Tracking Audit Consultant: what I detect, what I deliver, at what price
Freelance tracking audit consultant: GA4, GTM, server-side and CMP audit. 4-phase methodology, detailed deliverables, transparent fixed pricing.
By Ron Kopelman, freelance analytics consultant — updated May 18, 2026
A tracking audit is a complete diagnostic of your data collection setup — GA4 tags, GTM container, server-side, Consent Mode, dataLayer, tagging plan. My role as a tracking audit consultant is to detect every anomaly causing your business data to drift from reality, document each one, and deliver a prioritized, scoped action plan. The audit is never an end in itself: it opens the next phase — rebuilding your GA4 setup, migrating to server-side tracking, or bringing Consent Mode v2 into compliance.
Why audit your tracking in 2026
Three structural shifts explain why every site I audit has lost between 15% and 40% of its tracked conversions without noticing.
Consent Mode v2 has been mandatory since March 2024 to maintain access to Google Ads remarketing and modeled conversions. Half the sites I audit have a Consent Mode “declared” in the CMP but which never pushes gtag('consent', 'update', ...) signals at the right moment, or which sends them after tags have loaded. Result: lost conversions, modeled data not eligible.
The phase-out of third-party cookies and tightening of Safari ITP / Firefox ETP have neutralized classic pixels for 20–35% of users depending on browser mix. Without server-side tracking, a significant share of data no longer reaches ad platforms — and most sites don’t know it.
The GA4 migration left damage in roughly half the configurations I encounter: events poorly named, conversions imprecise, undeclared custom parameters, custom dimensions without values in BigQuery. An audit surfaces these accumulated debts.
The practical rule: if you haven’t audited your tracking in the last 18 months, your business numbers no longer match reality. It’s mechanical.
My audit method in four phases
The same protocol on every engagement, whether the site has 500K sessions or 30 million. Only scope adapts.
Phase 1 — Technical crawl and tag inventory
I inventory every tag actually deployed on your site: Google Analytics, ad pixels (Meta, LinkedIn, TikTok, Pinterest, Microsoft), third-party tools (Hotjar, AB Tasty, Contentsquare, Intercom). For each tag, I verify on which pages it fires, at which moment, with which parameters, and whether the CMP correctly blocks it before consent.
Tools: Screaming Frog, ObservePoint for high-volume audits, GTM/GA Debug extension for manual validation of critical journeys.
Phase 2 — GTM container audit
Full review of container organization: tag and trigger naming, variable structure, detection of orphaned or duplicated tags, validation of firing rules. Common findings: built-in variables misconfigured (Click URL used instead of Click Element URL), which is a classic source of silently broken tracking.
If server-side is in place, the sGTM container is audited too: routing, claims, event transformations, Meta CAPI integrations, Enhanced Conversions for Google Ads.
Phase 3 — GA4 review
I document the GA4 configuration property by property: data streams, conversions, audiences, custom parameters and dimensions, referral exclusions, internal filters, attribution. I verify that recommended events are correctly implemented (purchase, add_to_cart, generate_lead, etc.) with all expected parameters, and that conversions flow with a non-null value.
If BigQuery is connected, I audit the exports: presence of ga_session_id, user_pseudo_id, cleanliness of event_params, completeness of custom dimensions.
Phase 4 — Acceptance tests on critical journeys
This phase catches the real gaps: I replay conversion journeys in private browsing, on mobile, with consent refused. For an e-commerce: product search → product page → add to cart → checkout → confirmation. For B2B lead gen: key pages → form → confirmation → CRM relay. I note at each step which event fires, with which parameters, and where there’s drift between declared and actual.
What I detect in 90% of audits
After running missions across very different advertisers, I see the same problems recur with depressing regularity.
- Mis-wired Consent Mode:
consent defaultsignals fired after GTM loads, or never updated after acceptance. Consequence: Google Ads sees everything as “denied” and loses modeled conversions. - Duplicated pageviews: GA4 tag fired both by “Initialization” and “Page View” in GTM, or by source code and GTM tag simultaneously.
- Conversions without
value:purchaseorgenerate_leadevents firing without avalueparameter, invisible to modeling and uncomputable for attribution. - Broken mobile dataLayer: data layer present on desktop but empty or delayed on mobile.
- Ad tags outside the CMP: Meta Pixel or Floodlight loading before consent — direct GDPR/CNIL violation.
On a recent audit of an e-commerce with ~€10M tracked revenue, I documented 23 anomalies. After correction: 27% recovery of Google Ads conversions and 19% of Meta conversions — without changing a single line of application code, just by fixing tracking.
What you receive: the deliverable
A PDF + shared Notion containing:
- Executive summary (1 page): three major findings, quantified business impact, estimated effort
- Tag inventory with Consent Mode conformity per tag
- Per-anomaly fact sheet: description, screenshots, parameter involved, severity, business impact, recommended fix with exact code or GTM configuration
- Prioritized action plan: business impact × technical effort matrix
- Validation checklist to confirm fixes are effective
You walk away with an actionable document your developer or marketing analyst can execute without me.
Fixed audit fees
| Audit | Scope | Fixed fee |
|---|---|---|
| Essential | Brochure or single-country e-commerce, GA4 + web GTM + 1 CMP | €1,900 (~3 days) |
| Pro | Multi-country e-commerce or B2B lead gen + CRM, server-side, Consent Mode v2 | €3,200 (~5 days) |
| Platform | Media / SaaS / multi-domain, BigQuery, sGTM, Ads/CRM/CDP integrations | €5,500 (~8 days) |
Price includes the full report, the restitution session (1h video), and 30 days of email support. No hidden costs. For multi-brand or international groups, custom quote.
Frequently asked questions
How long does an audit take?
Three to eight days of consulting time on a 2-4 week calendar window. Essential audit: 2 weeks. Platform audit: 3-4 weeks.
Do you work in our environments or yours?
Always yours. I never use mirror or test accounts — production reality is what matters. You grant me read access to GTM, GA4, Search Console and BigQuery if applicable. NDA available if your legal team needs one.
Do you implement the fixes?
The report is designed so your dev or marketing team can implement directly. If you want me to drive implementation, that’s a separate engagement at 2-8 days depending on volume.
And if you find nothing wrong?
In six years across 60 audited sites, I’ve always found at least 8-10 anomalies. If it never happens, the report documents the clean state and you leave with a benchmark.