Meta Account Architecture
Structure wins on Meta too. Advantage+ campaigns, the feeder strategy, margin-based segmentation, and the math behind learning phase success.
1. Structure Over Creative
Great creative in a bad structure underperforms. True on Google. Just as true on Meta.
How you build the account determines whether the algorithm can actually learn.
The Meta Structure Principle
Structure beats creative. Fix attribution, segment by margin, fund ad sets properly. You'll outperform brands with better creative but worse structure.
The Four Structural Pillars
Attribution Settings
Correct conversion window (7-day click, 1-day view is default). Match to your actual purchase cycle.
Margin-Based Segmentation
Segment campaigns by profit margin, not product category. High-margin products can tolerate lower ROAS.
Proper Funding
Each ad set needs enough budget to exit learning phase. Starved ad sets never optimize.
Signal Quality
CAPI + Pixel events. Send clear, clean conversion data. Garbage in, garbage out.
2. Learning Phase Economics
Meta's algorithm needs data to optimize. The learning phase is where it gets that data.
Rush it or starve it, and you never reach the performance the account is capable of.
| Metric | Threshold | What It Means |
|---|---|---|
| Conversions per ad set | 50/week | Below this, the ad set stays in learning |
| Time in learning | < 7 days | If it takes longer, budget or audience is too narrow |
| Changes during learning | 0 | Any edit resets the learning phase |
| Budget per ad set | 5-10x target CPA | Daily budget minimum to fuel learning |
The Fragmentation Trap (Meta Edition)
Same problem as Google, different symptoms. Too many ad sets = none exit learning phase. Got 8 ad sets each getting 6 conversions per week? Consolidate to 2-3 getting 20+ each. Same budget. Much better results.
3. Advantage+ Shopping Campaigns
Advantage+ Shopping (ASC) is Meta's broadest, most automated campaign type. It works when you understand its constraints.
What it does: Broad targeting with full automation. Serves across prospecting and retargeting. One campaign, all audiences.
When to use it: Accounts spending €5K+/month on Meta with a proven product and sufficient creative volume (10+ active creatives).
When NOT to use it: New accounts with limited data, products that need educational content before purchase, or when you need strict budget control between prospecting and retargeting.
The existing customer cap: Set the "Existing Customer Budget Cap" to 10-25% to prevent ASC from burning budget on retargeting people who'd buy anyway. Without this cap, your "prospecting" campaign is often 50%+ retargeting.
4. The Feeder Strategy
Meta's algorithm concentrates on best-sellers. Great for those products, terrible for catalog breadth.
The feeder strategy solves this without fighting the algorithm.
How It Works
Primary Campaign
All products, all creatives. Broad "purchase new customer" optimization. This is your main engine.
Feeder Campaigns
Separate small campaigns, each focused on a specific product or category. Also purchase-optimized. Relevant creatives only.
Intentional Overlap
Let audiences overlap between primary and feeder campaigns. The feeders "train" the algorithm to diversify its focus within the primary campaign.
Why It Works
Feeder campaigns inject targeted activity for underperforming products. They teach the primary campaign's algorithm to diversify, without splitting your budget into starved, isolated campaigns that can't learn.
Good candidate for feeders
Products with decent margins and proven demand (on-site or organic) that the algorithm ignores in the primary campaign.
Bad candidate for feeders
New, unproven products or low-margin items. Test product-market fit first. The feeder strategy amplifies, it doesn't validate.
5. Margin-Based Segmentation
Most advertisers segment Meta campaigns by product category. That's wrong.
Segment by margin profile instead.
| Tier | CM2 Margin | Target ROAS | Strategy |
|---|---|---|---|
| High Margin | > 50% | 1.5-2.5x | Scale aggressively. These products fund growth. |
| Medium Margin | 30-50% | 2.5-4x | Protect efficiency. Scale carefully. |
| Low Margin | < 30% | 4x+ | Only if AOV or LTV compensates. Otherwise, deprioritize. |
This connects directly to your Profit Waterfall. Your CM2 margin per product determines how aggressively you can bid on Meta. If you haven't built your waterfall yet, start there.
See the Profit Waterfall →Google vs. Meta Structure
Same principles. Different implementation.
| Principle | Meta | |
|---|---|---|
| Don't fragment | Consolidate ad groups (5K impr/mo min) | Consolidate ad sets (50 conv/week min) |
| Segment by economics | Separate campaigns by bid target | Separate campaigns by margin tier |
| Feed the algorithm | 30+ conv/mo for tCPA, 50+ for tROAS | 50+ conv/week per ad set |
| Match message to stage | Keyword intent → ad copy | Creative angle → audience matching |
| Measure patiently | Respect the conversion window | Don't touch learning phase for 7 days |
The CMO Lens
Account structure is invisible to most founders. They see the creative, the budget, the results. But a CMO knows that structure determines whether the algorithm can actually learn. We've seen accounts double performance just by consolidating from 12 ad sets to 3. No new creative. No budget increase. Just structure.
Want Your Meta Structure Audited?
We'll review your Meta account architecture, identify structural issues, and show you exactly what to consolidate or restructure. 30 minutes. No pitch.