DAVID MATHEWS
SELECTED PROJECT PORTFOLIO
Performance Marketing • AI & Automation • Platform Strategy
From campaign optimization to building systems, experiments and platform strategy.
01
Platform Strategy & Governance
Performance Max
THE STORY
Learn → Build → Scale → Strategize
The projects show a progression from performance execution to technical ownership and platform strategy. Four chapters, with the flagship business cases called out.
Paid Search
AI & Automation
Paid Search Projects
NL Packaging Parity
UK T-shirt Deep Dive
UK T-shirt Visibility
Polos Parity
CASE STUDY 1
NL Packaging Parity
Sep – Oct 2024 • Paid Search • Netherlands
BUSINESS PROBLEM
Packaging coverage was incomplete versus the US benchmark, leaving demand uncaptured.
WHAT I DID
Rebuilt the campaign structure end-to-end, identifying missing campaigns/ad groups/keywords and coordinated Dutch ad-copy translation.
IMPACT
13 → 20 campaigns; spend +73% MoM with internal ROAS (iROAS) held around 100–110%. First solo Vistaprint project.
TOOLS / SKILLS
Google Ads • Campaign Architecture • Keyword Research • Language Localization • Looker
CASE STUDY 2
UK T-shirt PSNB Deep Dive
Aug – Oct 2024 • Paid Search • UK
BUSINESS PROBLEM
VTP (Profitability metric) fell ~70% and orders dropped ~75% YoY while margin and competitiveness were under pressure.
WHAT I DID
Diagnosed the collapse and delivered nine optimization recommendations; analysis contributed to the roadmap for the half year that followed
IMPACT
VTP $3.6K / cost $4.4K during the analysis period; strategic recommendations adopted into roadmap.
TOOLS / SKILLS
Google Ads • Business analysis • Category diagnosis • Looker
CASE STUDY 3
UK T-shirt Visibility Recovery
Nov 2024 – Mar 2025 • Paid Search • UK
BUSINESS PROBLEM
VTP had collapsed to 6.6% of UK PPAG Category after tROAS was pushed too high due to market competition.
WHAT I DID
Co-led a five-month test removing IRS (Bidding Automation)/tROAS constraints, optimizing for conversion value and rebuilding ad copy.
IMPACT
~$28.6K test investment; visibility recovered to ~37%; tuned to ~1.0 iROAS by March. Documented for QBR.
TOOLS / SKILLS
Google Ads • Experimentation • Bidding strategy • Ad copy • Looker
CASE STUDY 4
Polos Parity Project
Dec 2024 • Paid Search • 7 EU markets
BUSINESS PROBLEM
Polos campaign structures were incomplete/outdated across secondary European markets.
WHAT I DID
Scaled a repeatable parity playbook across BE, CH, IT, SE, NO, AT and FI.
IMPACT
~$3.9K/month aggregate long-tail spend; 2 → 7 active markets and 6 → 10 campaigns.
TOOLS / SKILLS
Google Ads • Multi-market execution • Campaign architecture
Performance Max Projects
CA GeoX Media Mix Test
CA Search → 100% PMax Transition
Smartly Integration
BCN Onsite Knowledge Transfer
CASE STUDY 5
CA GeoX Media Mix Test
June 2025 – August 2025 • AI Media / PMax • Canada
BUSINESS PROBLEM
Canada needed evidence for the right balance between Search and PMax.
WHAT I DID
Executed a geo experiment across approximately $1M combined media to compare media-mix cells.
IMPACT
PMax-heavy cell demonstrated ~2× efficiency versus Search; informed the adopted 70/30 PMax/Text strategy and later consolidation.
TOOLS / SKILLS
Google Ads • Geo experiment • PMax • Search • Measurement • Conversion Lift Study
FLAGSHIP - CASE STUDY 6
Canada: Search → 100% Performance Max
The clearest example of strategy, experimentation, execution and financial impact converging.
$5.33M
FY26 PMax portfolio
−97%
Search spend: $183K → $7K/mo
~$2.1M/yr
identified wasted spend
~2×
PMax efficiency vs Search in GeoX
OPEN LOOP
How do you move a $5M+ market from Search to PMax without dropping sales?
That is the story the next 2 cases studies are designed to unpack.
FLAGSHIP - CASE STUDY 6
Canada: Search → 100% Performance Max
May 2026 – June 2026 • AI Media / PMax • Canada
BUSINESS PROBLEM
Search was not demonstrating incremental sales relative to PMax, creating an opportunity to reallocate spend.
WHAT I DID
Led the switch-over through three budget-cap waves and manual bid adjustments while absorbing Search spend into PMax.
IMPACT
Search spend −97% ($183K → $7K/month); ~$2.1M/year wasted-spend opportunity identified; negligible sales drop through transition.
TOOLS / SKILLS
Google Ads • PMax • Budget governance • Experimentation • Strategy
CASE STUDY 7
BCN Onsite Knowledge Transfer
October 2025 – November 2025 • Performance Max + Paid Search
BUSINESS PROBLEM
Automation only scales if other people can use and extend it.
WHAT I DID
Created and delivered a 35-slide enablement session covering alerting, self-serve alert creation and prompt engineering.
IMPACT
Enabled two teams to build their own alerts and extend the capability beyond the original builder.
TOOLS / SKILLS
Google Slides • Prompt engineering • Enablement • Knowledge transfer
CASE STUDY 8
Smartly Integration — Canada
Jan–Mar 2026 • AI Media / PMax • Canada
BUSINESS PROBLEM
Canada needed to migrate campaign/asset-group operations into a new automation platform without disrupting performance.
WHAT I DID
Adapted the campaign structure and launched 24 new asset groups through Omnifeed → Automation Feed → Smartly.
IMPACT
24 asset groups launched in March 2026 with zero performance disruption; part of a $5.33M CA PMax portfolio.
TOOLS / SKILLS
Smartly • Omnifeed • Automation Feed • Performance Max
AI & Automation Projects
Publisher Alerts Foundation
Google Ads Scripts Suite
AI Media Alerting System
Gemini Search Theme Generator�AI Recommendation Engine�Placement & Distribution Intelligence�
CASE STUDY 9
Publisher Alerts Foundation
2025 • Global AI Media foundation
BUSINESS PROBLEM
Manual monitoring was becoming a constraint, but the technical solution did not yet exist.
WHAT I DID
Volunteered for the technical unknown; self-taught JavaScript (vibe coding), prompt engineering and Google Ads Scripts from zero and built the foundation for later alerting.
IMPACT
Established the technical base for the automation suite that followed.
TOOLS / SKILLS
Vibe Coding • Google Ads Scripts • Problem solving • Project ownership • Prompt Engineering
CASE STUDY 10
Google Ads Scripts Suite
2025–ongoing • Global AI Media
BUSINESS PROBLEM
Manual pulls and analysis slowed recurring optimization work across markets.
WHAT I DID
Built scripts from scratch for daily alerts, search terms, brand traffic and trending categories.
IMPACT
Created on-demand analysis capability across US, CA and ANZ; foundation for later AI workflows.
TOOLS / SKILLS
JavaScript • Google Ads Scripts • Google Sheets • Automation
CASE STUDY 11
AI Media Alerting System
Built 2025 • recognized Dec 2025 • Global AI Media
BUSINESS PROBLEM
Teams needed a faster way to detect silent performance drops, budget anomalies and unexpected campaign changes.
WHAT I DID
Designed and built an AI-informed alerting system in Google Ads Scripts; then helped EU and ANZ teams adapt it to their markets.
IMPACT
Rolled out globally across EU, ANZ and North America. Recognized with the Think Big Spot Bonus for end-to-end ownership and global enablement.
TOOLS / SKILLS
AI • Google Ads Scripts • JavaScript • Monitoring logic • Global enablement
CASE STUDY 12
Gemini Search Theme Generator
~Mar 2026 • AI Media / PMax • Canada
BUSINESS PROBLEM
Creating search themes at scale was a repetitive bottleneck during the Smartly launch.
WHAT I DID
Built a Gemini-powered Google Sheets tool that generates 50 search themes per asset group.
IMPACT
Accelerated creation for the 24-asset-group Smartly launch and turned a repetitive task into a reusable team tool.
TOOLS / SKILLS
Gemini • Google Sheets • Prompt design • Automation
CASE STUDY 13
AI Recommendation Engine
Current • Platform Strategy • Global
BUSINESS PROBLEM
Given the large volumes of email alerts through scripts, decision making across all different scripts because time consuming
WHAT I AM WORKING ON
Building an all encompassing AI Agent and Automation workflow using N8N to connect the all the scripts created
EXPECTED IMPACT
A one stop Vistaprint Google Ads Recommendation Engine that will deliver an AI Agent to converse with and pull information and alerts on slack
TOOLS / SKILLS
AI • Google Ads • Prompt Engineering • N8N Automation
CASE STUDY 14
Placement & Distribution Intelligence
2025–26 • Global AI Media
BUSINESS PROBLEM
Category-level reporting & placement analysis for Performance Max was manual and not available at the required level.
WHAT I AM DOING
Volunteered to design an automated workflow that turns recurring category-level reports & placement analysis into an agent that gives actionable recommendations
IMPACT
Taking high level strategic AI-assisted decisions on Performance Max in terms of placements & distribution
TOOLS / SKILLS
N8N Automation • Reporting • Google Sheets • AI
NEXT GENERATION
From automation to AI decision support
The next layer of the work is not just collecting data — it is turning recurring signals into recommendations.
DAILY MARKET LOOP
Google Ads Scripts
→
n8n
→
LLM Agent
→
Internal Data
→
Recommendations
→
Shared Slack
MONTHLY ANALYSIS AGENTS
Brand traffic analysis • Search term analysis • PMax placement intelligence • Channel distribution analysis
The design principle: automate the monitoring, then use agents to surface the decision.
RECOGNITION
Think Big Spot Bonus — Dec 2025
Recognition for ownership of the AI Media Alerting System.
WHAT WAS RECOGNIZED
• Owned the alerting system from idea through implementation and continuous improvement.
• Used AI to design alerts and Google Ads Scripts to build the solution.
• Helped EU and ANZ teams adapt and deploy the system for their markets.
• Challenged manual monitoring processes and created a scalable technical solution.
• Continued identifying opportunities to automate category reporting and invoicing.
THINK BIG
Platform Strategy & Governance
ANZ Holiday Tactical Plan
Reddit Max / AU CLS
WorkMagic / US CLS
Bidding & Budget Governance�Placement Governance & Distribution Control�
Platform Strategy: moving upstream
The work is shifting from campaign optimization to platform governance, planning and experimentation.
PRIMARY PILLARS AS A PLATFORM STRATEGIST
Bidding governance • Placement Governance & Distribution Control • Tactical Planning • Incrementality Testing
CASE STUDY 15
ANZ Holiday Tactical Plan
Current • Platform Strategy • ANZ Google
BUSINESS PROBLEM
Holiday demand requires tactical allocation while Performance campaigns operate under a 100% ROAS optimization target.
WHAT I DID
Designed the Holiday Tactical Plan with explicit Demand Gen budget and separate Performance campaign governance.
IMPACT
$195K Demand Gen budget; Performance campaigns managed separately against the 100% ROAS target.
TOOLS / SKILLS
Google Ads • Tactical planning • Budget governance • ROAS
CASE STUDY 16
Reddit Max + AU CLS Roadmap
Current • Platform Strategy • Reddit AU
BUSINESS PROBLEM
Reddit Max is a new campaign format that needs to be understood and tested without losing the existing purchase-conversion framework.
WHAT I DID
Designing campaign integration, split-text testing, budget allocation and a combined CLS framework to establish incrementality.
IMPACT
Strategic roadmap in development; outcome to be validated through CLS before scaling.
TOOLS / SKILLS
Reddit Ads • Reddit Max • CLS • Experiment design
CASE STUDY 17
WorkMagic CLS — US Reddit
Current • Platform Strategy • Reddit US
BUSINESS PROBLEM
Need to validate Reddit as a proven partner before removing fixed-spend constraints for Reddit in the US
WHAT I AM DOING
Working with WorkMagic to conduct a CLS test; successful validation enables a move toward ROAS optimization.
IMPACT
Success criterion: prove incrementality and unlock flexible ROAS-based optimization.
TOOLS / SKILLS
Reddit Ads • WorkMagic • CLS • Partner strategy
CASE STUDY 18
Bidding & Budget Governance
Current • Platform Strategy • Governance
BUSINESS PROBLEM
As a part of my new role in need to manage the rules in internal automated bidding tool to ensure changes being made are in sync with performance
WHAT I AM DOING
Working with each Global Campaigns and markets to come up with bidding rules to ensure changes are smooth with least human intervention
EXPECTED IMPACT
Automated Bidding Changes and Budget Management based on internally specified rules & targets
TOOLS / SKILLS
Google Ads • Bidding & Budget Governance
CASE STUDY 19
Placement Governance & Distribution Control
Current • Platform Strategy • Governance
BUSINESS PROBLEM
No analysis on how Pmax can be improved in terms of placements & constant and continuous checking on Channel Distribution
WHAT I AM DOING
Creating an AI Agent that pulls information from the scripts created around placements & channel distribution to integrate the information with internal performance
IMPACT
AI Agent that recommends placement exclusions for Pmax & integrates channel level analysis into performance shifts and reporting
TOOLS / SKILLS
Agentic AI • Google Ads Scripts • Performance Reporting Enhancement
THE THREAD
I don't just optimize the campaign.
I learn the system → find the constraint → build the solution → prove it → scale it.
Performance Marketing • Experimentation • Automation • AI • Platform Strategy