Campaign Structure SOP: The Clean Architecture

Module: Scaling Millions on Meta Instructor: Kevin Gundersen Revenue Rush University

The Structure

Every account I've built — from Inno Supps to the accounts I advise through Revenue Rush — runs the same basic structure. It works at $1K/day and it works at $150K/day. The details change, the skeleton doesn't.

Campaign 1: Prospecting — Creative Testing

• Objective: Conversions (Purchase)
• Budget: 15-20% of total spend, ad-set-level budgets
• Audience: Broad (25-55, United States, no interest targeting)
• Ad sets: 1 per creative concept being tested
• Budget per ad set: $50-100/day
• Purpose: Find winners. Expect negative or break-even ROAS here.

Campaign 2: Prospecting — Scaling

• Objective: Conversions (Purchase)
• Budget: 65-70% of total spend, ad-set-level budgets
• Audience: Broad (same as testing)
• Ad sets: 2-4, each with proven creatives only
• Budget per ad set: Scale based on performance ($200-5,000+/day)
• Purpose: Drive revenue. This is where profitability lives.

Campaign 3: Retargeting

• Objective: Conversions (Purchase)
• Budget: 10-15% of total spend
• Ad sets:
• Website visitors (7-day window) — exclude purchasers
• Add-to-cart abandoners (14-day window) — exclude purchasers
• Engaged on Instagram/Facebook (30-day) — if applicable
• Budget per ad set: $20-50/day at your stage, scaling with traffic volume
• Purpose: Convert warm traffic cheaply

Optional Campaign 4: Retention / Existing Customers

• Objective: Conversions (Purchase)
• Budget: 5% of total spend
• Audience: Past purchasers, subscription lapsed
• Purpose: Reactivation, cross-sell, new product launches

Audience Strategy

I'm going to say something that contradicts what most people teach: at your spend level, go broad.

Interest targeting was valuable in 2019. In 2025-2026, Facebook's algorithm is better at finding your customer from a broad audience + good creative than you are at defining interests. Every interest you add is a constraint that limits the algorithm's ability to explore.

When to use interests: Only during your very first week of ads, when you have zero pixel data and need to give the algorithm a starting point. After 50-100 conversions, switch to broad.

Lookalike audiences: Valuable at $5K+/day when you have 500+ purchasers. Use 1% Lookalike of purchasers. Don't stack Lookalikes — Facebook will already expand beyond the seed.

Exclusions to always run: - Exclude purchasers (last 180 days) from prospecting campaigns - Exclude website visitors from prospecting if running retargeting - Exclude each ad set's audience from other ad sets (prevent overlap)

Ad Format Priorities

In order of typical performance for DTC supplements:

1. Single video (15-30 seconds) — UGC testimonials dominate
2. Single image with bold copy — Works for retargeting and price-sensitive buyers
3. Carousel — Good for showing product range or before/after sequences
4. Collection ad — Best for accounts with 10+ SKUs, feeds into a catalog experience
5. Instant Experience — Resource-intensive to create, test at scale

Start with video and static. Add carousel once you're past $2K/day. Don't touch Collection or Instant Experience until $10K/day.

Daily Management Checklist

Every morning (10 minutes max): - [ ] Check spend vs. budget (any ad set overspending or underspending?) - [ ] Check CPA on active ad sets (any above 1.3x your target?) - [ ] Check frequency on retargeting (above 3.0? Refresh creative) - [ ] Check any new ad sets launched yesterday (is data flowing?)

Every Monday (30 minutes): - [ ] Kill underperformers in testing campaign (CTR <1.5% after $150 spend) - [ ] Graduate winners to scaling campaign - [ ] Review weekly CPA trend (stable, improving, degrading?) - [ ] Plan next week's creative tests

Do NOT do: - Make changes more than once per day - Make changes after 6pm (let the algorithm run overnight) - Make changes on weekends unless something is catastrophically broken - Panic after one bad day — judge on 3-day rolling averages minimum