating a custom dimension that maps landing pages to clusters like blog, product, comparison, and pricing. This lets you analyze performance by page type instead of individual URLs. For a step-by-step walkthrough of GA4 configuration, see /guides/getting-started-with-ga4.
What you can safely ignore
Not everything in GA4 deserves your attention. Predictive metrics like purchase probability are built for e-commerce and irrelevant for most B2B sites. The default Acquisition reports blend paid and organic in ways that obscure channel-level performance for small teams. The Advertising section is empty unless you run Google Ads. And the real-time report, while satisfying to watch, rarely drives actual decisions for B2B marketing teams with moderate traffic volumes.
When GA4 alone is not enough
Even a well-configured GA4 property requires someone who knows how to build Explorations, segment audiences, and cross-reference with Search Console data. For B2B marketing teams without a dedicated analyst, this is where the system breaks down. ClimbPast connects directly to your GA4 property and Google Search Console, letting you ask questions like "which blog posts drove the most demo requests last month" in plain English and get an answer in seconds without building a custom report. It also monitors your data for anomalies and sends Slack alerts when something changes, so you catch problems before they affect pipeline. See how it compares to native GA4 at /compare/climbpast-vs-google-analytics.