Skip to content

Services / Data & Automation / Attribution & Dashboards

Attribution that shows what actually works — not what's convenient.

Platform reports tell every channel its own truth. I build cross-channel attribution with incrementality tests and live dashboards — so the budget moves toward impact, not toward the loudest reporting.

Cross-channel view

One truth instead of five platform claims. Clear where the customer really comes from.

Incrementality tests

Measures whether a channel actually causes sales — not just credits itself conversions that would have happened anyway.

Live dashboard

Looker Studio with all channels, CAC, ROAS trend, cohorts. No monthly PDF, no slide battles.

What you can book

Six building blocks, individually or as a package

We usually start with the attribution audit. The model choice depends on your channel mix and data maturity.

Attribution audit

Channel-truth comparison, reporting gaps, model status (last-click / data-driven / MMM).

Model choice

Last-click for small setups, data-driven for mid-size, media-mix modeling from €1M budget. With a decision matrix.

Data pipeline

GA4 BigQuery export, shop data, ad-platform APIs in one warehouse. dbt transformations.

Dashboard build

Looker Studio or Metabase with channel mix, cohort analysis, LTV trend, CAC payback.

Incrementality tests

Geo holdouts, channel pauses with lift measurement. Methodology for overlapping audiences.

Team onboarding

Dashboard-reading workshop, decision framework, monthly review sync. Self-service maturity.

How do we differ?

Three ways to build attribution

Specialist tool, platform reports or your own pipeline — all three have their phase. Here's the honest comparison.

 Attribution tool
Tracify, Triple Whale, Cometly
Platform reports
GA4 + ad platforms directly
Truong Suarez
Own pipeline + setup hands-on
Setup effortLow (plug-and-play)Very lowMedium (2–4 weeks setup)
Model flexibilityTool defaults, limitedEvery platform its own modelLast-click / data-driven / MMM combined freely
IncrementalityAvailable (e.g. Tracify)Not possibleGeo holdouts + channel-pause methodology
Data ownershipTool databasePlatform silosYour BigQuery, your Looker, your dbt
Best phaseDTC with standardized channelsEarly phase, small budgetsGrowing setup, > 5 channels, in-house BI maturity

Comparison based on publicly available information, as of 2026. If a tool or platform reporting fits your constellation better, I'll tell you so in the intro call.

How we work

Five phases, one point of contact

Audit first, then pipeline. Model and incrementality come once the data flows.

01 · Week 1–2

Audit

Channel discrepancies, model status, data maturity. Output: triage report.

02 · Week 2–4

Pipeline

BigQuery setup, GA4 export, ad-API connections, dbt transformations. Output: data warehouse.

03 · Week 4–6

Dashboard

Looker Studio with channel mix, cohort, LTV. Output: live dashboard.

04 · Week 6–8

Incrementality

First geo holdouts, pause tests, initial MMM evaluation. Output: reliable impact per channel.

05 · quarterly

Refresh

Model adjustment, integrate new channels, update methodology.

Stack

What we work with

No black-box tools. Everything we use, you can run yourself — if you want to.

Warehouse

  • BigQuery (standard)
  • Snowflake (enterprise)
  • Postgres + dbt (smaller setups)

Ingestion

  • GA4 BigQuery Export
  • Fivetran / Airbyte
  • Custom APIs (Meta, TikTok, etc.)
  • Shopify / WooCommerce

Transformation

  • dbt
  • SQL
  • Python (notebooks)
  • MMM frameworks (Robyn, LightweightMMM)

Visualization

  • Looker Studio
  • Metabase
  • Mode Analytics
  • Streamlit (custom)

Recommended entry point

Two paths, depending on where you stand

For existing reports, the tracking audit reveals the discrepancies. For building a pipeline, the Growth Sprint is the better fit.

For you if

Your reports contradict each other

GA4 says X, Meta says Y, the shop says Z. You want to know which truth holds and how to set up consistent reporting.

€890fixed price

5–7 days · report + 30-min call

Starter Audit / Tracking

  • Channel-discrepancy analysis
  • Model-status check
  • Reporting gaps uncovered
  • Top 3 quick wins
Book the Starter Audit
Deeper plan

For you if

Pipeline + dashboard are new

You want your own attribution setup with BigQuery, dbt and Looker Studio. You need architecture, a pipeline plan and a model recommendation.

€2,490fixed price

2 weeks · report + workshop

Growth Sprint

  • Pipeline architecture (warehouse, ingestion)
  • Model recommendation (last-click / DDA / MMM)
  • Dashboard concept (charts, cohorts)
  • Incrementality test plan
  • Half-day workshop for the handover
Book the Growth Sprint

Not sure? The symptom triage on the audit page helps you choose. The audit price is credited toward a follow-up project.

When this becomes relevant

Typical starting points

Three recurring situations in which attribution & dashboards are the right tool.

FAQ

What clients often ask before the first engagement

Last-click, data-driven or MMM — which fits me?

Last-click with < €100k annual budget and 1–3 channels (fast, simple, enough). Data-driven attribution from €100k and 4–7 channels (GA4 or custom). Media-mix modeling from €1M or with a strong offline component (TV, OOH). I often combine DDA + light MMM.

Do I really need incrementality tests?

If your biggest channel has > 30% of the budget: yes. Geo holdouts or channel pauses often show that only 20–40% of the ROAS is incremental — the rest were conversions that would have happened anyway. For small setups (< €50k budget/month), direct-response tests rather than formal incrementality.

Isn't GA4 + platform reports enough?

For smaller setups: yes. GA4 + clean channel mapping is a valid baseline. As soon as 4+ channels run with mutual overlap audiences, the reports start excluding each other — then your own pipeline with source-of-truth logic pays off.

Tracify or your own pipeline?

Tracify for DTC e-commerce where setup time is short and AI attribution is wanted. Your own pipeline for mid-size companies with BI maturity, multi-brand complexity or specific industry KPIs (LTV, cohorts, subscription churn).

How much maintenance does the dashboard need?

Once the pipeline runs: 2–4 hours per month for data-anomaly checks, new campaign tagging, quarterly reviews. For structural changes (new platform, new audience logic) a module gets added. I offer maintenance as an optional retainer or hand it over to your data team.

Contact

Let's talk

Three paths — depending on where you are.