HOME/WORK

Evidence beats adjectives. Inspect the systems behind the claims.

Real systems, not a list of employers: the case studies, company chapters and sanitized artifacts behind the claims — problem, system built, and the operating output that remained.

System-level workNo badge theatreBuyer-readable evidence
proof / full loop
signal · l1

The offer, buyer journey and system map are set before anything gets built.

Every claim on this page maps to one of these layers — built or operated in a real system.

Evidence model

Every claim connects to a case, a chapter or an operating pattern.

This is the proof layer for the site. No invented metrics and no badge clouds — the useful proof is what kind of system was owned, repaired, built or operated, and where you can read the detail.

01Software and founder delivery

Aroko, custom CRM/email product, technical teams, client delivery, contracts and support.

02CRM, lifecycle and automation

HubSpot workflows, source fields, lifecycle stages, handoffs, reporting and governance.

03Tracking and attribution repair

GTM, GA4, events, UTMs, CRM handoff checks, BI joins and platform conversion quality.

04Commerce, ERP and warehouse systems

Store architecture, product feeds, order flow, ERP handoff, stock and fulfillment visibility.

05Standards and barcode infrastructure

GS1-related web platforms, catalog discipline, barcode/standards context and operational data quality.

Case studies

Problem → build → operating output.

Sanitized by design: systems and outputs described honestly, without confidential metrics or invented numbers. Filter by type, or click any card to expand the full structure.

B2B SaaS · Product launch

Paid growth architecture for a platform launch & PLG shift

A German low-code SaaS moving from trial-based acquisition to freemium — needing a paid system that could prove itself past the signup.

Problem

New product, new motion: multiple ICPs, multiple channels, and a freemium funnel where the conversion that matters happens weeks after the click.

System

Full campaign architecture across ICPs and funnel stages — LinkedIn, Google Search & Demand Gen, Reddit, G2, Capterra — with UTM governance, stage-specific landing pages, and a layered measurement model connecting website intent, product signals, CRM quality and BI reporting.

Output

One documented paid system: every campaign traceable from impression to revenue-quality outcome, product signals feeding paid optimization, and reporting the leadership team runs weekly.

CLICK TO EXPAND +
B2B SaaS · Measurement

GTM/GA4 rebuild after a website migration

A site migration quietly broke the tracking foundation — conversions misfired, platforms optimized on noise.

Problem

Post-migration: events double-firing or missing, conversion goals pointing at dead URLs, ad platforms and analytics disagreeing badly enough that nobody trusted any number.

System

Full event taxonomy rebuilt in GTM/GA4: diagnostic events, verified conversion goals, platform alignment, SPA-safe triggers, and reconciliation checks between platforms and the CRM.

Output

A tracking layer the whole funnel could stand on — verified event-by-event, documented, and stable enough to train bidding on real outcomes again.

CLICK TO EXPAND +
B2B · Attribution

Fixing the CRM handoff where attribution died

Landing pages had UTMs; the CRM had "direct". The leak sat exactly at the click-to-contact handoff.

Problem

Contacts created through API and cross-subdomain paths arrived with no source data; first-touch fields were overwritten on later form fills. Marketing couldn't defend spend.

System

Source-preserving handoff: first-touch UTM properties written once per contact, custom marketing-touch events synced from product analytics, and a shared user key across CRM, product and warehouse.

Output

Pipeline finally carrying its origin: campaign-level attribution in the CRM, "unknown" collapsed to a residual, and channel decisions made on lineage instead of folklore.

CLICK TO EXPAND +
B2B software · Reporting

Central warehouse reporting for a marketing team

Tool-by-tool reporting meant hours of manual exports and numbers that never matched between meetings.

Problem

CRM, ads, analytics and web data lived in silos; monthly reporting was a copy-paste ritual and every stakeholder kept a private spreadsheet version of the truth.

System

Data pipelines and automated syncing between the marketing tools, a central warehouse as the single reporting source, and Looker Studio dashboards built on defined, documented metrics.

Output

Standing dashboards replaced the export ritual: one definition per metric, one place to look, and marketing KPIs reported monthly without archaeology.

CLICK TO EXPAND +
E-commerce · Fashion

Store, feed and campaign system for a fashion brand

One connected commerce system: the platform, the product feeds, the paid engine and the analysis loop.

Problem

A growing fashion brand with the classic split-vendor mess: a store, an agency, a feed problem — and nobody owning the whole chain from product data to purchase.

System

Built the e-commerce site end to end (payments, shipping gateways, variation logic, live chat), Merchant Center feeds with optimization rules, and full-funnel Google + Facebook campaigns with behavioral analysis via analytics, heatmaps and session recordings.

Output

One owner, one system: product changes flowed to feeds, feeds to campaigns, campaigns to measured purchases — with merchandising and sales actions executed inside the same loop.

CLICK TO EXPAND +
Tavazo Europe · Full system launch

Software, e-commerce, ERP, warehouse automation and launch marketing

A broader commercial build: not only a website or ads, but the operating system around a European commerce launch.

Problem

A launch needed more than a storefront: software and e-commerce architecture, design and development, operational handoffs, ERP setup, warehouse process automation, advertising and reporting had to work together.

System

Architected and built the commercial layer across software, e-commerce UX, operational workflows, ERP installation, warehouse automation logic, launch advertising and measurement.

Output

A connected launch system: customer-facing commerce, internal operations, inventory/warehouse process, marketing execution and reporting designed as one system rather than separate vendor tasks.

CLICK TO EXPAND +
DACH SMBs · Lead gen

Multi-client lead-generation systems for the DACH market

Ads, tracking, CRM automation and the websites themselves — built and proven across SMB clients.

Problem

SMBs needing pipelines, not campaigns: each with different offers, budgets and sales processes, none with in-house marketing operations.

System

Per-client lead-gen stacks: Google Ads with GTM conversion tracking, HubSpot automation (sequences, segmentation, lifecycle), omnichannel LinkedIn outreach with conditional logic, plus conversion-focused WordPress builds.

Output

Repeatable systems handed to clients as documented operating assets: trackable spend, automated follow-up, and websites that carried their own conversion weight.

CLICK TO EXPAND +

Confidentiality note: client-specific revenue figures and internal metrics are deliberately excluded. References and deeper detail available in conversation.

Artifact layer

Some proof should be visual, because the work itself is visual and operational.

Sanitized references based on real working files: command centers, GTM files, launch wireframes and search/reporting tools. They prove the type of system thinking behind the work, without exposing private customer data.

How to evaluate it

Every project should leave behind an operating asset.

A useful engagement does not end as “we ran campaigns” or “we made a dashboard.” It leaves behind a map, definitions, handoffs, implementation logic and a decision rhythm.

01

Map

What exists, what breaks, which owner touches the next step.

02

Build

Pages, flows, automations, events, dashboards or operating documents.

03

Govern

Definitions, source logic, QA checks and ownership rules.

04

Operate

Weekly decision loops and prioritized improvements.

Your situation probably rhymes with one of these. Let's find out which.

Use the service recommender

Evidence first, scope second.

Run the free audit against your own stack, open a company chapter, or send the system problem directly. The right route depends on what is already built and what can be trusted.

Free Growth Audit → x