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 effort | Low (plug-and-play) | Very low | Medium (2–4 weeks setup) |
| Model flexibility | Tool defaults, limited | Every platform its own model | Last-click / data-driven / MMM combined freely |
| Incrementality | Available (e.g. Tracify) | Not possible | Geo holdouts + channel-pause methodology |
| Data ownership | Tool database | Platform silos | Your BigQuery, your Looker, your dbt |
| Best phase | DTC with standardized channels | Early phase, small budgets | Growing 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.
Audit
Channel discrepancies, model status, data maturity. Output: triage report.
Pipeline
BigQuery setup, GA4 export, ad-API connections, dbt transformations. Output: data warehouse.
Dashboard
Looker Studio with channel mix, cohort, LTV. Output: live dashboard.
Incrementality
First geo holdouts, pause tests, initial MMM evaluation. Output: reliable impact per channel.
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.
5–7 days · report + 30-min call
Starter Audit / Tracking
- – Channel-discrepancy analysis
- – Model-status check
- – Reporting gaps uncovered
- – Top 3 quick wins
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 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
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.
Cross-channel
Multichannel sales data
When every channel credits itself the conversion — how a consistent truth gets built.
Scaling
Ads won't scale anymore
When last-click reports send you in the wrong direction — what better attribution changes.
Tracking
Tracking despite consent loss
Server-side setup with hashed conversions as the data foundation for reliable attribution — GDPR-compliant.
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.
Let's talk
Three paths — depending on where you are.