Adobe Analytics vs GA4 for B2B attribution: an honest take

Before I say anything critical of GA4, I'll acknowledge the obvious: GA4 is free. Adobe Analytics costs $30,000-200,000 per year depending on traffic volume and features.

That price gap means most companies default to GA4 regardless of whether it's the right tool. And for a lot of companies, it's fine. E-commerce, SaaS with short sales cycles, content publishers. GA4 works.

For a $3.78M industrial B2B pipeline with 9-month sales cycles and multi-stakeholder procurement? GA4 fails in predictable and expensive ways.

What GA4 gets wrong for B2B

Session stitching across long time horizons. GA4's user identity model relies on cookies (and Google accounts when available) to stitch sessions into a unified user journey. In a 9-month sales cycle, even the Google account-based stitching has gaps. Corporate device switches, new browser installs, IT policy changes — all break the session thread.

GA4 might see 4 separate users where there's actually 1 buyer at 4 points in a 9-month journey.

Multi-stakeholder company attribution. GA4 tracks at the user level, not the company level. When three people from the same company visit your site over six months, GA4 sees three users. It has no concept of "these are all Acme Corporation, and together they represent one procurement process."

This is the same 70% overstatement problem I've described elsewhere, but at the analytics layer rather than the conversion layer.

Data retention. GA4's default retention window is two months. Maximum is 14 months (with the blended model). B2B sales cycles regularly exceed 14 months for large enterprise deals.

By the time a deal closes, the early touch data might be outside the retention window. You can't analyze the full journey.

Event model complexity. GA4's event-based model is flexible but requires significant configuration to capture meaningful B2B signals. "Demo request submitted" needs a custom event. "Spec sheet downloaded" needs a custom event. "Engineering contact form submitted" needs a custom event. Each one requires implementation work.

GA4 without heavy custom event implementation is basically a traffic counter. It tells you pages were visited but not what mattered.

What Adobe Analytics does instead

Adobe Analytics uses a dimension/metric architecture built for complex behavioral analysis. More importantly, it was designed for enterprise use cases where sessions are long, journeys are complex, and user identity is ambiguous.

Persistent eVars. Custom variables that persist across sessions indefinitely (until expiration, which you control). A first-touch campaign value captured in January stays associated with that user through October.

Visit and visitor distinction. Adobe natively distinguishes between a single visit and the aggregate visitor history. Analyzing a buyer's entire history across all their visits is a first-class operation.

Company-level attribution. With proper implementation, Adobe can store company domain or Salesforce account ID as a custom dimension, enabling company-level attribution that maps directly to your CRM.

Workspace flexibility. Adobe's Analysis Workspace allows you to build custom attribution models, journey visualizations, and cohort analyses that aren't possible in GA4's native interface.

None of this is free. Adobe's implementation requires dedicated analytics engineers. The learning curve is steep. The license cost is real.

The honest recommendation

Use GA4 if: You're managing under $500K/year in B2B paid media, you have a sales cycle under 90 days, or you don't have the budget/team to implement Adobe properly.

Supplement with BigQuery exports and manual attribution analysis. Accept that your journey data is incomplete. Make decisions on directional data.

Use Adobe Analytics if: You're managing a complex pipeline above $500K/year with long cycles, you have the budget, and you can staff a dedicated analytics implementation. The data quality difference is real and compounds in impact over time.

Don't do neither. The worst outcome is having neither tool implemented properly and making $650K/month budget decisions based on Google's native reporting. That's what I see most often. And it's the most expensive mistake of all.