Every marketer's LinkedIn now says "AI-powered." Almost none of them will show you what that means on a Tuesday morning. This page does. It describes the actual operating system behind my current role: five brands, one lean team, and a weekly rhythm where AI does specific jobs, and human judgment does the rest.
The setup
North 49 Brands and the Rymar Group run five distinct businesses: Spartan Fitness Equipment (commercial and home fitness, carrying lines like Life Fitness, Precor and TRX), Vintage Iron Cycles and Synergy Bikes (e-bikes and electric cycles), VO3 Fitness, and Rymar. Different customers, different price points, different seasons, one marketing function, which I lead.
That means strategy, budget, campaigns, content, e-commerce, events and reporting for all five, delivered by a small team I direct. I act as a team multiplier: by designing these AI-assisted workflows, I free up our designers and specialists from repetitive, manual execution so they can spend their energy on creative campaigns and strategic positioning. The system eliminates wasted coordination overhead, allowing a lean team to deliver enterprise-scale results.
The weekly rhythm
Monday is the numbers. GA4, Meta Ads, Google Ads and Shopify data for each brand, reviewed against the month's targets. I decide what gets more budget, what gets paused, and what needs creative refreshed. This is a judgment exercise; AI doesn't make these calls.
Tuesday and Wednesday are production. Campaign builds, email sends, landing pages, product launches. This is where AI earns its place, detailed below.
Thursday is the brands' own worlds. In-store promotions, corporate events, product launches, supplier co-op marketing. The physical side that no tool replaces.
Friday is the system itself. Reporting, next week's plan, and one improvement to the machine: a better template, a cleaner automation, a page that converts a little better than it did last week.
The Operating System: Context-Aware Local AI Database
I do not copy-paste generic prompts into web interfaces. Instead, my workflow is built on local shared folders connected to context-aware AI tools like Co-work, Antigravity, and Codex. This setup interfaces directly with a centralized local database containing two layers of business context:
- Strategic Core Files: Hand-built research documents detailing competitor analysis, visual brand voice & tone, customer personas, social media strategy, and business fundamentals (values, milestones, and brand representation guidelines).
- Dynamic Monthly Data: Regularly updated store logs, online sales analytics, website codebase repositories, product spec sheets, and inventory lists.
How the tools orchestrate campaigns
Because the AI operates with this full local database context, I can execute highly calculated, data-driven campaigns for any of the five brands without the risk of generic outputs or brand confusion:
- Data-Driven Campaign Building: I can query the AI system to identify the best-selling or best-priced products for a specific target persona (e.g., "Analyze Spartan Fitness inventory logs and suggest a campaign targeting boutique studio owners using our top-rated commercial treadmills"). The system instantly cross-references sales data with our persona guides to draft a calculated campaign bundle.
- Strict Voice Guardrails & Banned Words: To prevent brand cross-contamination, each brand directory contains its own visual guidelines and a custom "banned words and rules list." Vintage Iron's rugged, lifestyle e-bike copy is programmatically kept distinct from Spartan's high-performance commercial tone.
- Performance Integration: The system synthesizes actual GA4 website traffic logs, Google Analytics, and Klaviyo email metrics, ensuring that our next ad angle is backed by real performance curves, not creative guesswork.
- Technical SEO & Coding Efficiency: In web engineering, using Antigravity and Codex directly inside local codebase folders means the AI understands our custom theme architecture. It generates landing page copy, metadata structures, and custom scripts with full file context—minimizing reworks, cutting down iterations, and allowing us to ship exact changes immediately.
Where it doesn't
Just as important is what AI never does here:
- It doesn't decide. Budget, positioning, pricing support, what gets published, those calls stay human, informed by the data.
- It doesn't replace taste. A Harley-Davidson background teaches you that brand is discipline. Every output is edited against each brand's voice before it goes anywhere near a customer.
- It doesn't touch the numbers unsupervised. Analysis starts from real GA4 and platform data that I pull and verify; AI helps interpret, never invent.
- It isn't a volume machine. The goal was never "more content." It operates as a talent multiplier—handling the structural templates and data compilation so that our team can focus on taste, positioning, and direct customer interactions.
The other half: building it yourself
AI-assisted thinking pairs with hands-on execution. I write and ship the HTML, CSS and JavaScript changes on the Shopify and WordPress stores myself, landing pages, UX fixes, tracking, CRO experiments, without waiting on a developer queue. When the person who sets the strategy can also change the website by lunchtime, the whole portfolio moves faster.
What this means if you're hiring
The tools will change again next year. What you're actually hiring is the operating system: a senior marketer who can own a multi-brand function, lead a lean team, put AI to work where it belongs, keep judgment where it belongs, and show the receipts, 3.5x ROAS, 60% organic growth, 20% retention lift, Dealership of the Year, from every role that came before this one.