Tracking audit
Tracking audit: every gap between reality and your analytics
Built the growth engine that took cnvrg.io to its acquisition by Intel.
I personally investigate the agreed campaign and measurement setup: events, tags, server delivery, naming, attribution settings and customer-source records where relevant. You receive source-linked findings, severity and decision impact, reproduction details and a scoped repair plan. Access and coverage limits remain explicit; a tracking audit cannot promise to find every possible gap.
- Typical audit finding3-7critical tracking gaps per engagement across paid channels
- Most common root causedLbroken dataLayer push on checkout - missed by every pixel
- Across engagements$100M+in total ad spend managed across roles - each one required defensible measurement
The full scope of a tracking audit
"If you cannot trust your tracking, every optimization after it is a guess with a dashboard attached."
Yaniv Goldenberg, fractional growth operator
Pixel inventory
Every pixel on your site: Meta Pixel, GA4, Google Ads tag, TikTok Pixel, LinkedIn Insight Tag. Which ones fire? On which pages? Are they firing duplicates? Are they missing on key conversion pages?
DataLayer audit
Your dataLayer is the source of truth for all event-based tracking. I review every push: correct event names, complete parameters, right timing, no missing items on purchase events. A broken dataLayer means every platform tracking the same events is broken simultaneously. To record tracking evidence before treating analytics as reliable, see GA4 and GTM tracking evidence checker.
Server-side event coverage
I inspect the actual event and destination configuration, permissions, source records and deduplication where relevant. Browser and server delivery can observe different parts of a journey. I test the supported receipt and document gaps; an audience profile alone cannot quantify missing conversions.
UTM hygiene
Consistent naming conventions, complete tagging on all paid campaigns, no accidental tagging on organic links, no encoding errors. UTM problems silently corrupt channel attribution - paid traffic looks like organic, campaigns look like direct.
Attribution window review
Record the actual attribution settings, source definitions, observation windows and customer journey, then compare like with like. A seven-day or thirty-day window can serve different questions; the configured model does not become truth or automatically wrong because the buying cycle differs.
Cross-channel reconciliation
Compare like with like: platform-claimed conversion counts with matched conversion records, and attributed value with the corresponding billing or CRM revenue. Check windows, identifiers, refunds and duplicate rules. Do not compare a count of conversions directly with an amount of revenue or infer a universal failure threshold.
Audit deliverables
A tracking audit delivers a findings document - not a generic checklist, a specific list of what is broken in your stack with severity ratings and fix specifications.
Each finding includes: what is wrong, which campaigns or decisions it is affecting, what a correct implementation looks like (specific code or configuration), and an effort estimate for fixing it. You can hand this to your dev team or have me implement it.
The findings include the source definitions, eligible denominator, matches, missing observations and uncertainty that access supports. A before-and-after event or match-rate check can validate a repair. Economic return still needs actual costs, contribution and an appropriate comparison.
A tracking repair can improve decision evidence, but neither a gap nor a first budget change proves the audit paid for itself. Evaluate actual intervention costs, eligible customer contribution and the comparison or observation window needed for the decision.
Next: full-funnel growth leak analysis
A broader private diagnostic can examine agreed acquisition, conversion, lifecycle and economic questions. The free linked AI visibility assessment is a separate public-surface check; it does not inspect your private tracking or establish every point where revenue is lost.
The free AI visibility assessment reviews selected public-surface readiness. A private tracking investigation needs its own authorized sources, scope and evidence; this assessment does not diagnose private measurement.
When to start where
Tracking audit vs GA4 audit vs attribution consulting
Start with a tracking audit when
- You run several paid channels and cannot explain how campaign conversions relate to the matching CRM or billing records. You need a documented source comparison before deciding whether to commission a larger repair.
Start narrower when
- You only run Google channels - a GA4 audit is more focused and faster. You already know server-side is missing - go straight to a server-side tracking implementation. You want strategy plus measurement in one engagement - that is marketing attribution consulting.
Connected specialties
FAQ
Frequently asked questions
What is a tracking audit?
How is a tracking audit different from a GA4 audit?
What do you look for in a UTM audit?
What is a dataLayer audit?
How much does a tracking audit cost?
Can you fix the problems the audit finds, or only report them?
Both are possible. Each finding includes what a correct implementation looks like, as specific code or configuration, so you can hand it to your dev team or have me implement it. Implementation of fixes is quoted separately from the audit, and the audit itself is agreed for the scope.
What happens on the call before a tracking audit?
I ask about your stack and your channels, and tell you whether you need a full tracking audit or a narrower GA4 audit first. The call is a 15-minute discovery call where I assess your stack complexity, and a fixed quote follows it.
Why do attribution windows matter in a tracking audit?
Inspect each source’s actual configured windows, event and customer definitions and timing, then reconcile comparable records. A seven-day or thirty-day setting serves a defined reporting question; no universal business-cycle rule makes a source’s attributed total the truth. Document the differences and decision limits.
How do you check whether my ad platforms are double-counting conversions?
I compare conversion counts with matched conversion records and attributed value with corresponding billing or CRM revenue, using agreed definitions and windows. I test event identifiers and deduplication where access allows, then document missing or unmatched records. A count-to-revenue comparison or an arbitrary discrepancy threshold cannot establish that tracking is broken.
Next step
Find out what your tracking is missing
Book a call. I will ask about your stack and your channels, and tell you whether you need a full tracking audit or a narrower GA4 audit first.
What slips through
Most common tracking audit findings
Double-counted purchases
A browser tag and a server can report the same purchase. Check the event identity and destination deduplication rules against the source record before interpreting ROAS. Duplicate collection can distort reporting, but its size must be measured for the actual implementation.
UTM stripping on redirect
Illustrative failure path: an email link passes through a redirect that drops campaign query values. Test the actual destination and eligible source records before attributing missing credit to that redirect. I reproduce the fault, scope and test the approved repair and verify receipt; no fixed twenty-minute remedy or resulting revenue change is promised.
Attribution window mismatch
Inspect and record the actual Meta, GA4 or other source configurations rather than assume platform-wide defaults. Compare eligible records, attribution definitions, windows, time zones, duplicates and refunds, retaining unmatched journeys and unknowns.
Tracking configuration reference: Google Analytics docs · My role history: Yaniv Goldenberg on LinkedIn
Work with Yaniv
Fractional Head of Growth for B2B SaaS, B2C, and e-commerce. Revenue that compounds, not spikes. Any channel, any motion, any stage.
Explore the Marketing Attribution Consultant hub