Performance Marketing Knowledge Pack Files

Placement Quality — MFA Detection, Scoring & Supply Chain Validation

MFA Detection Checklist

Score each placement against these signals. Any 3+ flags = high MFA risk, recommend exclusion.

Signal 1: Ad Density

Threshold	Risk	Action
Peak ad density >50% of screen	High — auto-reject	Exclude immediately
Average ad density >35% of screen	High	Flag for review
Mobile ad density >30%	Violation of Coalition for Better Ads	Exclude
Ad density 20-30%	Moderate	Monitor performance
Ad density <20%	Normal	No action

Calculation: Sum heights of all ads within main content portion / total height of main content. Exclude headers, footers, navigation.

Signal 2: Ad Refresh Behavior

Indicator	Risk
Ads auto-refresh while page is in background tab	High
Ad refresh interval <30 seconds	High
Multiple ad auctions per slot per page view	High
1,000+ impressions served to single user in one session	Critical (Adalytics threshold)
Standard refresh (60+ seconds, user-initiated)	Normal

Signal 3: Traffic Source Composition

Traffic Mix	Risk
>70% paid/arbitrage traffic	High MFA signal
>50% paid traffic with low direct/organic	Moderate-High
Balanced mix (organic, direct, search, social, paid)	Normal
Majority organic/direct traffic	Low risk

Check via DSP reporting (some DSPs surface traffic source signals) or third-party tools.

Signal 4: Content Quality

Indicator	Risk
AI-generated or syndicated content with no original reporting	High
No author attribution, no editorial voice	High
Content optimized purely for ad arbitrage (clickbait headlines, slideshow pagination)	High
Thin content surrounding excessive ad placements	High
Original content with bylines, sources, editorial standards	Normal

Signal 5: Site Design & UX

Indicator	Risk
Templated/cookie-cutter layout (identical to other MFA sites)	High
Aggressive pop-ups, auto-play video with sound, full-screen overlays	High
No clear navigation, confusing UX designed to generate accidental clicks	High
Professional design with clear content hierarchy	Normal

ACT Classification Quick Reference

Category	Primary Signal	Secondary Signals
Arbitrage	>50% paid traffic sourcing	Usually accompanied by high ad density
Clutter	Ad-to-content ratio >35% average	Rapidly refreshing placements, MFA layout
Template	Generic/AI content + cookie-cutter design	Fastest-growing category; harder to detect

Source: DeepSee.io ACT Framework

Placement Quality Scorecard

Rate each placement on a 0-100 scale across four dimensions:

Dimension	Weight	Score Criteria
Viewability	30%	>70% = 100, 50-70% = 70, 30-50% = 40, <30% = 0
MFA risk	25%	0 MFA signals = 100, 1-2 signals = 50, 3+ signals = 0
IVT rate	25%	<2% = 100, 2-5% = 70, 5-10% = 30, >10% = 0
Performance	20%	Above avg CPA/CTR = 100, avg = 70, below avg = 30, no conversions = 0

Quality Score = weighted average. Action thresholds:

• 80-100: Allowlist candidate
• 60-79: Monitor monthly
• 40-59: Flag for review, consider exclusion
• <40: Exclude immediately

Supply Chain Validation

Three-layer transparency check (IAB Tech Lab):

Layer	What	Check
ads.txt / app-ads.txt	Publisher declares authorized sellers	Verify seller IDs match what you see in bid stream
sellers.json	SSP declares identity of all sellers	Cross-reference with ads.txt entries
SupplyChain Object	Full path from publisher to buyer	Verify each hop is authorized; fewer hops = lower fraud risk

Run automated validation weekly. IAB Tech Lab provides validation service. Flag any broken chains or unauthorized resellers.

IVT Classification (MRC Standard)

Tier	Type	Detection	Examples
GIVT (General)	Routine filtration	Lists, parameter checks	Bots, spiders, crawlers, non-browser user agents, pre-fetch
SIVT (Sophisticated)	Advanced analytics	Multipoint corroboration	Hijacked devices, adware, malware, misappropriated content

All MRC-accredited vendors must apply GIVT detection. SIVT detection strongly encouraged.

Benchmark: TAG-certified channels achieve 1.41% IVT rate (88% reduction vs industry average).

Verification Vendor Selection Guide

Use Case	Recommended
Full-stack brand safety + MFA + IVT	IAS or DoubleVerify (both GARM-aligned, MRC-accredited)
MFA-specific detection (web + CTV + mobile)	Pixalate (cross-platform MFA risk scoring)
Supply chain validation	IAB Tech Lab automated service
IVT-focused detection	HUMAN Security (2,500+ signals per impression, sub-12ms)
Attention measurement	IAS (attention metrics) or Adelaide (attention-based media quality)

Pre-bid vs post-bid: Use both layers. Pre-bid blocks known risks before spending. Post-bid discovers new patterns and measures actual placement quality. Pre-bid alone misses novel MFA sites. Post-bid alone wastes spend on known risks.