At DARM 2025, Jeff Sauer, Co-founder, Measure U, delivered a timely message to revenue and marketing leaders: ad tracking, as many teams still understand it, is broken.¹
The shift did not happen overnight. It unfolded quietly across browser updates, privacy laws, operating system changes, and the steady erosion of third-party tracking. What worked reliably just three years ago now produces partial truth — incomplete attribution, inconsistent reporting, and underfed ad algorithms.
The result is not just data confusion. It is budget hesitation.
As Sauer explained, when marketing teams cannot confidently attribute bookings, they pull back on spend. And when platforms receive weaker conversion signals, smart bidding systems optimize against flawed inputs.
The feedback loop deteriorates.
The Core Problem: Client-Side Tracking Loss
Traditional “sprinkle pixels everywhere” tracking relies on client-side execution — meaning scripts fire within the user’s browser. Today, between 25–50 percent of that signal can be lost due to:
- Apple’s App Tracking Transparency (ATT) framework
- Browser-level blocking (Safari, Firefox, Chrome policy changes)
- Ad blockers (now used by roughly one-third of internet users)
- Expanding state-level privacy regulations
When ad platforms only see a fraction of actual conversions, performance appears weaker than reality. Sauer illustrated a simple example: if 100 bookings originate from Google Ads but only 60 are attributed correctly, the platform optimizes around 60 signals, not 100.
Less signal means weaker optimization.
Weaker optimization means rising cost per acquisition.
The Upgrade: Server-Side Tracking
Sauer’s recommended structural fix is server-side tracking.
Instead of relying on browser-executed scripts, server-side tracking routes conversion data through controlled server infrastructure before passing it to ad platforms via APIs (such as Meta’s Conversions API).
The implications are significant:
- Higher data resilience against browser blocking
- Improved conversion recovery
- Stronger bidding signal to advertising platforms
- Greater control over what data is shared and retained
In practice, this restores confidence. Cleaner data strengthens attribution. Stronger attribution supports smarter allocation decisions.
For teams spending more than $5,000 per month on advertising, Sauer argues that server-side infrastructure is no longer optional — it is foundational.¹
Clean Data as a Framework
Sauer also introduced a five-part clean data model:
- Collect meaningful data
- Label it clearly
- Enhance it through system integration
- Assemble dashboards that align with business goals
- Narrate insights for action
While most teams focus heavily on collection, fewer prioritize enhancement and narration. Yet those final steps are what transform data from reporting into decision-making.
AI-Powered Reporting: The Emerging Layer
Perhaps the most forward-looking portion of the session centered on AI integration.
Through Model Context Protocol (MCP), AI assistants can now securely access marketing tools such as Google Analytics and Google Ads, allowing users to query live data conversationally rather than navigating traditional dashboards.
Instead of logging into multiple systems, teams can ask:
- Which campaigns delivered the highest revenue per booking last week?
- Is there a drop-off in the booking funnel?
- Which channels are underperforming relative to spend?
The technology is new — introduced in early 2025 — but Sauer demonstrated how AI-assisted reporting dramatically reduces friction in analysis.
The critical caveat: AI is only as effective as the data it accesses. Without clean, server-side reinforced data, AI simply analyzes flawed inputs faster.
The Takeaway for Revenue Leaders
The conversation around AI often overshadows the more pressing infrastructure issue.
Before dashboards become intelligent, the signal must be trustworthy.
Server-side tracking restores conversion visibility.
Clean labeling strengthens interpretation.
AI accelerates insight once the foundation is stable.
The competitive edge will not belong to those who adopt AI first. It will belong to those who combine clean first-party data with AI-assisted decision-making.
