Advertising Mastery

Lineage: upgraded from IDS_Advertising_Mastery.md. Practitioner base — Meta/Facebook: Nick Shackelford (Structured Social), Cody Plofker (Jones Road Beauty), Ben Heath, Charlotte Henry, Barry Hott, Jon Loomer, Andrew Faris, Emanuel Cinca, Thomas Micaletti, Depesh Mandalia. Google/YouTube: Mike Rhodes (AgencySavvy), John Moran (Solutions 8), Brett Curry (OMG Commerce), Kasim Aslam, Isaac Rudansky, Neil Patel. TikTok: Harry Coleman (Beast of Ecom), Jordan Welch, Davie Fogarty (The Oodie), Jack Bloomfield, Elise Darma. Creative/UGC: Raul Galera, Ash Melwani, Rachel Jimenez, Kristen LaFrance. Attribution: Andrew Faris, Cody Plofker, Taylor Holiday (Common Thread Collective), Richard Gelis. Scaling: Raphael Paulin-Daigle, Ezra Firestone (Smart Marketer), Ryan Deiss, Chase Chappell, Scott Cunningham. Cross-platform: John Moran, Mike Rhodes, Aaron Young, Luke Carthy. Current as of July 2026.

---

#THE ONE-PAGE VERSION

1. Know your numbers before you touch a campaign. Breakeven ROAS = 1 ÷ contribution margin before ad spend (LUCE_06 §2.4's definition — not gross margin). Target ROAS = breakeven × 1.3–1.5. If you can't state these cold, you're not ready to scale.
2. MER > ROAS. NC-MER > blended MER. Platform-reported ROAS is directional at best; New Customer MER is the only number that tells you whether the brand is actually growing or just re-selling to the same list.
3. Andromeda/Advantage+ CPA advantage is tiered by spend: −38% at $10k+/month, only −14% under $2k/month. Above $10k/month you can lean on automation for delivery; below it, creative quality still does most of the work.
4. Meta's July 2026 removal of the off-platform activity opt-out (folded into "Activity from other businesses") expands retargeting and lookalike source pools — re-segment by recency rather than treating it as a free scale-up in quality.
5. TikTok's post-JV algorithm is retraining on US-only data — expect real week-to-week reach volatility through 2026. TikTok Shop's July 2026 rulebook replaced violation points with Account Health Rating (AHR), recalculates Store Rating continuously, and consolidated ad spend under GMV Max.
6. Google's consensus is PMax + Standard Shopping hybrid, not PMax alone — Standard gives you back the query-level visibility PMax hides. Demand Gen is the only conversion-optimized video type left since Video Action Campaigns died April 2025; YouTube Shorts is the cheapest reach on the platform.
7. Attribution is a triangulation exercise, not a single source of truth. Use platform-reported + MTA for daily decisions, MMM + incrementality for quarterly budget allocation, post-purchase survey for brand-awareness validation. No single number is "correct."
8. Creative velocity is the actual scaling constraint above $10k/month. More budget accelerates fatigue; only a systemized creative production pipeline (hook matrices, AI-assisted drafting, structured UGC briefs, variant batching) keeps pace.
9. The 4-phase scaling model still holds: Foundation ($0–5k/mo, learning) → Acceleration ($5k–30k/mo, MER 3–4×) → Scale ($30k–150k/mo, full channel portfolio) → Efficiency Optimization ($150k+/mo, MMM + incrementality quarterly).
10. Contribution-margin bidding beats ROAS bidding at scale. Bid on profit, not revenue — apply margin-weighted conversion value rules so the algorithm optimizes for what actually pays your bills.
11. Incrementality varies wildly by channel. TikTok tends to run 60–80% incremental (creates genuine new demand); Google Brand Search often runs only 20–40% incremental (captures demand that existed anyway). Budget allocation should reflect this, not just platform-reported ROAS.
12. Cheap-inventory channels (Pinterest, Threads, Reddit) earn a budget line only after your core two channels are stable — sequencing prevents diluting measurement and creative capacity too early.
13. Q4 is a scaling month, not a testing month. Build your creative bank in Q1–Q3; CPMs run 2–3× normal during BFCM week, and cold-testing new concepts into that spike is how good quarters go bad.
14. The agency-vs-in-house decision is a spend-threshold decision, not a philosophy: in-house becomes economical north of ~$50k/month in ad spend, assuming you can afford a dedicated media buyer.
15. This module is the deep-dive twin of LUCE_04. If you haven't run LUCE_04's testing-phase playbook and cleared consistent breakeven ROAS, come back to this module after you have — the frameworks here assume real data volume to work with.

---

#SECTION 1: THE ADVERTISING OPERATOR MINDSET

#1.1 The Three Laws of Elite Media Buying

The amateur asks: "What's my ROAS?" The intermediate asks: "What's my blended ROAS?" The expert asks: "What is my MER, my NC-MER, my contribution margin, and my payback period?"

The fundamental shift: you are not running ads. You are building a customer-acquisition machine with known inputs and predictable outputs.

Law 1 — Economics Before Tactics. Know your numbers cold before touching a campaign:

Breakeven ROAS = 1 ÷ Contribution Margin % before ad spend (LUCE_06 §2.4)
Example: 60% CM → breakeven ROAS = 1.67×
Target ROAS = Breakeven ROAS × 1.3–1.5 (cushion for overhead + reinvestment — canonical multiplier, LUCE_06 §7)
LTV-adjusted ROAS = Breakeven ROAS × (1 ÷ (1 − (LTV Multiplier − 1)))

Law 2 — Creative Is the Variable, Distribution Is the Constant. At scale, the algorithm distributes efficiently across whatever it's given. Creative quality determines CPM efficiency, CTR, and CVR. Roughly 80% of your time should sit in creative strategy, not campaign structure — this compounds the LUCE_04 mandate ("creative is the targeting") into an operating discipline, not a slogan.

Law 3 — Attribution Is a Story, Not a Truth. Every attribution model lies in a specific, knowable direction. The goal is triangulation across platform-reported, incrementality testing, post-purchase survey, and MER — not finding "the correct number" (Section 7).

#1.2 The Media Buyer's Hierarchy of Needs

Level 5: Portfolio Thinking (MER, payback optimization)
Level 4: Channel Orchestration (cross-platform sequencing)
Level 3: Creative Systems (production, testing, iteration)
Level 2: Campaign Architecture (structure, bidding, targeting)
Level 1: Economic Foundation (margins, LTV, CAC targets)

Work bottom-up. Most operators skip to Level 2 without securing Level 1 — this is the single most common reason a scaling attempt stalls out around $10–20k/month: the campaign architecture is sophisticated, but nobody actually knows the breakeven number the architecture is supposed to serve.

#1.3 Key Metrics Hierarchy — Mid-2026

---

#SECTION 2: META ADVERTISING MASTERY

#2.1 Account Architecture — The 2026 Operating Model

Complex campaign trees remain over. Meta's algorithm — now running on the Andromeda infrastructure — needs consolidation, not fragmentation, to learn efficiently.

The Proven 3-Campaign Architecture:

CAMPAIGN 1: ACQUISITION (Prospecting)
├── Campaign Budget Optimization (CBO) ON
├── Bid Strategy: Highest Volume (or Cost Cap if profitable)
├── Ad Set 1: Broad (no interests, no demo beyond age/country)
│   └── 3–5 ads (varied creative formats/angles)
├── Ad Set 2: Interest Stack (1–3 broad interests max)
│   └── 3–5 ads (same top performers)
└── Ad Set 3: Lookalike (1–3% LTV customers)
    └── 3–5 ads (same top performers)

CAMPAIGN 2: RETARGETING (Warm Audiences)
├── Ad Set Budget (ABO) — fixed daily spend
├── 0–30-day website visitors (excl. purchasers), segmented 0–7 / 8–30
├── Video viewers 50–75%
├── IG/FB engagers 30–60 days
└── Ads: social proof heavy, urgency, objection handling

CAMPAIGN 3: RETENTION (Customer Winback)
├── ABO — small fixed budget
├── Past purchasers 90–365 days
├── Exclude 30-day purchasers
└── Ads: cross-sell, replenishment, loyalty messaging

Why broad still works: Meta's lookalike modeling on your pixel and CAPI data operates on a graph no manual interest stack can approximate. Broad targeting lets the system leverage that full graph. Interest targeting remains, at best, a secondary test — not a primary strategy.

What changed since the original module — segmentation inside retargeting. The July 2026 removal of the off-platform activity opt-out grew the raw size of your 0–30-day retargeting pool. Left as one flat audience, it now dilutes toward lower-intent users. Split it 0–7 / 8–30 days and treat the 0–7 segment as your highest-intent, highest-frequency-tolerance audience.

#2.2 Campaign Budgeting Philosophy

The Nick Shackelford / Cody Plofker Scaling Model:

Phase 1 (Testing, $100–300/day): ABO on all campaigns. Test creative. Find winners.
Phase 2 (Scaling, $300–1,000/day): CBO on acquisition. Retargeting stays ABO.
                                     Begin budget consolidation.
Phase 3 (Scaling, $1,000–5,000/day): Increase CBO budget 15–20% every 48–72 hours
                                       when MER is healthy.
Phase 4 (Scale, $5,000+/day): Duplicate winning CBOs. Add Advantage+ Shopping
                                Campaigns (ASC) as a parallel structure.

The 20% Rule: never increase a campaign budget more than 20% in 24 hours without resetting the learning phase. The algorithm needs stability to optimize — this is the same rule as LUCE_04 Section 2.4, restated here because at Phase 3–4 spend levels, violating it is far more expensive per mistake.

Cost Cap vs. Highest Volume vs. Value Optimization:

• Highest Volume: scaling, strong creative, when you trust the delivery system to find efficient volume.
• Cost Cap: a hard CPA ceiling exists, contribution margin is being actively managed, or you're testing new audiences.
• Value Optimization: 50+ purchase events/week per ad set, and you want higher-AOV customers specifically.

#2.3 Advantage+ Shopping Campaigns — The Andromeda-Era Standard

What changed since the original module: ASC is no longer simply "the fully automated option you turn on at scale." Since Andromeda's full rollout (October 2025), roughly 78% of all Meta spend runs through Advantage+, and its CPA advantage is measurably tiered:

The operating implication: above $10k/month, lean into ASC as a primary structure — the system has enough purchase-event volume weekly to outlearn manual audience-building. Below $2k/month, ASC is still worth running (it's rarely worse than manual), but don't expect it to compensate for a weak creative pipeline; the CPA delta is small enough that creative quality remains the dominant variable, as covered in LUCE_04.

ASC setup:

• Budget: start at 20–30% of total Meta budget.
• Existing customer audience: upload customer list, set spend cap 10–20%.
• Creative: load all proven winners (5–10 assets minimum) — ASC performs in proportion to what it's fed, not despite it.
• Catalog: connect for dynamic elements.
• Attribution: 7-day click, 1-day view (standard; see Section 2.8).

When ASC outperforms manual: catalog-heavy stores (fashion, home goods, multi-SKU), accounts with 1,000+ purchases/month of pixel data, high retargeting volume.

When manual outperforms ASC: new brands with weak pixel data, single-product stores, when you need granular creative-testing control.

#2.4 Creative Testing Methodology — The Machine

The Barry Hott / Charlotte Henry Method:

Step 1 — Hypothesis Matrix. Every test is a hypothesis: "We believe [creative element X] will improve [metric Y] because [consumer psychology reason Z]." Example: "We believe a testimonial hook showing real skin transformation will improve hook rate by 15% because viewers see themselves in the before-state."

Step 2 — Variable Isolation. Test one variable at a time: hook (first 3 seconds), body copy angle, CTA text, visual format (UGC vs. polished), music/sound design, aspect ratio (9:16 vs. 1:1 vs. 4:5).

Step 3 — Test Structure:

Testing Ad Set:
- Budget: $30–50/day
- Audience: Broad
- 3–5 variations of the same concept
- Run 3–5 days minimum (50+ impressions per variation)
- Winner threshold: 1.5× average CTR AND lower CPA

Step 4 — Iteration Velocity. Top-tier creative teams ship 10–20 new creative concepts per week — not 10–20 individual ads, 10–20 concepts (each with 2–3 executions = 20–60 actual ads). Section 8.2 shows how AI-assisted drafting makes this achievable without an agency-sized team.

The Creative Testing Funnel:

Week 1: 20 new concepts tested
Week 2: 5 survivors (25% win rate is excellent)
Week 3: 2 strong performers scaled
Week 4: 1 control asset → iterate on this winner
Month 2: 20 new concepts testing against proven control

#2.5 Creative Frameworks That Print Money

The Hook Formula Library:

The 3-Second Rule: the first 3 seconds determine whether the ad lives or dies. In those 3 seconds: stop the scroll (pattern interrupt), signal relevance (who this is for), create curiosity (why keep watching).

The AIDA+ Framework for DTC:

Attention: Hook (3 seconds)
Interest: Problem amplification (10–15 seconds)
Desire: Solution + differentiation (20–30 seconds)
Action: CTA + urgency (final 5 seconds)
+Social Proof: Woven throughout — not a separate section

This is the same AIDA+ formula introduced at the LUCE_04 level (Section 2.7); here it's the backbone of the full YouTube direct-response script in Section 5.3.

#2.6 Fatigue Management System

Fatigue Detection Metrics:

Alert Level 1 (Monitor): Frequency >3.0 | CTR down 15%
Alert Level 2 (Action Required): Frequency >3.5 | CTR down 20% | CPA up 20%
Alert Level 3 (Replace): Frequency >4.5 | CTR down 30% | CPA up 35%

The Refresh Protocol:

1. Change the hook (first 3 seconds) — quickest fix, often extends life 2–4 weeks.
2. Change the format (same message, different execution).
3. Change the angle entirely.
4. Archive and rotate back after 6–8 weeks (audiences refresh).

Evergreen Creative Strategy: build a creative "bank" of 15–20 proven assets and rotate on a 6–8 week cycle. Some brands run the same top 5 ads for 12–18 months by managing frequency through rotation alone.

#2.7 Bidding Strategies Deep Dive

Cost Cap — Setting It Correctly:

Step 1: Run Highest Volume for 7 days to establish baseline CPA
Step 2: Set cost cap at 110–120% of baseline CPA
Step 3: Allow 2–3 days to exit learning
Step 4: If delivery is throttled (underspending >30%), raise cap 10%
Step 5: If CPA is comfortably under cap, lower cap 5–10%

Bid Cap (Advanced): used only to control individual auction bids, not just CPA. Use for maximizing impression share on specific placements, when CPMs spike (Q4), or advanced arbitrage plays.

ROAS Target ("Minimum ROAS"): effective only with 50+ weekly purchase events. Set at 80–90% of your target ROAS to give the algorithm room to operate. Typically underspends — use only when efficiency matters more than volume.

#2.8 Attribution Settings and the July 2026 Retargeting Shift

The Attribution Window Debate:

• Meta default: 7-day click, 1-day view.
• Conservative: 7-day click only.
• Aggressive: 7-day click, 7-day view (inflated numbers).

Recommendation (Cody Plofker framework): use 7-day click, 1-day view for optimization. This is the algorithm's signal input — changing it doesn't change how Meta tracks; it changes what events Meta optimizes for.

What you actually need, layered:

1. Meta reported (directional — relative performance).
2. GA4 (different cookie model — cross-channel comparison).
3. Northbeam/Triple Whale/Rockerbox (MTA — channel contribution).
4. Post-purchase survey ("How did you hear about us?") — declared attribution.
5. Incrementality tests (true causality — geo holdout or ghost ads).

What changed since the original module: July 2026's removal of the standalone off-platform activity opt-out (folded into "Activity from other businesses") means Meta's own signal — the input to layer 1 — now draws from a larger off-platform data pool than it did in 2025. This doesn't fix attribution's fundamental unreliability; it means layer 1's numbers may shift for reasons that have nothing to do with your campaigns. Treat any sudden MER or ROAS jump around a Meta policy update as a measurement artifact until layers 2–4 confirm it.

---

#SECTION 3: TIKTOK ADS MASTERY

#3.1 Why TikTok Is Different — And the Post-JV Reality

TikTok is not a social platform with an ad network bolted on. It's an entertainment platform with an ad product, and the content norms are inverted from Meta: native, organic-feeling content outperforms polished ads by 3–5×. The user's psychology: "I am here to be entertained. Show me something interesting, not an ad."

TikTok Wave Algorithm:

1. ~200 initial views (bot/quality check)
2. ~2,000 views if engagement passes threshold (>10% like rate, >50% hold rate)
3. ~20,000 views (second filter)
4. ~200,000 → viral potential

Higher organic engagement signals boost paid delivery efficiency — this logic applies to paid content as much as organic.

What changed since the original module: the TikTok USDS joint venture (Oracle/Silver Lake/MGX ~45–50% ownership, ByteDance retaining 19.9%) closed January 22–23, 2026. Ban risk — the shadow over every 2025 TikTok media plan — is resolved. But the US algorithm is retraining on a US-only data environment, and delivery has been measurably more volatile through 2026 than pre-divestiture. At scale, this means: hold larger creative reserves than you would on Meta, and don't read one bad week as platform decay or one great week as the new baseline.

#3.2 TikTok Campaign Architecture

CAMPAIGN LEVEL:
- Campaign Objective: Product Sales (Catalog) or Website Conversions
- Budget: Campaign-level for scale
- CBO equivalent: "Campaign Budget" in TikTok

AD GROUP LEVEL:
- Ad Group 1: Broad (18–55, all genders, no interest targeting)
- Ad Group 2: Interest targeting (3–5 relevant interests)
- Ad Group 3: TikTok-specific behaviors (video interactions, creator followers)
- Ad Group 4: Custom audiences (website visitors, email lists)

AD LEVEL:
- 3–5 ads per ad group
- Mix of Spark Ads (boosted organics) and non-Spark (regular ads)

#3.3 Creative Strategy for TikTok

The Native Content Principle: trending sounds (licensed for ads), TikTok text overlays/captions, vertical 9:16 only (never repurpose Meta square content), authentic voiceover or camera-talking style, POV hooks ("POV: You just discovered...").

TikTok Hook Formulas:

1. "I tried [product] for 30 days and here's what happened"
2. "Why I switched from [competitor] to [brand]"
3. "This is the [product] everyone is talking about"
4. "Rating [product] as a [relevant person]"
5. "Honest review of [product] — is it worth it?"
6. "Things I wish I knew before buying [product category]"
7. "Watch me try [product] for the first time"

Spark Ads: boost your own organic content as ads. Social proof carries over (likes/comments/shares stay on the post), CTR runs 17–30% higher than standard in-feed dark posts, and CPMs improve because TikTok rewards native-feeling content.

Spark Ad strategy:

1. Post organic content on brand or creator accounts.
2. Wait for organic engagement signal (>5% engagement rate).
3. "Whitelist" the post via Ads Manager.
4. Scale spend on proven organic posts.

Creator Whitelisting: find UGC creators or micro-influencers (10K–500K followers), whitelist their posts as dark ads not appearing on their own feed. Benefits: authentic accounts, built-in social proof, lower CPMs.

#3.4 TikTok-Specific Metrics

TikTok's golden metric: Average Watch Time ÷ Video Length = Hold Rate. Target >40% — if users watch 40%+ of your ad, it will distribute.

#3.5 TikTok Shop Integration — GMV Max and the July 2026 Rulebook

TikTok Shop's native checkout has made it the fastest-growing GMV channel for beauty, health, and home categories — often growing faster than the brand's own DTC site.

What changed since the original module:

• GMV Max consolidates ad campaigns, LIVE boosts, and affiliate-driven content buying into one automated budget-allocation system, replacing the earlier practice of running these as separate manual campaigns.
• Account Health Rating (AHR) replaced the old violation-points penalty system in July 2026 — a continuous composite score, not a periodic tally.
• Store Rating recalculates continuously.
• 60-day After-sales Handling Time replaces Customer Complaint Rate as the primary service metric.
• Shipping still requires USPS labels bought through TikTok Shipping (since January 2026); 2-day shipping and 48-hour tracking-scan rules remain in force.

The conversion delta that should drive your content strategy: Shop-tagged content converts at 3.7% vs. 1.8% for non-Shop-tagged content. At scale, this is the single highest-leverage tag you can add to a video — tag every product mention.

TikTok Shop strategy at scale:

1. List your top 3–5 products on TikTok Shop.
2. Activate LIVE shopping (7-day streaming schedule, 2–3 hours/session).
3. Connect affiliate creators via the Creator Marketplace, run commission through GMV Max.
4. Use Shoppable Ads — video ads linking directly to TikTok Shop.
5. Track TikTok Shop revenue separately from Shopify — it's additive, not cannibalistic, but only if your attribution stack (Section 7.1) keeps them distinct.

Affiliate economics at scale: referral fee 6% (3% for a new seller's first 30 days), affiliate commissions typically 5–25% set by seller. As GMV grows, negotiate tiered commissions — top-performing affiliates on lower commission, new/unproven affiliates on higher commission to earn pickup. This is portfolio thinking applied to affiliates, the same logic as Section 7.3's channel allocation.

---

#SECTION 4: GOOGLE ADS MASTERY FOR E-COMMERCE

#4.1 Google's Role in the Acquisition Funnel

Google captures demand; Meta and TikTok create it. Meta/TikTok show ads to people who weren't thinking about your product — they create demand. Google shows ads to people already searching — it captures demand. Never reduce Google to fund Meta if Google is profitable — they serve different functions, and cutting one to feed the other misreads what each channel is for.

#4.2 The Google E-commerce Campaign Stack

What changed since the original module: "PMax is the only way to run Shopping" is stale advice. The mid-2026 consensus is a hybrid — PMax for automated breadth, Standard Shopping for control and query-level visibility on your most important SKUs and brand terms.

CAMPAIGN 1: Brand Search (protect your brand name)
- Keywords: [brand name], [brand] + product types
- Bidding: Target Impression Share (top) 90%+
- Budget: small (brand searches are cheap)

CAMPAIGN 2: Performance Max (PMax)
- Feed-based (catalog required)
- All assets: headlines, descriptions, images, videos
- Signal audiences: past purchasers, email lists, website converters
- Bidding: Target ROAS or Maximize Conversion Value

CAMPAIGN 3: Shopping Standard (run alongside PMax, not instead of it)
- More control over product-level bidding
- Use for your top 20% of SKUs
- Custom labels: margin tier, star rating, top seller

CAMPAIGN 4: Dynamic Search Ads (DSA)
- Captures long-tail queries Shopping misses
- Use page feeds for accuracy
- Exclude branded terms (bid separately)

CAMPAIGN 5: Demand Gen (video — the only conversion-optimized video
             campaign type since Video Action Campaigns were killed
             April 2025)
- In-feed and in-stream video assets
- Include YouTube Shorts placement for the cheapest reach on
  the platform (CPM ~$4–4.85, CTR ~1.24%)
- Target: custom intent audiences (searched your category recently)

#4.3 Performance Max — Advanced Optimization

The John Moran / Mike Rhodes PMax Framework:

Asset group strategy: segment by product category, not audience:

Asset Group 1: [Category A] — matching images, headlines about Category A
Asset Group 2: [Category B] — matching images, headlines about Category B
Asset Group 3: [Top Sellers] — bestselling products, best UGC images
Asset Group 4: [Clearance/Sale] — promotional messaging

Signal audiences (priority order): converters list (strongest signal) → high-value customers (LTV segment) → email list → website visitors (30-day) → custom intent (category-keyword searchers). PMax signals are suggestions, not restrictions — don't over-invest in perfecting them at the expense of the asset group's actual creative quality.

Budget and ROAS targets: start tROAS at 200–300% (2–3×), let it learn 30 days, then adjust based on actuals. Never set tROAS more than 20% above recent actuals — PMax underspends when the target is unreachable.

What to actually monitor: asset group performance reports, search terms insight report (limited but useful), audience insights (who's actually converting), placement exclusions (manually exclude low-quality Display placements).

#4.4 Shopping Campaigns — Advanced Bidding

The custom label system:

Custom Label 0: Margin Tier (high/medium/low/negative)
Custom Label 1: Inventory Status (in-stock/low-stock/clearance)
Custom Label 2: Performance Tier (bestseller/mid/slow)
Custom Label 3: Seasonality (seasonal/evergreen/holiday)
Custom Label 4: Price Tier (<$25/$25–75/$75–150/$150+)

Bid higher on High Margin + Bestseller. Bid lower or exclude Negative Margin + Slow Mover.

#4.5 GA4, Merchant Center, and Profit Bidding

Critical tracking setup: GA4 ↔ Google Ads linked (conversion import); Merchant Center ↔ Google Ads linked (Shopping feed); Enhanced Ecommerce GA4 events (view_item, add_to_cart, begin_checkout, purchase); Enhanced Conversions (hashed email/phone with conversion events); conversion modeling enabled for cookieless measurement.

The Profit Bidding Approach (Advanced): bid on profit, not revenue, using conversion value rules that discount conversions by COGS:

Product A: 70% margin → conversion value = Revenue × 0.70
Product B: 40% margin → conversion value = Revenue × 0.40

This tells Google to optimize for profit rather than raw revenue — the same principle as Section 11.1's contribution-margin bidding, applied at the Google Ads layer specifically.

---

#SECTION 5: YOUTUBE ADVERTISING FOR E-COMMERCE

#5.1 Why YouTube Is Underutilized by DTC Brands

A customer acquired through YouTube shows roughly 2.2× higher LTV on average than a Meta-acquired customer (Brett Curry, OMG Commerce data). Reason: YouTube is lean-back, long-form, high-intent — a viewer who watches 30–90 seconds and then buys is highly qualified.

#5.2 YouTube Ad Formats for DTC

In-Stream Skippable (most used): 15+ seconds, skippable after 5. First 5 seconds are the hook — make them unskippable in effect. You pay only if the viewer watches 30+ seconds or clicks. Best for demo-heavy products, storytelling brands, repurchase reminders.

In-Stream Non-Skippable (15–30 seconds): forces full viewing, higher CPM. Use for brand awareness and retargeting-sequence companions.

Bumper Ads (6 seconds): non-skippable pre-roll. Use for brand recall and retargeting-sequence support. One clear benefit + logo.

Discovery/In-Feed Ads: appear as recommended videos, user-initiated click = high intent. Good for educational content and product explainers.

YouTube Shorts (the 2026 addition): CPM ~$4–4.85, CTR ~1.24% — the cheapest reach on the platform, run through Demand Gen. The natural entry point for a smaller creative budget before investing in long-form in-stream production.

#5.3 The YouTube Direct Response Formula (Brett Curry / Ezra Firestone Method)

0:00–0:05 — HOOK (non-skippable window)
  • Pattern interrupt visual
  • Bold claim or question
  • Who this is for — self-selection

0:05–0:30 — PROBLEM AGITATION
  • Amplify the problem this product solves
  • Empathize with the viewer's frustration
  • Use "before" language

0:30–1:00 — SOLUTION INTRODUCTION
  • Introduce product as the natural solution
  • Key differentiators (a story, not a list)
  • Social proof woven in (numbers, testimonials)

1:00–1:30 — DEMONSTRATION + PROOF
  • Show product working
  • Before/after if applicable
  • Testimonial clips

1:30–2:00 — CTA + URGENCY
  • Specific CTA ("Click below to get 20% off today only")
  • Repeat offer value
  • Closing social proof (star rating, customer count)

The 5-second hook test: watch your first 5 seconds with sound off. If you can't understand what the ad is about and feel intrigued, it fails.

#5.4 YouTube Targeting

Audience types: in-market segments (people actively researching your category — the highest-intent audience Google offers); custom intent segments (built from what people searched in the past 7 days — e.g., "best moisturizer for dry skin" for a skincare brand); life events (moving, marriage, having a baby); YouTube remarketing (channel viewers, site visitors, cart abandoners).

Sequential advertising (advanced): show ads in a predetermined sequence — Ad 1 awareness (problem agitation, 30s) → Ad 2 consideration (solution + demo, 60s, shown to Ad 1 viewers) → Ad 3 conversion (offer + urgency, 30s, shown to site visitors). This mirrors an email nurture sequence in video form.

---

#SECTION 6: PINTEREST, THREADS, AND REDDIT — THE EXTENDED CHEAP-INVENTORY PLAYBOOK

LUCE_04 Section 5 introduced these as a lean-tier extension. At scale, each earns a fuller architecture of its own — but the sequencing rule from LUCE_04 still applies: add a channel only once your primary two are stable.

#6.1 Pinterest Advertising

Pinterest carries ~450M+ monthly active users, skews 85% female, and shows the highest purchase intent of any social platform — it's a planning platform, not a scroll-and-forget feed. Path to purchase runs 30–90 days, but conversion is high once it happens. Average order values run roughly 2× other social channels for home, fashion, beauty, and food.

Campaign types: Shopping Ads (catalog, direct product cards — essential for e-commerce); Standard Pins (static images, vertical 2:3 ratio); Video Pins (6–15s short or 15–60s medium, autoplay muted — use text overlay); Carousel Pins (product variation showcase); Idea Pins (organic-first, Stories-format content marketing).

Structure:

Campaign: Product Category
└── Ad Group: Broad Interest (your category) → Shopping + Static Images
└── Ad Group: Keywords (search intent) → Shopping ads for top products
└── Ad Group: Retargeting (website visitors) → Dynamic product ads from catalog
└── Ad Group: Actalike (lookalike equivalent) → Based on customer email list

Creative standard: vertical 2:3 (1000×1500px) always outperforms square; light, bright, high-contrast backgrounds perform best; add text overlay (brand name or key benefit, mobile-readable); Pinterest trends run 60+ days ahead of seasonal timing — plan creative accordingly; ask "would a real pinner save this to a board?"

Mid-2026 pricing: retail CPC $0.50–0.70, best-case CPA $7–8.

#6.2 Threads Ads — The Early-Mover Window

New for 2026. Threads ads run through the same Meta Ads Manager infrastructure, currently in an early-mover beta pricing window: CPC ~$0.68, CPM ~$4.82 — meaningfully cheaper than Reels or Feed. This window is a function of low advertiser density, not a structural advantage of the platform; expect it to normalize toward Meta's other placements as more advertisers arrive.

How to use it at scale: don't build separate Threads-native creative initially — reuse your best-performing Meta cold-acquisition creative (the format transfers reasonably well) and treat the early CPC advantage as a temporary discount on testing volume. Revisit the cost delta quarterly; the moment Threads CPC approaches Reels CPM-equivalent, fold it back into your standard Meta placement mix rather than managing it separately.

#6.3 Reddit Advertising

Reddit's native tracking is weak relative to Meta/Google, and its audience punishes anything that reads as a hard sell. It works for niches with an active, opinionated subreddit and a product that survives blunt discussion (functional, problem-solving categories do better than aspirational/lifestyle ones).

Mid-2026 pricing: CPC ~$1.25. Pair every Reddit campaign with strict UTM discipline (Section 7.1) since platform-reported numbers alone are unreliable here — treat early results skeptically and confirm with GA4/MTA before scaling budget.

#6.4 When to Add Each Channel — The Decision Framework

Add a new channel when all five hold:

1. Current channels are profitable (MER above target).
2. You have creative production capacity for the new format (Section 8).
3. You have measurement systems to track the new channel distinctly (Section 7.1).
4. You have a 60-day budget for channel ramp-up without judging it prematurely.
5. The channel's demographic matches your ICP.

---

#SECTION 7: ADVANCED CROSS-PLATFORM STRATEGY

#7.1 The Attribution Stack — Building a Complete View

The fatal mistake: trusting any single attribution model.

Layer 1: Platform Reported (directional — relative comparison only)
         • Meta Ads Manager · Google Analytics 4 · TikTok Ads Manager

Layer 2: Multi-Touch Attribution (MTA)
         • Northbeam, Triple Whale, Rockerbox, or Elevar
         • Shows channel contribution across touchpoints
         • Better for budget allocation decisions

Layer 3: Media Mix Modeling (MMM) — for $1M+/month spenders
         • Statistical model using historical revenue and spend
         • Not real-time, most accurate for budget planning
         • Tools: Meridian (Google), Robyn (Meta), Recast

Layer 4: Incrementality Testing
         • Geo holdout tests (turn off a channel in X markets)
         • Ghost ads / holdout groups
         • "True lift" measurement — run quarterly minimum

Layer 5: Post-Purchase Survey
         • "How did you hear about us?" with attribution options
         • Segment by new vs. returning customer
         • Compare to platform-reported → understand over/under-counting

The synthesis process: don't try to reconcile these into one number. Use them for different decisions — day-to-day optimization: platform-reported + MTA; quarterly budget allocation: MMM + incrementality; brand-awareness validation: post-purchase survey. See LUCE_06 for the full dashboard build that operationalizes this stack.

#7.2 The Full-Funnel Cross-Platform Playbook

COLD AWARENESS:
- TikTok (entertainment-first, broad reach)
- YouTube In-Stream / Shorts (attention, long-format or cheap reach)
- Meta Prospecting (scale + lookalike)
- Pinterest (visual discovery, planning audiences)
- Threads (early-mover cheap reach, Section 6.2)

WARM CONSIDERATION:
- Meta Retargeting (0–30-day website visitors, segmented)
- YouTube Remarketing (video viewers sequence)
- Google DSA (captures searches from warmed prospects)
- Pinterest Retargeting (catalog-based)

HOT INTENT / CART ABANDON:
- Meta Dynamic Product Ads (catalog-based personalization)
- Google Shopping (brand + product search capture)
- Email (the best retargeting channel you own)
- SMS (for opt-in subscribers)

POST-PURCHASE / RETENTION:
- Email (primary)
- Meta Retention Campaign (exclude 30-day, target 90–365-day)
- TikTok for brand engagement
- Loyalty program activation (owned channel, not ads)

#7.3 New Platform CAC vs. Blended CAC — The Budget Allocation Decision

The law of diminishing returns: every channel has an efficiency curve. The first dollar on a channel is most efficient; the thousandth dollar is least efficient. Scaling across channels beats exhausting one.

Portfolio Budget Allocation (Andrew Faris framework, updated for 2026 channel mix):

Starting portfolio ($10K/month budget):
- Meta: 55% ($5,500)
- Google: 25% ($2,500)
- TikTok: 15% ($1,500)
- Other/Test (Pinterest/Threads/Reddit): 5% ($500)

Scaling portfolio ($100K/month budget):
- Meta: 40% ($40,000)
- Google: 30% ($30,000)
- TikTok: 18% ($18,000)
- YouTube: 7% ($7,000)
- Pinterest/Threads/Reddit: 5% ($5,000)

At $100K+/month:
- Meta: 32–38%
- Google: 25–30%
- TikTok: 18–22%
- YouTube: 8–12%
- Pinterest/Threads/Reddit: 3–5%
- Influencer/UGC distribution: 5–10%

What changed since the original module: TikTok's allocation share has risen relative to the 2025 model, reflecting its lower CPMs and GMV Max consolidation making spend easier to manage at scale; Meta's share has compressed slightly to make room, reflecting the diminishing marginal edge of Advantage+ automation once an account is already well past the $10k/month tier where it performs best.

#7.4 Incrementality Testing — The Advanced Playbook

Geo holdout test:

1. Identify test and control markets (similar demographics, similar revenue).
2. Pause one channel in the test market, maintain in control.
3. Run 3–4 weeks (minimum for statistical significance).
4. Incremental revenue = control performance − test performance.
5. True incrementality factor = incremental revenue ÷ platform-reported revenue.

Ghost ads (conversion lift study): Meta and TikTok both offer built-in conversion lift studies — a holdout group sees a PSA instead of your ad, measuring true causal lift versus would-have-purchased-anyway baseline.

Typical findings:

• Meta organic-social effect: 30–50% of Meta-attributed revenue would have happened without ads.
• Google Brand Search: 20–40% incremental (many would have found you organically).
• TikTok: often 60–80% incremental — it tends to create genuine new demand rather than capture existing intent.

This is why a naive platform-reported ROAS comparison consistently overrates Brand Search and underrates TikTok — factor incrementality into any budget-reallocation decision, not just reported ROAS.

---

#SECTION 8: CREATIVE PRODUCTION SYSTEMS AT SCALE

#8.1 The Creative-Scale Relationship

The fundamental law: at scale, creative velocity is the limiting factor. More budget accelerates fatigue; more creative is the only fix.

#8.2 The AI-Assisted Creative Production System — Advanced

LUCE_04 Section 6.4 introduced the hook matrix and variant batching for a lean operator running 3–5 concepts a week. At $10k+/month, the same system needs to scale to 8–40 concepts/week — the workflow, not the underlying logic, is what changes.

1. The Hook Matrix at scale. Expand the matrix from 3 angles × 7 hook types to your full angle library (typically 6–10 angles by the time a product has run for months) × 7 hook types = 42–70 candidate hook lines per drafting session. Run this weekly, not ad hoc — a dedicated creative strategist (per the table above) owns the matrix and routes surviving hooks to the shoot calendar.

2. Structured UGC briefs at volume. With 3–5 creators in rotation, standardize the brief template (LUCE_04 Section 6.4) into a repeatable intake form so every creator gets the same product brief, angle assignment, and DO NOT list — variance in brief quality is a common, invisible cause of inconsistent creative performance across a creator roster.

3. Variant batching as the default, not the exception. At scale, every shoot should be planned to yield 6–9 executions minimum (hook swaps × CTA swaps), not treated as a single ad. This is the single highest-leverage move for hitting the concepts/week targets in Section 8.1 without proportionally increasing shoot volume.

4. AI-assisted analysis, not just drafting. Beyond hook generation, use an AI assistant to summarize weekly winner patterns (which hook types, which angles, which formats are winning) from your ad platform exports — feeding that summary back into the next week's hypothesis matrix (Section 2.4, Step 1) closes the loop between what's tested and what's hypothesized next.

Governance rule at every scale tier: AI drafts, a human approves. No AI-generated hook, claim, or script reaches a live ad or a creator brief without a compliance and accuracy pass — this is non-negotiable given ad-policy and health/comparison-claim risk.

#8.3 Creative Research Systems — Finding What to Test

The 5-Source Creative Research System:

1. Reddit mining: search your product category, find how customers describe problems in their own words; use that language verbatim in ads.
2. Amazon review mining: 1-star reviews of competitors reveal problems to solve; 5-star reviews of your product reveal what to amplify. Tools: Jungle Scout, Helium 10.
3. Customer interview synthesis: 20 customer interviews → common language patterns → hook angles.
4. Competitor spy tools: Meta Ad Library (free); AdSpy ($149/mo flat — historical data, engagement metrics); Foreplay.co (save/categorize winning ads); Minea ($49/$99/mo — cross-platform creative intelligence, including TikTok).
5. Winning-ad pattern study: analyze your own top performers for common elements (hook style, format, length, voiceover vs. text) and extract the winning formula to iterate on.

#8.4 The Testing Velocity Compounding Effect

Creative testing compounds. A brand testing 10 concepts/week for 12 months has tested 480 concepts; a brand testing 2/week has tested 96. At a 10–20% win rate: 480 concepts → 48–96 proven winners; 96 concepts → 9–19 proven winners. The brand with 80 proven winners has 5–8× more scale options than the brand with 15 — an asymmetric, compounding advantage.

Test more → Find more winners → More scale options →
More revenue → Fund more creative production → Test more

---

#SECTION 9: SCALING FRAMEWORKS

#9.1 The 4-Phase Scaling Model

Phase 1: Foundation ($0–$5K/month ad spend). Establish pixel/CAPI data, find creative winners, reach profitable unit economics. Run ABO to isolate winners. Test 3–5 creative angles/week. Target ROAS: breakeven or better. Focus: learning, not scaling.

Phase 2: Acceleration ($5K–$30K/month). Identify scalable creative-audience combinations, achieve consistent MER. Transition top performers to CBO. Add a second channel (Google, if not already running). Creative: 8–12 new concepts/week. Target MER: 3–4×.

Phase 3: Scale ($30K–$150K/month). Portfolio expansion, cross-channel sophistication, contribution-margin defense. Full channel portfolio active. MTA attribution implemented. Creative team: 2–3 dedicated people. Target MER: 3–5× (compression at scale is normal). Introduce ASC and PMax fully.

Phase 4: Efficiency Optimization ($150K+/month). Profit maximization, incrementality testing, media mix modeling. MMM implemented quarterly. Geo incrementality tests monthly. Budget allocation driven by data, not intuition. MER: 3–4× blended — accept compression, compensate with LTV programs.

#9.2 Horizontal vs. Vertical Scaling

Vertical (increase budget on winners): simple, but creates diminishing returns and creative fatigue faster. Appropriate when the creative pipeline is strong.

Horizontal (duplicate winners into new audiences/platforms): maintains efficiency longer, more complex account management. Appropriate once a winning formula is found.

The hybrid approach (recommended):

1. Scale vertically 15–20% every 72 hours
2. When CPA degrades >20%, duplicate the winning ad set with a new audience
3. When horizontal options are exhausted, refresh creative
4. Rinse, repeat

---

#SECTION 10: MEDIA BUYING OPERATIONS

#10.1 The Daily Rhythm of a Professional Media Buyer

7:00 AM — Morning check (5 min): MER yesterday vs. target; campaigns that over/underspent; account issues (disapprovals, policy flags, AHR changes on TikTok Shop); GA4 previous-day revenue confirmation.

9:00 AM — Optimization window: pause underperformers (CPA >150% of target, 3+ days); scale ads >20% above target; budget adjustments; creative fatigue check.

Midweek (Wed–Thu) — Creative analysis: review hook rates, hold rates, CTRs across all running ads; flag creatives needing refresh; brief next week's concepts.

End of week — Strategic review: weekly MER vs. NC-MER; channel budget allocation review; CAC by channel vs. LTV targets; creative pipeline status; next week's testing calendar.

#10.2 The Media Buyer's Toolkit — Mid-2026 Pricing

Free but essential: Meta Ads Manager, Google Ads Editor, Google Analytics 4, Looker Studio (reporting dashboards), Meta Creative Hub.

#10.3 Agency vs. In-House — The Decision Framework

Hire in-house when: spending $50K+/month on paid media; consistent creative output is needed; you want deep brand/product knowledge baked into media buying; you can afford $60K–$120K/year for a skilled media buyer.

Use an agency when: under $50K/month (agencies bring platform-rep access and beta features); testing a new channel (agency expertise shortens the ramp); you want a fractional CMO model; you don't want to manage a media buyer directly.

The agency contract checklist:

• Retainer fee structure clear (avoid pure % of spend — misaligned incentives).
• Creative services included or separate?
• Reporting cadence: weekly + monthly minimum.
• Account ownership: your accounts, your data — never agency-owned.
• 30–60 day out clause.
• Performance benchmarks written into the contract.

---

#SECTION 11: ADVANCED CONCEPTS

#11.1 Contribution Margin Bidding

Traditional bidding targets ROAS. Advanced bidding targets contribution margin per order.

Contribution Margin = Net Revenue − COGS − Payment Processing − Shipping − Returns
CM% = CM ÷ Revenue × 100

Target CPA = LTV × Target CM%

Worked example: LTV = $200, target 3-month payback, 50% gross margin: target 3-month CM = $200 × 50% = $100 → max CAC = $100 for 1:1 payback in 3 months. If a 6-month payback is acceptable, max CAC scales proportionally with the payback window; if a 12-month payback is acceptable, max CAC can approach the full LTV × GM%.

The Payback Period Framework:

Feed this target CPA into Google's Profit Bidding (Section 4.5) and Meta's Cost Cap (Section 2.7) so both platforms optimize toward the same margin-aware number, not just revenue.

#11.2 The New Customer Revenue (NCR) Operating System

Developed by Taylor Holiday (Common Thread Collective): treats new-customer acquisition as the primary business metric. Core principle: returning-customer revenue is an outcome of past acquisition; new-customer revenue is the leading indicator of future health.

NCR Dashboard — track weekly:

1. NC-MER (new customer revenue ÷ total ad spend) — health signal.
2. New customers acquired — volume signal.
3. Blended CAC (total spend ÷ new customers) — efficiency signal.
4. Day-30 LTV of the new-customer cohort — quality signal.
5. Repeat purchase rate (60–90 day) — retention signal.

NCR decision rule: NC-MER above target → increase spend (acquisition is profitable). NC-MER below target → fix unit economics before scaling; this is a CPA or conversion problem, not a volume problem.

#11.3 Why This Matters More in 2026

The Andromeda-era CPA tiering (Section 2.3) makes NC-MER even more important than it was in the original module: a mid-size account leaning on Advantage+'s automation can see blended CPP look artificially healthy while the algorithm quietly re-serves existing customers more efficiently than it acquires new ones. NC-MER is the check against that failure mode — run it alongside blended MER every week, not as an occasional audit.

---

#SECTION 12: Q4 AND PEAK SEASON PLAYBOOK

#12.1 The Q4 Preparation Timeline

8 weeks before BFCM: creative production at 3–4× normal volume (higher fatigue at peak spend needs a deeper bank); set total Q4 budget and daily pacing; finalize the BFCM offer (stacking discounts, free gifts, bundles); build dedicated sale landing pages; draft the pre-BFCM email warmup sequence.

4 weeks before BFCM: all creative assets complete and approved; campaigns pre-built and set to "paused"; bid strategies set (switch to Maximize Conversions during peak); Google Shopping feed optimized (images, titles, descriptions); inventory confirmed with supplier.

2 weeks before BFCM: "teaser" campaigns live (early access, waitlist); Meta campaign warm-up (small spend to refresh audiences); budget pre-loaded in platforms.

BFCM week: Monday — "Early Access" campaign for email list. Tuesday–Wednesday — scale acquisition. Thursday (Thanksgiving) — organic momentum, minimal new spend. Black Friday — all campaigns at maximum budget. Cyber Monday — maintain with "last chance" urgency creative. Tuesday after BFCM — wind down offers, pivot to "extension" if still profitable.

#12.2 Q4 CPM Management and Seasonality

Q4 tactics:

1. Pre-load retargeting audiences before the CPM spike (October warmup).
2. Shift budget toward owned channels (email, SMS) during peak-CPM days.
3. Use Cost Cap to avoid overpaying at peak CPM.
4. Launch 4–5 AM ET on Black Friday, ahead of competitor wake-up.
5. Watch dayparting — the first 4 hours of Black Friday often show the best MER before competition peaks.

Who should cold-test in Q4: operators with a creative bank of 10+ proven winners built during Q1–Q3, per LUCE_04 Section 7.3's seasonality guidance. Everyone else scales known winners and banks new concepts for the following January.

---

#DECISION TREES

#Tree 1 — When to move from LUCE_04's testing playbook into this module's frameworks

START: You've cleared LUCE_04's testing phase.

IF monthly ad spend < $5,000 AND MER is inconsistent
  → Stay in LUCE_04's frameworks. This module's contribution-margin
    bidding, MMM, and incrementality testing need data volume you
    don't have yet.

IF monthly ad spend $5,000–$30,000 AND MER is consistently ≥3×
  → Enter Phase 2 (Section 9.1). Add a second channel. Move
    retargeting to the segmented 3-audience architecture (Section 2.1).
    Begin NC-MER tracking (Section 11.2) alongside blended MER.

IF monthly ad spend $30,000+ AND full channel portfolio is active
  → Enter Phase 3/4. Implement MTA (Section 7.1 Layer 2). Move
    bidding to contribution-margin targets (Section 11.1). Run
    quarterly incrementality tests (Section 7.4).

IF NC-MER diverges meaningfully from blended MER at any spend level
  → Stop scaling. Diagnose whether Advantage+/GMV Max automation
    is re-serving existing customers before adding more budget.

#Tree 2 — Which attribution layer decides a given question

"Should I pause this ad set today?"
  → Layer 1 (platform-reported) + Layer 2 (MTA). Fast, directional,
    good enough for a same-day call.

"Should I shift 15% of budget from Meta to TikTok this quarter?"
  → Layer 3 (MMM, if $1M+/month) or Layer 4 (incrementality test).
    Platform-reported alone will systematically overrate Brand
    Search and underrate TikTok (Section 7.4) — don't decide this
    on Layer 1 alone.

"Is my brand-awareness spend working?"
  → Layer 5 (post-purchase survey), cross-checked against Layer 4
    findings for the same channel.

"A sudden MER spike appeared right after a Meta policy change."
  → Treat as a measurement artifact (Section 2.8) until Layers 2–4
    confirm the lift is real, not just a bigger reported pool.

---

#KPI TABLE — TARGETS, WARNINGS, KILL SWITCHES

---

#THE 2026 REALITY LAYER

Andromeda's tiered CPA advantage reshapes the scaling math. The −38%/−14% split (Section 2.3) means the ROI on moving from $2k to $10k/month in Meta spend is partly an automation unlock, not just a volume increase — factor this into how you prioritize scaling one channel versus diversifying, since crossing that threshold changes what the algorithm can do for you.

Meta's July 2026 retargeting policy change is a measurement event, not just a targeting improvement. Larger off-platform-sourced pools mean platform-reported numbers can shift for reasons unrelated to campaign performance. Cross-check any sudden lift against Layers 2–4 of the attribution stack (Section 7.1) before reallocating budget on the strength of it alone.

TikTok's post-JV volatility is a real operating condition through 2026, not a temporary blip. Build wider confidence intervals into TikTok forecasting than you would for Meta, and hold a deeper creative reserve for the platform specifically.

TikTok Shop's AHR/GMV Max consolidation (July 2026) merges ads and operations risk. A slow 3PL or a bad batch of returns now shows up in your Account Health Rating, which gates your reach and affiliate marketplace access — treat AHR as a media-buying KPI, not just an ops metric (see the KPI table above).

Google's video options collapsed to one type, and Shopping split back into two. Demand Gen is now the sole conversion-optimized video campaign (VAC died April 2025); simultaneously, the "PMax only" consensus from 2024–2025 gave way to a PMax + Standard Shopping hybrid for visibility. Both changes point the same direction: less blind trust in full automation, more deliberate use of the control layers still available.

Incrementality gaps are now large enough to misallocate real budget if ignored. TikTok's 60–80% incrementality versus Google Brand Search's 20–40% is not a rounding difference — a budget decision made purely on platform-reported ROAS will systematically overfund Brand Search and underfund TikTok. Run at least one incrementality check per major channel per year even below the $1M/month MMM threshold.

Threads' early-mover CPC window is a 2026-specific opportunity. Treat the ~$0.68 CPC as a temporary testing subsidy, not a durable cost advantage — revisit quarterly and fold Threads back into standard Meta placement management once pricing normalizes.

---

#FAILURE MODES

---

#SOPs & CADENCES

Daily (per Section 10.1): morning check (MER, spend pacing, account/AHR issues); midday optimization window (pause/scale per KPI thresholds); confirm revenue reconciliation between platform and GA4/Shopify.

Weekly:

• Full MER + NC-MER review, channel-by-channel.
• Creative fatigue audit (frequency, hook rate, hold rate) against Section 2.6 alert levels.
• Hypothesis matrix session (Section 2.4 Step 1) — draft next week's test concepts using the hook matrix (Section 8.2).
• TikTok Shop sellers: check AHR and Store Rating trend specifically.

Monthly:

• Recompute breakeven ROAS and target CPA against current margins (Section 11.1) — landed costs and platform fees drift.
• Review portfolio budget allocation against the Section 7.3 targets for your current spend tier.
• Audit whether any channel's incrementality assumption needs re-testing.
• Review agency/in-house economics if spend has crossed a threshold (Section 10.3).

Quarterly:

• Run incrementality tests (geo holdout or ghost ads) on at least your two largest channels.
• Revisit Threads/Pinterest/Reddit pricing and allocation (Section 6).
• If $1M+/month, run or refresh the MMM model (Section 7.1 Layer 3).
• Rebuild the creative bank ahead of the next Q4 cycle (Section 12.1's 8-week timeline starts in early October).

---

#WEEK-1 ACTION PLAN

1. Day 1: Compute your current breakeven ROAS, target ROAS, and NC-MER from last 30 days of data (Section 1.1, 11.2). If you can't compute these cleanly, fix your tracking (LUCE_06) before doing anything else in this module.
2. Day 2: Audit your Meta account architecture against the 3-campaign model (Section 2.1) and your retargeting segmentation against the 0–7/8–30-day split.
3. Day 3: Determine your current scaling phase (Section 9.1) and confirm your creative production cadence matches the Section 8.1 table for your spend tier — if it doesn't, that's this week's actual bottleneck, not campaign structure.
4. Day 4: If running TikTok Shop, check AHR and Store Rating directly in Seller Center; if not yet running Shop-tagged content, tag your next batch of videos and compare CVR against untagged content over the next two weeks.
5. Day 5: Review your attribution stack (Section 7.1) — confirm you have at minimum Layers 1–2 (platform-reported + MTA) running, and schedule your first incrementality test if you don't have one on the calendar.
6. Day 6: Run the hypothesis matrix session (Section 2.4) for next week's creative tests, using the AI-assisted hook matrix workflow (Section 8.2) if you haven't set it up yet.
7. Day 7: Set your Q4 prep calendar reminder now (Section 12.1's timeline starts 8 weeks before BFCM) — this is the single most commonly missed step at every spend tier.

---

#SELF-TEST

1. A brand has 60% gross margin and wants a 6-month payback period with an LTV of $300. What's the maximum CAC that satisfies this, using the payback framework in Section 11.1?
2. Explain why platform-reported ROAS systematically overrates Google Brand Search and underrates TikTok, using the incrementality figures from Section 7.4.
3. An account spending $1,500/month on Meta sees Advantage+ underperforming expectations. Per Section 2.3, is this surprising, and what should the operator focus on instead of switching bidding strategies?
4. What are the five layers of the attribution stack (Section 7.1), and which layer(s) should decide a same-day pause/scale decision versus a quarterly budget reallocation?
5. Name the TikTok Shop metric that replaced Customer Complaint Rate in July 2026, and explain why it matters for an ads-focused operator, not just an operations one.

<details> <summary>Answers</summary>

1. 6-month payback with a 3-month baseline in the worked example scales proportionally: 3-month CM at 60% margin = $300 × 60% × (6/12 as a share of LTV realization, using the framework's proportional logic) — concretely, max CAC for a 6-month payback ≈ LTV × GM% × (6/12) = $300 × 0.60 × 0.5 = $90. (If the full annual LTV is already the 12-month figure, a 6-month payback allows up to half of LTV × GM% as CAC.)
2. TikTok runs 60–80% incremental (it creates demand that mostly wouldn't exist otherwise), while Google Brand Search runs only 20–40% incremental (much of that traffic would have found the brand anyway). Platform-reported ROAS doesn't discount for this — it credits both channels at face value, making Brand Search look artificially efficient and TikTok look artificially weaker than its true contribution.
3. Not surprising — under $2,000/month, Advantage+'s CPA advantage is only −14% versus −38% at $10k+/month. The operator should focus on creative quality and testing velocity (Sections 2.4, 8.2), not on tuning bidding strategy, since the algorithm doesn't have enough purchase volume at this spend level to outlearn a strong creative pipeline.
4. Platform-reported, MTA, MMM, incrementality testing, post-purchase survey. Same-day pause/scale: Layer 1 (platform-reported) + Layer 2 (MTA). Quarterly budget reallocation: Layer 3 (MMM, if $1M+/month) and/or Layer 4 (incrementality testing).
5. 60-day After-sales Handling Time. It matters to an ads-focused operator because it feeds directly into Account Health Rating (AHR), which gates reach and affiliate marketplace access — a fulfillment problem now shows up as an advertising problem before it shows up as a return-rate problem.

</details>

---

#CROSS-REFERENCES

• ← LUCE_04 (Advertising): the core module this one extends — come from there once you've cleared consistent breakeven ROAS and are scaling past $5k/month.
• → LUCE_06 (MER & Measurement): the operational home for the full attribution stack (Section 7.1), MER/NC-MER dashboard construction, and incrementality test tracking referenced throughout this module.
• → LUCE_05 (Marketing): organic, email, SMS, and influencer channels that inform the post-purchase survey layer of attribution (Section 7.1 Layer 5) and the retention campaigns in Section 2.1.
• → LUCE_07 (Brand Building) / LUCE_17 (Influencer & UGC System): the creative-sourcing and creator-relationship depth behind Section 8's production system and Section 3.5's TikTok Shop affiliate economics.
• → LUCE_09 (Finance & Scaling): the contribution-margin, payback-period, and cash-flow mechanics that Section 11.1's bidding framework assumes.
• → LUCE_08 (Store CRO) / LUCE_18 (CRO Advanced): the on-site conversion layer that determines whether Section 4.5's profit-bidding inputs and Section 1.3's CVR benchmarks are being met.

---

LUCE — Launch. Unit Economics. Compound. Exit.

Next: → LUCE_05_Marketing.md — organic, email, SMS, influencer, and content strategy for brands that own their audience instead of renting it from an ad platform.