Analytics and tracking infrastructure rebuild
Leadership had stopped trusting the numbers, and the cause sat underneath the campaigns: years of undocumented tag debt.
What they asked for
The numbers in the reports look off. Check the dashboards.
What was underneath
Years of undocumented tag debt sat underneath every report. Leadership hadn’t stopped needing the data; they had stopped trusting it.
The starting point
Performance reports were going out with a caveat that the numbers might not be accurate. The data had been unreliable long enough that leadership had stopped making decisions with it.
GA4 was partially configured, and GTM held years of tags added without documentation, naming conventions, or governance. Attribution across web, app, and lifecycle was inconsistent, and nobody had a complete picture of what was firing, what was double-counting, and what wasn’t firing at all.
The martech stack had grown tool by tool without a central owner. Braze, Intercom, and Microsoft Dynamics were running as silos: Braze wasn’t connected to Dynamics at all, so behavioural segmentation was blind to customer status, loan history, and account events.
Two problems compounding each other, then: a broken tracking layer making reporting unreliable, and a disconnected stack holding lifecycle strategy back from what it could do.
The decision
Use the Webflow ecosystem rebuild as the forcing function to fix the tracking layer at the same time. That was what made the work possible without a separate project cycle: every page launching in Webflow could launch with clean tagging from day one instead of a retrofit afterward.
Run the martech audit in parallel: map every tool, integration, and data flow, and surface what was missing, what was underused, and what was costing money without delivering value.
How I measured it
Data accuracy rate. The question leadership needed answered was whether they could trust what the reports showed, and accuracy at the event level was the only thing that answered it.
Secondary: cost savings from the martech audit. Unused tools and redundant subscriptions surfacing from the diagnostic work were a direct business outcome.
The build
Created a new GA4 property from scratch with a full event taxonomy supporting lifecycle, acquisition, and retention measurement simultaneously: standardized naming, custom dimensions, and conversion event configuration.
Rebuilt GTM completely in parallel with the Webflow migration, clearing years of tag debt and consolidating pixel management for Meta, LinkedIn, and other channels under GTM governance. Brought Analytics, DevOps, Product, and Legal into the process during the migration rather than after. Integrated Branch Metrics for mobile attribution, which had been essentially unmeasured.
Ran the martech audit across the full stack, mapping data flows between Braze, Intercom, Dynamics, and supporting tools. Surfaced the Braze–Dynamics integration gap, escalated it to the COO, and got it onto the product roadmap. Identified about $18K in annual savings from unused features and redundant subscriptions, and delivered a tech roadmap for future scaling.
40% → 80%
Data accuracy rate
~$18K
Annual martech savings identified
The rebuild didn’t improve campaign performance directly. It made campaign performance legible. When leadership can trust the data they make decisions with it, and when they can’t they stop looking at it entirely. The outcome that mattered wasn’t better numbers. It was numbers that could be used.
Tech
GA4
Analytics property rebuilt from scratch
Google Tag Manager
Full implementation and governance
Branch Metrics
Mobile attribution and deep-link tracking
Microsoft Dynamics CRM
Martech audit and integration mapping
Braze
Lifecycle platform and integration assessment
Looker
Measurement and reporting
Previous
DTC Newsletter web and SEO rebuild
Next
Building the MSOps payments division
Tell me what you’re working on.
If you’re working through a customer journey, system, or operational problem and think my experience might help, I’d like to hear about it.
Consulting on lifecycle, conversion, and automation—and open to the right senior role.