Skip to content
VERQEN — home

Service 05 · Service

Analytics audits: know whether your numbers can be trusted.

Most measurement problems are not missing dashboards. They are duplicate events, conversions counted twice, consent silently dropping data, and a CRM that disagrees with GA4 by 40%. We test the whole chain with real actions and tell you what is actually true.
Audit my analyticsSee how we work
Timeline
Audit 1–2 weeks; implementation 2–4 weeks
Starting from
From $2,500
Shape
Audit, then implementation, then optional monthly monitoring
Deliverables
8 items

The problem

You are making decisions on numbers nobody has verified.

Budget gets moved, pages get changed and agencies get judged on data that was configured once and never tested. When GA4 and the CRM disagree, teams usually pick the number that supports the decision they already wanted.

You probably recognise some of these

  • GA4 conversions and CRM leads differ by a margin nobody can explain.
  • The same conversion appears to fire twice.
  • Paid and organic both claim the same conversions.
  • Traffic looks fine but “direct” is enormous.
  • A consent banner went live and reported traffic changed overnight.
  • Nobody can say which events are real and which are legacy.

Who needs this

Is this for you?

  • Companies whose GA4 and CRM numbers do not reconcile
  • Teams that migrated from Universal Analytics and never validated the result
  • Businesses running meaningful paid spend on unverified conversion data
  • Multi-domain or multi-subdomain setups
  • Companies implementing consent management in regulated markets
  • Anyone about to build dashboards on data they have not audited

Why it happens

What usually causes it

  1. GA4 configured by following a checklist rather than modelling the actual business.
  2. GTM containers that accumulated tags nobody owns.
  3. Conversions defined on page views instead of confirmed actions.
  4. dataLayer implemented inconsistently across templates.
  5. Cross-domain tracking not configured, so one journey becomes two sessions.
  6. UTMs applied without governance, so channel data is unreliable.
  7. Consent mode implemented incorrectly, so data is dropped rather than modelled.
  8. No CRM connection, so nothing validates the front-end numbers.

Scope, in detail

What exactly we check

Not a category list — the actual checks. If something is not on here and you need it, say so and we will tell you whether it fits the scope.

Collection

  • GA4 property, stream and data-retention configuration
  • GTM container inventory — every tag, trigger and variable
  • dataLayer implementation across templates
  • Duplicate and double-firing events
  • Bot and internal traffic filtering
  • Data quality: (not set), unassigned, self-referrals

Conversions

  • Every conversion definition against the real business action
  • Form starts versus submissions versus confirmed leads
  • Ecommerce events and revenue accuracy
  • Phone, chat and booking interactions
  • Google Ads and Meta conversion imports
  • Offline and CRM-side conversions

Attribution & governance

  • Attribution model and reporting window
  • Channel grouping accuracy
  • UTM taxonomy and enforcement
  • Cross-domain and subdomain configuration
  • Consent mode behaviour and its measured effect
  • CRM field mapping and lead-source capture

Live testing

  • Real form submission end to end
  • Real purchase or signup end to end
  • Phone click, booking and download events
  • Website → GTM → GA4 → Ads → CRM reconciliation
  • Consent granted and consent denied paths

Deliverables

What you actually receive

Every important issue arrives in the same format: problem, evidence, why it matters, root cause, solution, exact actions, owner, priority and measurement.
Audit my analytics
  • Analytics audit with each issue as Problem → Evidence → Impact → Exact fix
  • A measurement plan mapped to the business: what counts as a lead, a qualified lead, a customer
  • Event and conversion specification, including dataLayer contracts for developers
  • GTM container cleanup plan — what to remove, what to rebuild, in what order
  • Cross-domain, UTM and consent configuration requirements
  • CRM connection and lead-source capture design
  • Reconciliation report: GA4 versus Ads versus CRM, with the explained gap
  • A dashboard that answers decisions, not one that displays everything

Process

How the work runs

Each step has an exit condition. We do not move on because time passed.

  1. Model the business first

    Before touching a tag, we define what actually matters: what a lead is, what a qualified lead is, what a customer is worth, and which actions are worth measuring at all.

    • Lead definition
    • Value model
    • Decision list

  2. Audit what exists

    Full inventory of GA4, GTM, ads platforms and CRM fields. Every event and conversion gets classified: correct, wrong, duplicate, or legacy.

    • GA4
    • GTM
    • Ads
    • CRM fields

  3. Test with real actions

    We submit real forms, complete real purchases, click real phone links, and follow each one through the whole chain. If the numbers do not make sense, we investigate before reporting.

    • Real submissions
    • Chain tracing
    • Consent paths

  4. Fix and rebuild

    Implement the corrected event and conversion set, clean the container, and connect the CRM — in a staged order so each change can be verified independently.

    • Staged releases
    • Verification per change

  5. Reconcile and document

    Report the remaining gap between platforms with an explanation for it, and hand over documentation so the setup survives staff changes.

    • Reconciliation
    • Documentation
    • Handover

What we need from you

Access and data

We ask for read access only, and only for what the question needs.
  • GA4 (editor access for implementation, read for the audit)
  • Google Tag Manager container access
  • Google Ads and Meta accounts, if paid spend matters
  • CRM access or a lead and revenue export
  • A test path — a form we can submit, or a test purchase route
  • Your definitions: what counts as a lead, and what a customer is worth

Measurement

What results can be measured

Agreed before the work starts, so nobody argues about the definition afterwards.
  • Tracking accuracy — the measured gap between GA4 and CRM, before and after
  • Duplicate conversion rate eliminated
  • Share of sessions with a resolvable channel and source
  • Conversion-to-CRM match rate
  • Consent-path data completeness
  • Number of business decisions the dashboard can actually answer

Afterwards

What happens when the project ends

You get a documented measurement setup and the reconciliation baseline. Measurement drifts every time the site or the tag stack changes, so many clients keep a monthly check that catches breakage in weeks rather than quarters.
How scoping and pricing works

Why trust us

How we keep ourselves honest

  • We test with real actions rather than reading the interface and assuming.
  • When the numbers do not make sense, we investigate before reporting — we do not pass an unexplained number to you.
  • We will tell you when the honest answer is “this data cannot be trusted for that decision”.

Honest limitations

What this service will not do

Stated up front, because finding out later is worse for both of us.
Some historical data cannot be repaired. We can fix collection going forward and explain the break, but we cannot recreate what was never collected.
Consent and privacy regulations set a hard floor on what is measurable. We work within them; we do not offer ways around them.
Perfect attribution does not exist. We aim for numbers that are reliable enough for the decisions you actually make.

Quick answers

Analytics & Revenue Measurement in nine questions

The same facts, stated plainly. Useful if you are skimming, and useful to any system trying to understand what we actually offer.
What is this service?
Make the numbers trustworthy enough to make business decisions with — GA4, GTM, ads, consent and CRM.
Who is it for?
Companies whose GA4 and CRM numbers do not reconcile; Teams that migrated from Universal Analytics and never validated the result; Businesses running meaningful paid spend on unverified conversion data; Multi-domain or multi-subdomain setups; Companies implementing consent management in regulated markets; Anyone about to build dashboards on data they have not audited.
What problem does it solve?
You are making decisions on numbers nobody has verified.
What do you check?
4 areas covering collection, conversions, attribution & governance, live testing.
What do you deliver?
Analytics audit with each issue as Problem → Evidence → Impact → Exact fix; A measurement plan mapped to the business: what counts as a lead, a qualified lead, a customer; Event and conversion specification, including dataLayer contracts for developers; GTM container cleanup plan — what to remove, what to rebuild, in what order; Cross-domain, UTM and consent configuration requirements; CRM connection and lead-source capture design; Reconciliation report: GA4 versus Ads versus CRM, with the explained gap; A dashboard that answers decisions, not one that displays everything.
How long does it take?
Audit 1–2 weeks; implementation 2–4 weeks
How much does it cost?
From $2,500. Audit, then implementation, then optional monthly monitoring.
What access is needed?
GA4, GTM, ads accounts, CRM, and a test conversion path
What happens after?
You get a documented measurement setup and the reconciliation baseline. Measurement drifts every time the site or the tag stack changes, so many clients keep a monthly check that catches breakage in weeks rather than quarters.

Questions

Questions about analytics & measurement

Something else? hello@verqen.com

Evidence and further reading

Worked investigations and working notes on this subject.

Related services, problems and industries

The problems this service usually solves, and where we see them most often.