Attribution Tool-agnosticB2B SaaSE-learning + coaching

Marketing Attribution Consultant

Your ads might be profitable. You cannot prove it, so the budget conversation is a fight every quarter. I rebuild tracking and attribution so every sale traces back to a channel and a campaign, including the ones that close on a call, and I send that data back to the ad platforms so they optimise for buyers.

Book a free 30 minute audit See how attribution unlocked €1M ARR

Ayoub Kaddouri

Full stack growth consultant · 500+ webinars · €40M+ in webinar sales · LinkedIn

Updated Oct 2026

Attribution is a data pipeline first and a model second. Most teams buy the model and skip the pipeline.

Short answer

A marketing attribution consultant builds the pipeline and the model that tells you which channels drive revenue. What I do:

Tracking architecture: UTM rules, server-side events, CRM fields, identity stitching across web, forms, calls and sales
An attribution model that fits your sales cycle, not GA4's default
Offline and sales-assisted conversions sent back to Meta, Google and LinkedIn
Reporting from spend to closed revenue that finance signs off on

I am Ayoub Kaddouri, a full stack growth consultant. I work across SEO, technical marketing, paid ads, reporting and martech instead of one channel, which is why clients hire me when the problem sits between departments. Ten years running paid ads, eight on email, six on SEO and six on WhatsApp automation. Attribution is the layer I get called for most by SaaS and education companies. For Alphorm I rebuilt attribution from scratch: ads that were profitable but unprovable became a scalable €1M ARR channel mix. I have written the reference guides on attribution tools, GA4 models and custom models in SQL, and I am not affiliated with any tool vendor.

€1M

ARR unlocked at Alphorm

Tool affiliations

10 yrs

Paid media measured

01 Why attribution is broken at most companies

Last click by default. GA4 and the CRM both credit the final touch. Branded search and direct win, demand creation looks useless and gets cut.
Consent and iOS gaps. 40 to 60 percent of EU traffic is invisible to client-side tracking. Models run on half the data and call it the total.
The sale happens offline. Webinars, calls, proposals, bank transfers. None of it reaches the ad platform, so the algorithm optimises for form fills.
Three sources of truth. Ad platforms, GA4 and the CRM each report a different number. Nobody trusts any of them, so decisions default to gut.

The honest position

No model is correct. The goal is a model that is consistent, understood by finance, and good enough to make the next budget decision better than the last one.

02 What I build

AuditEvery tracking point, every UTM, every CRM field, every report. Where identity breaks between web, forms, calls and closed deals.
Tracking architectureUTM taxonomy enforced, server-side tagging, Conversions API for Meta, Enhanced Conversions for Google, LinkedIn CAPI. Lead source and campaign captured as CRM properties on the contact and the deal.
Identity stitchingAnonymous visitor to known lead to opportunity to customer, across devices where possible. This is the step most tools skip.
Attribution modelRule-based or custom, chosen for your cycle length and volume. Built in your warehouse or BI tool, not locked in a vendor.
Feedback loopClosed revenue, deal stages and lead quality sent back to the ad platforms as offline conversions. Within weeks the algorithms optimise for buyers.
ReportingOne dashboard from spend to pipeline to revenue by channel and campaign, with the model stated. Finance reviews it with me before it goes live.

03 Choosing the model

Your situation	Model I usually recommend	Why
Short cycle, self-serve, under 30 days	Last non-direct click or linear	Few touches, the last one matters most
B2B SaaS, 30 to 90 day cycle, sales-assisted	Position-based (40/20/40) or W-shaped	Demand creation and the demo request both deserve credit
Coaching and e-learning with webinars	Custom: webinar attendance as a weighted milestone	The webinar is the conversion event, standard models ignore it
3,000+ conversions a month	Data-driven, validated against a rule-based model	Enough volume for the algorithm to mean something

I keep two models side by side in every report. When they agree on channel rankings, trust the ranking. When they disagree, that is where the budget decision needs a human.

04 Tools: when you need one and when you do not

I have compared HockeyStack, Triple Whale, Northbeam, Hyros, Dreamdata and GA4 in detail in the attribution tools guide. The short version: under roughly €30K a month in ad spend, you do not need a paid tool. You need clean tracking, a CRM with source fields, and a BI dashboard. Above that, a tool saves engineering time, but it does not fix a broken pipeline. I recommend tools based on your stack, and I am not paid by any of them.

05 Who this is for

Good fit

B2B SaaS spending €10K+ a month on ads with sales-assisted deals
E-learning and coaching companies where sales close on webinars and calls
Companies with GA4, the CRM and ad platforms all reporting different numbers
Teams about to buy an attribution tool and unsure if they need one

Not a fit

E-commerce under €10K a month in ad spend (platform reporting is enough)
Companies without a CRM
Looking for a GA4 setup only (I can do it, but it will not fix attribution)

Want to know which channels actually pay?

30 minutes. Bring your ad spend by channel and your CRM. I will show you where the identity breaks and what the first fix is.

Book a free 30 minute audit See how attribution unlocked €1M ARR

06 Frequently asked questions

What does a marketing attribution consultant do?

They build the data pipeline and the model that connect marketing spend to revenue: tracking architecture, UTM rules, server-side events, CRM fields, identity stitching, an attribution model that fits your sales cycle, and the feedback loop that sends closed revenue back to ad platforms. The output is reporting that finance accepts and that makes budget decisions better.

Do I need an attribution tool like HockeyStack or Triple Whale?

Under roughly €30K a month in ad spend, usually not. Clean tracking, a CRM with source fields and a BI dashboard cover it. Above that, a tool saves engineering time, but it will not fix a broken pipeline. I compare the tools without affiliation in my attribution tools guide and recommend based on your stack.

Which attribution model is best for B2B SaaS?

For a 30 to 90 day sales-assisted cycle, position-based (40 percent first touch, 40 percent last touch, 20 percent middle) or W-shaped is the practical default. Data-driven only makes sense above roughly 3,000 conversions a month. I always report two models side by side so disagreements are visible.

How do you attribute sales that happen on calls or offline?

By capturing the lead source and campaign on the CRM contact at the first touch, carrying it to the deal, and sending the closed deal back to Meta, Google and LinkedIn as an offline conversion with its value. This also lets the ad platforms optimise for buyers instead of form fills.

Why do GA4, my CRM and my ad platforms show different numbers?

Different attribution windows, different identity methods and different scopes. Ad platforms count any touch in their window, GA4 splits credit across channels, the CRM records what a rep typed. Discrepancies of 20 to 40 percent are normal. The fix is one agreed model and one report, not forcing three systems to match.

How long does an attribution project take?

The audit takes one to two weeks. Tracking architecture and CRM fields take two to four weeks. The model and dashboard another two. Expect six to eight weeks to a trusted report, and another month before the offline conversion feedback loop changes ad performance.