The work that speaks for itself
Real projects, real outcomes. Every engagement below started with a measurement plan and ended with clean, documented tracking. Details have been generalised to respect client confidentiality, but the results are genuine.
GA4 Rescue for a Growing Marketing Agency
A UK-based marketing agency with 12 clients on retainer had inherited GA4 setups that were producing inconsistent numbers. Client reports were being questioned. Trust was eroding.
Conflicting data across clients
Each client had GA4 configured differently. Some had no conversion events, others had duplicate page views inflating session counts, and a few were still running UA alongside GA4 and double-counting every interaction. The agency tried fixing this internally, but their team lacked the technical depth to standardise the tags. The primary constraint was time. The account managers were spending days each month manually cleaning data before presenting it. The agency's reporting was built on unreliable data and clients were starting to notice discrepancies. The agency's clients were seeing different conversion numbers in GA4 compared to their CRM systems. The data inconsistencies were eroding trust between the agency and its clients. When a client asked why a specific campaign was underperforming, the agency could not confidently answer because the underlying data was flawed.
Standardised measurement framework
We designed a single measurement plan template covering the events every client needed: form submissions, phone clicks, email clicks, scroll depth, and key page views. Then we audited and rebuilt each client's GA4 and GTM containers against that template. We consolidated tagging into a single GTM container per client, wrote custom Javascript variables to handle dynamic forms, and built a Looker Studio template for consistent reporting. We stripped out the legacy Universal Analytics code that was causing the double-counting, ensuring a clean slate. For lead generation clients, we standardized form submission events, tracking both successful submits and field-level validation errors. We migrated all custom dimensions to a standard naming convention. For the technical implementation, we built a master GTM template that could be easily adapted for each client's specific CMS. We removed all legacy code that was firing duplicates.
Zero client questions about data accuracy
Within three weeks, all 12 clients were on identical measurement frameworks. Reported conversion figures finally matched client-side CRM data. The agency's monthly reporting time dropped from 12 hours to 4 hours per account. The technical debt was cleared, meaning new clients could be onboarded with the standard tracking in a fraction of the time. Most importantly, there were zero client queries about data accuracy in the six months following the rebuild. The new setup allowed the agency to onboard new clients with a standard tracking framework in a fraction of the time. The 4 hours of monthly reporting was fully automated via Looker Studio. The data was finally trusted by both the agency and the clients. The agency could focus on strategy rather than defending their metrics.
Consent Mode v2 for a Regulated E-Commerce Business
An online retailer in a regulated sector needed full GDPR compliance without losing the marketing data their ad campaigns depended on. Their existing consent banner was blocking everything, and Google Ads performance had collapsed.
Compliance was killing marketing data
The existing cookie consent solution was an all-or-nothing banner: accept everything or nothing. When users declined, GA4 and Google Ads received zero data. The business was flying blind on campaign performance and could not justify ad spend. They had tried alternative consent platforms, but the fundamental issue remained: the strict setup broke attribution. The constraint was legal. They needed to remain strictly compliant with GDPR and could not risk a fine, but they also needed data to run their marketing effectively. The fundamental issue was balancing aggressive marketing targets with strict regulatory requirements. The business had previously ignored consent, and when legal forced a strict banner, the marketing team lost visibility. They tried different banner designs to increase opt-in rates, but the gains were minimal. They needed a way to measure campaign effectiveness without violating user trust or risking regulatory action. The ad spend was high, and flying blind meant they could not optimize their bidding strategies.
Granular consent with modelled data recovery
We implemented Consent Mode v2 with a self-hosted CMP. We configured four consent categories with granular gtag signals, and set up server-side GTM to move tracking off the browser. We tested the setup extensively to ensure no tags fired before user interaction. Even when users declined marketing cookies, Consent Mode sent cookieless pings that Google could use for behavioural modelling. This approach respected user privacy while maintaining data flow. We started by overhauling their Google Tag Manager setup, moving from a standard client-side implementation to a server-side container hosted on Google Cloud. We integrated a new self-hosted CMP that accurately categorized cookies. Then, we configured Consent Mode v2, mapping the consent states to Google's specific requirements. We ensured that when a user denied tracking, the browser only sent cookieless, anonymized pings to the server. The server then handled the data routing, ensuring no personal data ever reached third-party servers without explicit consent.
Compliant, measurable, and defensible
The business achieved full GDPR compliance with auditable consent records. Google Ads conversion reporting recovered approximately 65 percent of the conversion data that was previously lost, all through behavioural modelling. The self-hosted CMP eliminated the recurring third-party consent provider fee. Furthermore, the server-side GTM setup improved page load speed by reducing client-side tag bloat, making the site faster for users. The business found the sweet spot between compliance and performance. The modelled data allowed Google Ads algorithms to resume optimizing bids, preventing a total collapse of their return on ad spend. The legal team was satisfied with the auditable consent logs. The self-hosted CMP meant they were not paying monthly fees based on traffic volume. The server-side setup also gave them complete control over their data pipeline, opening the door for more advanced analytics in the future.
Server-Side Tracking for an Adblock-Heavy Audience
A B2B SaaS company discovered that 42% of their target audience used adblockers, and their client-side GA4 was missing nearly half their traffic. Their subscription funnel data was incomplete. Marketing couldn't measure campaign ROI.
42% of traffic was invisible to analytics
The company's tech-savvy audience (developers and IT decision-makers) had high adblocker adoption. About 42 percent of their traffic was invisible to analytics. Client-side GA4 and Google Ads tags were being blocked outright. The marketing team was making budget decisions based on data that represented barely half of actual visitors. Trial signups were not being attributed to the right campaigns. The constraint was technical: no amount of client-side tweaking could bypass the filter lists these users had installed. The developers actively block third-party trackers, meaning standard analytics scripts were failing silently. The marketing team was completely unaware of the scale of the problem until they noticed a massive discrepancy between GA4 traffic and their backend server logs. They tried incentivizing users to disable adblockers, but this failed completely. They were forced to make budget decisions based on an incomplete picture, meaning they might have been cutting campaigns that were actually driving valuable trial signups.
First-party endpoint bypassing adblockers
We deployed a server-side GTM container on the company's own subdomain. We configured a custom first-party endpoint for GA4 data collection, and routed Google Ads conversion tracking through the same server container. From the browser's perspective, all tracking data went to the company's own domain, making it invisible to adblocker filter lists. We also set up server-side validation to strip PII before forwarding any data to Google, ensuring privacy remained intact. We built a server-side tracking architecture designed specifically for resilience. We set up a first-party tracking endpoint on a subdomain. We rewrote their tracking scripts to send data directly to this endpoint rather than directly to Google's servers. From the browser's perspective, this looked like a standard API request to their own infrastructure, bypassing the filter lists entirely. Once the data reached the server-side GTM container, we implemented strict data governance rules. We scrubbed IP addresses and user agents before forwarding the anonymized event data to GA4 and Google Ads.
Full visibility across the funnel
Post-migration, GA4 session counts increased by 38 percent because the previously invisible adblocked traffic was now being measured. Campaign attribution became reliable enough to cut underperforming ad channels, saving approximately 2400 GBP per month in wasted spend. The first-party endpoint also improved cookie duration, giving more accurate returning-visitor metrics across longer time windows. The marketing team finally had an accurate picture of their funnel. The 38 percent increase in measured sessions completely changed their understanding of user acquisition costs. They identified that their highest-converting technical audience was the one most likely to use adblockers, meaning they had previously undervalued their most effective campaigns. The 2400 GBP saved per month was reinvested into the newly identified high-performing channels. The server-side infrastructure also proved more reliable during traffic spikes, ensuring no data was lost.
How we run an engagement
Every project is treated as an engineering task, not just a marketing add-on. We follow a strict process to ensure data is accurate, compliant, and useful.
Audit the existing setup
We review your current GA4 property, GTM container, and data layer. We document exactly what is broken and what is working. We sit with your team to understand the business decisions the data needs to support.
Write the measurement plan
Before writing any code, we document the event taxonomy. We map out the exact parameters for every interaction, ensuring the naming convention is consistent. Your developers get a clear specification to follow.
Implement and validate
We build the container architecture, set up the tracking endpoints, and configure the properties. Then we run extensive QA tests. We verify the data in preview mode, in production, and inside the analytics platforms themselves.
Ready to be our next case study?
Every engagement starts the same way, a free, honest conversation about your tracking. No hard sell. No jargon. Just clear thinking about what your data should be doing for you.
Let's talk - free consultation