Service · 05
Measurement and analytics: numbers you can defend in a board meeting
GA4 and analytics consulting covering GA4, GTM and server-side tagging, attribution and incrementality. Most reporting problems are measurement-design problems, and no dashboard fixes a measurement plan that was never written.
72 named engagements since 2012 · SEO Head at Eye Advertising · SEO & GEO instructor at Learn N’ Digital · Arabic and English
The problem
GA4 dashboards are rarely the problem
When a marketing team says their reporting is broken, the request that arrives is usually for a better dashboard. It almost never helps, because the dashboard is faithfully displaying numbers that were never designed to answer the question being asked of them.
The failures I find are upstream:
- No measurement plan — nobody wrote down what a conversion is before the tags went in, so three teams count it three ways.
- GA4 configured by migrating Universal Analytics habits, so events exist but the model underneath them does not.
- Consent implemented as a banner rather than as a data model, so a large share of traffic is missing and nobody has quantified how much.
- Attribution arguments that cannot be settled because the platforms are each counting their own assisted conversions.
- Organic performance reported as sessions, which no CFO has ever made a decision with.
What is included
What GA4 and analytics consulting actually delivers
Measurement plan and KPI definition
Written before anything is implemented: what the business decision is, what metric answers it, how that metric is defined, and who owns it. Boring, and it is the deliverable that saves the most money.
You get: a measurement plan document with every metric defined in one agreed way
GA4, GTM and server-side implementation
Event and parameter design that matches the plan, container structure that someone else can maintain, consent handled as a data model rather than a banner, and server-side tagging where the data loss justifies it.
You get: a configured, documented implementation plus a container spec
Attribution and reconciliation
Why the platforms disagree, which number to use for which decision, and where incrementality testing is worth the effort instead of arguing about last-click.
You get: a reconciliation note and an agreed attribution position per channel
Reporting that ties to outcomes
Organic performance expressed as pipeline, revenue or cost avoided, in the language your board already uses.
You get: a reporting template with the definitions attached, so the numbers survive a challenge
How it runs
The four stages of a GA4 measurement build
01 · Week 1
Decisions first
What decisions the reporting has to support. Everything else follows from this and it is usually the conversation nobody had.
02 · Week 1–2
Plan
Metric definitions, event design, consent model and ownership, agreed in writing before implementation.
03 · Week 2–4
Implement
GA4, GTM, consent and server-side where warranted — built to the plan, then documented so you are not dependent on me.
04 · Post-implementation
Validate
Debug, reconcile against platform and back-office numbers, and record the known gaps rather than pretending they do not exist.
Scope
What GA4 and analytics consulting does not cover
This is the service where a solo consultant is most often the wrong answer, and I would rather say so here than in a proposal. A full enterprise data-warehouse build with a modelling layer and a BI team is not what I do.
What I do well is the part that goes wrong most: designing the measurement before it is implemented, getting GA4 and GTM to reflect it, settling the attribution argument with evidence, and producing reporting that survives a challenge from finance. If your problem is genuinely a warehouse problem, I will tell you that.
Questions
GA4 and measurement questions clients ask
Both, though I would rather implement to a plan than inherit an implementation. If GA4 is already in place I audit it against what you need it to answer, then fix the gaps rather than rebuilding from scratch.
Only if you have quantified the data loss and it matters to a decision you are making. It adds cost and maintenance, and plenty of teams are sold it before anyone has measured whether client-side collection is actually failing them.
As a data model, not a banner. What is collected in each consent state, what that does to your reported numbers, and how to report honestly with a known gap — rather than reporting a partial dataset as if it were complete.
I can make it evidential rather than rhetorical: show why each platform reports what it does, recommend which number to use for which decision, and identify where an incrementality test would actually resolve it. Nobody can make every platform agree, and anyone promising that is selling something.
Directly. Without agreed measurement, an SEO or GEO engagement has no defensible before-and-after, and you are left arguing about whether it worked. I set the baseline before the work starts, which is also how I keep myself honest.
Usually, yes. Most of my measurement work is design and diagnosis — the plan, the event model, the reconciliation — while your team or your agency implements and maintains it.
Free AI visibility audit
Request a GA4 measurement baseline
The audit establishes where you stand before any work begins, which is the number every later report gets measured against.
Three working days, by email. No call unless you want one, and no follow-up sequence. What the audit covers.
Thank you for your response. ✨
Start your measurement work with a free AI visibility audit
Fixed scope, fixed price. You get the findings whether or not we work together afterwards.
Related:
Technical SEO & Site Architecture
Generative Engine Optimization
Free AI visibility audit