Skip to content
Get a quote

Marketing & ADS

Your data already says what is not working. We make it readable

Correct tracking, testing at the right statistical threshold and attribution that does not get lost along the way, without selling A/B testing to anyone whose traffic cannot yet use it.

  • You have steady traffic, but do not know why it does not convert better
  • You have Google Analytics installed, but nobody has looked at it seriously in a long time
  • You run paid campaigns and want to know what actually works

01 · What it costs you today

  • You have traffic, but nobody tells you why it does not sell more

    A diagnosis problem, not a traffic problem. Tracking installed once at launch and abandoned since answers no real question.

  • You tested a button and do not know what actually happened

    Testing with no statistical discipline: it looks like it works for a week, then everything returns to normal, with no reliable conclusion.

  • The reports show clicks, not money

    The PPC agency sends activity figures every month, but nobody connects them to how much money the campaign made you.

  • The Analytics numbers no longer match reality

    Since the cookie banner appeared, part of your conversions has vanished from the report, sometimes a gap of over 10% of the total, and you do not know how much.

02 · What you get

  • A measurement audit, from objective to event

    What exists in GA4 and GTM, what is missing, and a coherent taxonomy tied to the real business objectives.

  • GA4 and GTM tracking put right

    With consent compliant with Consent Mode v2, not just “installed at some point” and left there.

  • Attribution connected to your CRM

    A disciplined UTM convention, extended conversion windows for long cycles, and every lead's source visible at individual level.

  • An honest assessment of whether you need server-side tracking

    Recommended only where volume and media budget justify it, not sold to everyone out of habit.

  • A testing programme calibrated to your real traffic

    Classic A/B testing where volume allows, or qualitative and proxy methods where it does not, with an explicit threshold for moving from one to the other.

  • Reporting tied to results, not to activity

    Cost per qualified lead, qualification rate confirmed in the CRM, and the status of every hypothesis tested, the disproved ones included.

03 · How we work

  1. 01

    Measurement audit

    We check what exists, what is missing and whether the reported figures match commercial reality.

  2. 02

    Fixing tracking and consent

    GA4, GTM and Consent Mode v2, configured correctly rather than merely present.

  3. 03

    Diagnosis or test plan

    Qualitative below the statistical threshold, quantitative above it, depending on the real traffic of the page being tested.

  4. 04

    Running and reporting

    Cycles of 2 to 4 weeks, each with a hypothesis and decision threshold set in advance, then monthly recalibration.

04 · Results

  • 500k → 650klei per month, FineBar

    With the commercial markup raised from 15 to 20 per cent, towards 20 to 25.

These are the figures reported by the clients themselves, for their own accounts, over the periods shown. We do not turn them into derived percentages and we do not present them as reproducible.

05 · What people usually ask us

“We do not have enough traffic for A/B testing.”
Correct, and that is exactly why we do not sell it to everyone. We build a correct measurement plan and a qualitative programme, with an explicit threshold for moving to statistical testing as traffic grows.
“We already have Google Analytics installed.”
Having GA4 does not mean the data is complete or correct. Consent, wrongly configured events and broken attribution can make the figures differ seriously from commercial reality.
“We tried before and saw nothing.”
Most of the time that is a symptom of testing below the statistical threshold or of stopping early, not proof that the method does not work.

06 · Frequently asked questions

How much traffic do I need for A/B testing?
It depends on your baseline conversion rate, but a practical marker is at least a few thousand visitors and a few hundred conversions a week on the page being tested. Below that, we use qualitative methods.
What do you do if we do not have enough traffic?
Heatmaps, session recordings, proxy metrics and controlled before and after comparisons, methods that identify real friction without a statistical sample.
We already have Google Analytics, is an audit still worth it?
Almost always. We frequently find wrongly configured events or data distorted by the consent banner, invisible without a dedicated check.
What is server-side tracking and do we need it?
Infrastructure that is useful at high traffic and large media budgets. At small volumes the cost outweighs the benefit, and we say so plainly.
What happens if a test does not confirm the hypothesis?
We report it as such. A properly tested negative saves budget on directions that do not work, it is not a failure to hide.

Shall we talk about your project?

Tell us where you stand. If we are not the right answer for you, you will hear it in the first conversation.

Request a quote
← All Marketing & ADS services