Advanced

agency-data

11 min read

Clean Room Data Activation with IAB Taxonomy Enrichment

For agency data teams activating first-party data in clean rooms. Enrich content signals with IAB 3.0 taxonomy. 3.2x improvement in contextual match rates.

Who It's For

Agency data teams, trading desks, and brand data strategists activating first-party data in clean room environments like AWS Clean Rooms, Snowflake, or LiveRamp

Problem Solved

First-party data lacks the contextual signals needed for programmatic targeting in a cookie-less world. Manual taxonomy classification is impossible at scale

Ready to implement?

Schedule a Demo View Documentation

Why Mixpeek

3.2x improvement in contextual match rates, full IAB 3.0 coverage across 700+ categories, and privacy-compliant enrichment that works within clean room constraints

Overview

As cookies deprecate, agency data teams must extract more value from first-party data in clean rooms. This use case shows how Mixpeek enriches content with contextual signals that power better targeting without relying on third-party identifiers.

Challenges This Solves

Cookie Deprecation

Third-party cookies becoming unavailable for targeting

Impact: Traditional audience targeting breaking down, CPAs increasing

Limited Contextual Signals

First-party data lacks standardized taxonomy classifications

Impact: Cannot activate data for contextual targeting in programmatic

Clean Room Constraints

Data cannot leave clean room environment for enrichment

Impact: External enrichment services incompatible with privacy requirements

Scale Requirements

Millions of assets need classification for activation

Impact: Manual classification impossible, automated solutions required

Implementation Steps

Mixpeek analyzes creative assets and content at ingestion, automatically enriching with IAB 3.0 taxonomy, brand safety scores, and contextual signals that activate in clean room queries

Deploy Within Clean Room

Set up Mixpeek processing within your clean room environment

import { Mixpeek } from 'mixpeek';

// Mixpeek runs within clean room boundary
const client = new Mixpeek({
  apiKey: process.env.MIXPEEK_API_KEY,
  deployment: 'clean-room',  // Clean room deployment mode
  dataResidency: 'aws-clean-rooms'
});

// Process first-party content assets
await client.buckets.connect({
  collection_id: 'brand-content',
  bucket_uri: 's3://clean-room-bucket/content/',
  extractors: [
    'iab-taxonomy-v3',      // IAB 3.0 classification
    'brand-safety',         // GARM brand safety
    'sentiment-analysis',   // Content sentiment
    'entity-extraction',    // Named entities
    'image-embedding'       // Visual similarity
  ],
  settings: {
    taxonomy_depth: 'tier3',  // Full IAB 3.0 depth
    confidence_threshold: 0.7,
    output_format: 'clean_room_compatible'
  }
});

Enrich Content with Taxonomy

Add IAB 3.0 classifications to all content

// Process content and add taxonomy signals
async function enrichContent(assetId: string) {
  const enriched = await client.extract({
    asset_id: assetId,
    extractors: ['iab-taxonomy-v3', 'brand-safety', 'sentiment-analysis']
  });

  return {
    asset_id: assetId,
    iab_categories: enriched.iab_taxonomy.map(t => ({
      tier1: t.tier1,           // e.g., "Technology & Computing"
      tier2: t.tier2,           // e.g., "Consumer Electronics"
      tier3: t.tier3,           // e.g., "Smartphones"
      confidence: t.confidence
    })),
    brand_safety: {
      garm_categories: enriched.brand_safety.garm_flags,
      risk_level: enriched.brand_safety.risk_level,
      safe_for_brands: enriched.brand_safety.safe_for_all_brands
    },
    sentiment: enriched.sentiment.overall,
    entities: enriched.entities
  };
}

Activate in Clean Room Queries

Use enriched signals for targeting queries

// Clean room query using enriched signals
const cleanRoomQuery = `
  SELECT
    brand_assets.asset_id,
    publisher_inventory.placement_id,
    mixpeek_enrichment.iab_categories,
    mixpeek_enrichment.brand_safety
  FROM brand_assets
  JOIN publisher_inventory
    ON mixpeek_enrichment.iab_categories && publisher_inventory.content_categories
  WHERE
    mixpeek_enrichment.brand_safety.safe_for_brands = true
    AND mixpeek_enrichment.iab_categories @> ARRAY['Technology & Computing']
    AND mixpeek_enrichment.confidence >= 0.8
`;

// Execute within clean room
const activationSegment = await cleanRoom.query(cleanRoomQuery);

Measure Activation Performance

Track contextual targeting effectiveness

// Monitor activation metrics
const metrics = await client.analytics.getActivationMetrics({
  collection_id: 'brand-content',
  date_range: { start: '2024-01-01', end: '2024-03-31' },
  breakdown: ['iab_category', 'brand_safety_level']
});

// Key metrics
const report = {
  total_assets_enriched: metrics.total_processed,
  iab_coverage: metrics.with_iab_classification / metrics.total_processed,
  brand_safe_percentage: metrics.brand_safe / metrics.total_processed,
  avg_categories_per_asset: metrics.avg_iab_categories,
  activation_match_rate: metrics.matched_inventory / metrics.total_processed
};

Feature Extractors Used

Image Embedding

Generate visual embeddings for similarity search and clustering

Retriever Stages Used

Expected Outcomes

3.2x improvement in contextual targeting matches

Contextual Match Rate

100% of assets classified to IAB 3.0 Tier 2+

IAB Coverage

45% improvement in CTR for contextually targeted campaigns

Campaign Performance

99.5% accuracy in brand safety classification

Brand Safety

50K+ assets enriched per hour

Processing Speed

Frequently Asked Questions

Related Resources

More agency-data Use Cases

Creative Asset Intelligence for Media Planning

For agencies managing 100K+ creative assets. Unlock insights from creative performance data. 25% improvement in media efficiency.

Ready to Implement This Use Case?

Our team can help you get started with Clean Room Data Activation with IAB Taxonomy Enrichment in your organization.

Schedule a Demo Read the Docs