Engagement loop
Nodes that have been refreshed and proposals that the daily reviewers have emitted in the last 30 days. Sign in to flag any as helpful, incorrect, or needing more depth — the next morning the reviewer routines factor your signal into their scoring.
source.hipaajournal-com.hipaa-privacy-rule-2026
HIPAA Journal comprehensive overview of the HIPAA Privacy Rule, updated for 2026, covering PHI definition, permitted uses and disclosures, marketing restrictions, and written authorization requirements.
source.emarketer-com.content-faq-on-customer-data-platforms-composable-architectures-2026
eMarketer FAQ article published February 27, 2026 covering composable CDP architecture layers, team requirements, mid-market adoption barriers, AI integration patterns, and CDP market sizing.
archetype.salesforce-ecosystem-enterprise-evaluating-cdp
An enterprise with Salesforce CRM and Salesforce Data 360 as the incumbent CDP layer, now evaluating whether Data 360's segmentation and Marketing Cloud Next's Agentforce orchestration are sufficient for their use cases, or whether composable augmentation (Hightouch or Fivetran Activations over an existing CDW) adds meaningful value — particularly for CDW-native ML models, cross-cloud data, or segment volumes that require warehouse-native computation.
archetype.marketing-led-mid-market-composable
A mid-market organization ($25M–$500M revenue, 50K–1M customer profiles) where Marketing (CMO or VP Marketing) is RACI-Accountable for the CDP evaluation and operations. Evaluating composable or marketing-accessible CDP tooling without the budget for enterprise packaged platforms (AEP, Salesforce Data 360) and without deep data engineering staffing. The evaluation is marketing-led — not IT/CDO-driven. Primary candidate paths: BlueConic (marketer-operated Growth Plays, no-SQL), GrowthLoop (agentic no-code CDW activation if CDW exists), mParticle (mobile-first B2C apps). Distinguishing from enterprise archetypes by budget ceiling (~$400K/year), marketing Accountable (not CTO or CDO), and mid-market scale where composable tooling's operational simplicity outweighs its flexibility cost.
recommendation.salesforce-ecosystem-cdp-deepen-or-composable-augment
For an enterprise with Salesforce CRM and Salesforce Data 360 (incumbent or planned) as the CDP layer, the evaluation is not "Salesforce vs. composable" but "where does the canonical customer profile live and which activation paths need to be served?" Path 1 (Deepen) commits to Data 360 + Marketing Cloud Next + Agentforce as the unified stack. Path 2 (Augment) keeps Data 360 as the Salesforce-ecosystem profile layer while adding a CDW-native composable activation layer (Hightouch or Fivetran Activations) for CDW-held ML models and non-Marketing-Cloud channels. The choice is driven by CDW data access requirements, ML model location, and channel breadth — not by cost alone.
constraint.hipaa-phi-cdp-healthcare
Under the Health Insurance Portability and Accountability Act (HIPAA), covered entities (providers, health plans, clearinghouses) and their Business Associates may not use Protected Health Information (PHI) for marketing purposes without explicit written authorization from the individual, with narrow exceptions. CDPs operated by or for covered entities are Business Associates; vendor BAAs are required. The standard architectural mitigation is HIPAA de-identification — removing the 18 Safe Harbor identifiers — before PHI-derived data enters marketing pipelines or activation destinations.
constraint.aep-destinations-per-sandbox-100
Adobe Experience Platform enforces a hard limit of 100 destination instances per sandbox. System-enforced — AEP blocks creation of the 101st destination. Organizations building multi-channel activation architectures with AEP Real-Time CDP must account for this ceiling when designing their destination catalog. Multi-sandbox strategies become architecturally necessary for enterprise organizations whose activation surface exceeds 100 distinct endpoint connections (e.g., regional separation, business unit isolation, channel diversification across email, paid-media, CRM, streaming, analytics, and personalization destinations).
constraint.aep-batch-destination-per-destination-audience-250
Adobe Experience Platform performance guardrail recommending a maximum of 250 audiences per destination instance for streaming and file-based (batch) activation destinations. Not system-enforced — AEP does not hard-block at 250 — but performance degradation is expected above this ceiling. Distinct from constraint.aep-activation-external-audience-limit-20, which is a system-enforced hard limit specifically for external audiences (Federated Audience Composition, custom upload, Audience Composition). Adjacent performance guardrails from the same source: 50 simultaneous activations per destination, 50 mapped attributes per destination, 100 batch audiences per hour.
concept.real-time-decisioning
The class of decisioning systems that select an activation in response to a live signal. Almost always misused without specifying a Signal-to-Activation latency tier.
source.hightouch-com.docs-syncs-lightning-sync-engine-2026
Hightouch Lightning sync engine performs change data capture (CDC) directly in the warehouse, up to 100x faster than the Basic engine for large datasets (100k+ rows). Required for Journeys, Warehouse Sync Logs, Match Booster, Identity Resolution, and Audience Snapshots.
source.hightouch-com.docs-events-overview-2026
Hightouch Events collects customer behavior from websites, mobile apps, and backend systems, stores it in the warehouse, and streams events to downstream destinations in real-time. Supports browser/mobile SDKs, server-side SDKs, HTTP API, Kafka, Pub/Sub, and webhook inbound sources.
source.hightouch-com.docs-events-event-streaming-2026
Hightouch Event Streaming sends events collected by Hightouch directly to destinations in real-time without going through the warehouse. Enables latency-sensitive workflows (message triggering, consent updates, real-time personalization) within seconds. Any catalog destination supporting event forwarding can be used.
source.hightouch-com.docs-customer-studio-journeys-2026
Hightouch Journeys enables visual, tile-based multi-step customer lifecycle flows including welcome sequences, winback campaigns, and nurture tracks. Triggers syncs across 300+ destination integrations via Send-to-Destination tiles; exit criteria remove users when key conditions change.
source.docs-tealium-com.server-side-connectors-about-2026
Tealium documentation for EventStream connectors: real-time API integrations that transmit data between Tealium and external vendors. Connectors are triggered by event feeds or AudienceStream audience membership changes and support automatic retries and frequency caps.
source.docs-snowflake-com.en-release-notes-2026-other-2026-04-20-kafka-connector-v4-ga-2026
Snowflake Kafka Connector v4.0 reached GA on April 20, 2026. Ground-up rewrite on Snowpipe Streaming architecture delivering up to 10 GB/s throughput with 5-10s latency, exactly-once delivery, flat throughput-based pricing, and server-side schema evolution. Replaces v3 Classic.
source.docs-snowflake-com.en-developer-guide-snowflake-ml-model-registry-model-explainability-2026
Snowflake documentation for Model Explainability: uses SHAP library to compute Shapley values attributing model predictions to input features. Supports XGBoost, CatBoost, LightGBM, Scikit-learn, and Snowpark ML classes. Currently in preview release (Snowpark ML 1.6.2+).
vendor.hightouch
Composable CDP and reverse ETL platform. One of four Leaders in the 2026 Gartner Magic Quadrant for CDPs — the first warehouse-native CDP vendor to reach Leader status. Activates data from Snowflake, BigQuery, Databricks, and Redshift to 400+ destinations across 40+ source connectors without requiring a separate data copy or proprietary profile store.
source.siliconangle-com.growthloop-real-time-causal-decisioning-2026
Independent tech media analysis (SiliconANGLE, Apr. 15, 2026) of GrowthLoop's Composable AI Decisioning platform launch. Key: causal measurement approach (multi-armed bandit algorithms, not correlative), runs natively on BigQuery/Databricks/Snowflake/Redshift without data copying, pub/sub and Kafka queues for continuous measurement, Agentic Context Graph for causal history, and same-session message adjustment capability. Author Paul Gillin quotes former CEO Chris O'Neill positioning the shift from linear campaign models to iterative learning systems.
source.hightouch-com.docs-2025
Official Hightouch product documentation covering the full platform: Composable CDP (Customer Studio, audience building, identity resolution, Match Booster, reverse ETL, real-time and batch event triggering) and AI Decisioning (AID — AI agents that select message, channel, and timing per user). 40+ data source connectors and 400+ destination integrations. Last updated October 3, 2025.
org-dim.raci-role.informed
Stakeholders kept updated on CDP decisions and outcomes but not involved in the decision process. For CDP decisions, Informed roles typically include downstream marketing execution teams (who receive audience exports), business-unit leaders (who need to understand activation scope), and compliance officers (who receive audit-trail reports). Recommendations for this role are outcome-focused, non-technical, and packaged for executive or compliance review.
org-dim.raci-role.consulted
Subject-matter experts whose input is sought but who do not work directly on implementation. For CDP decisions, Consulted roles typically include IT Security (data governance and access controls), Legal/Privacy (GDPR, CCPA, HIPAA compliance), Finance (TCO approval and budget sign-off), and the data science or ML team (AI/ML model portability and CDW architecture requirements). Recommendations for this role must address the specific concerns of each function — security posture, regulatory compliance triggers, cost modeling, and model integration patterns.
org-dim.operational-profile
Dimension describing an organization's operational capability characteristics: the nature and scale of high-volume customer-facing operations — contact center, field sales, physical retail. Orthogonal to org-chart structure and industry; governs which specialized CDP vendor and use-case paths are applicable.
vendor.growthloop
Agentic composable CDP running natively on BigQuery, Snowflake, Databricks, and Redshift with no data ingestion or vendor-managed profile store. Three AI agents: Audience Agent (auto-generates audience segments from warehouse data), Journey Agent (builds and launches personalized campaigns across channels), and Insights Agent (real-time performance analysis). April 2026: launched Composable AI Decisioning with Decisioning Node (real-time channel/offer allocation), Always-On Lift Measurement (causal/multi-armed-bandit — not correlative attribution, independently confirmed by SiliconANGLE), Agentic Context Graph (causal knowledge base), and Universal Journeys (cross-channel journey engine). Enterprise customers include Costco, Albertsons, Ford, Indeed, Priceline, and Allegro. Allegro case study: 2x ROAS, 60%+ GMV increase, 4x CTR on Meta. TJC financial partnership January 2026. G2 Momentum Leader. No confirmed 2026 Gartner MQ or Forrester Wave placement.
constraint.ccpa-data-subject-rights-2026
The California Consumer Privacy Act (CCPA) as amended by the California Privacy Rights Act (CPRA) grants California residents six data subject rights: Right to Know, Right to Delete, Right to Opt-Out of Sale or Sharing, Right to Correct, Right to Limit Use of Sensitive Personal Information, and Right to Non-Discrimination. Regulations effective January 1, 2026 added Automated Decision-Making Technology (ADMT) rules: businesses using ADMT for significant decisions must provide pre-use notices, honor opt-out rights, and respond to access requests describing the logic and likely outcomes of automated processes. Risk assessment and cybersecurity audit requirements also took effect January 1, 2026. Enforced by the California Privacy Protection Agency (CPPA) — the first dedicated data privacy enforcement agency in the US.
source.snowflake-com.en-blog-new-regions-egress-cost-optimizer-2026
Snowflake blog post announcing the Egress Cost Optimizer (ECO), which enables organizations to expand to new regions and share data cross-cloud with up to 96% savings compared to standard egress pricing. ECO rate: $16.896/TB-month.
source.hightouch-com.solutions-healthcare-2026
Hightouch's healthcare-specific product page covering HIPAA compliance, BAA availability, and how the warehouse-native (no-storage) architecture simplifies PHI governance compared to traditional CDPs that refuse to sign BAAs.
source.hightouch-com.platform-composable-cdp-2026
Hightouch's own product page for its composable CDP platform. Describes warehouse-native architecture, 300+ destination integrations, AI-powered audience builder, no-MTU pricing, and HIPAA / SOC 2 compliance posture.
source.getcensus-com.reverse-etl-2026
Census's product page for its reverse ETL platform. 200+ connectors, 20+ warehouse integrations (Snowflake, Databricks, BigQuery, Redshift), real-time Live Syncs for BigQuery, no data storage model, and Census Embedded for SaaS builders.
source.gdpr-info-eu.art-7-gdpr-2016
Authoritative mirror of GDPR Article 7 (conditions for consent), covering the four requirements for valid consent — freely given, specific, informed, and unambiguous — plus withdrawal rights and the burden-of-proof rule on controllers.
source.docs-snowflake-com.en-user-guide-cost-exploring-data-transfer-2026
Snowflake's official documentation on how to explore and monitor data transfer (egress) costs, complementing the "understanding" guide. Covers tooling for cost attribution, cross-region/cross-cloud transfer reporting, and the Egress Cost Optimizer (ECO).
source.docs-databricks-com.aws-en-oltp-projects-reverse-etl-2026
Databricks official documentation for reverse ETL using Lakebase on AWS — covering synced tables, sub-10ms query latency, 1,000 concurrent connections, and the workflow for activating gold-layer lakehouse data into operational apps and CRM systems.
source.cloud-google-com.blog-products-data-analytics-hightouch-composable-cdp-built-on-bigquery-2026
Google Cloud blog post describing how Hightouch's composable CDP integrates with BigQuery as the warehouse-of-record for customer data, enabling data teams to activate BigQuery data to 300+ destinations without data duplication.
source.cdp-com.basics-cdp-vendors-2026
CDP Institute's 2026 comparison of nine leading CDP vendors, covering packaged and composable options. Identifies Hightouch as a new Gartner MQ Leader and describes the 2026 bifurcation between platformization CDPs (Salesforce, Adobe, Oracle) and agentification CDPs (Hightouch, composable players).
source.cdp-com.articles-cdp-for-healthcare-2026
CDP Institute article on deploying CDPs in healthcare. Key requirement: vendor must sign a BAA. Identifies Hightouch, Ours Privacy, Freshpaint, Piwik PRO, and Treasure Data as vendors with documented HIPAA compliance postures.
source.inma-org.blogs-digital-subscriptions-4-data-platforms-media-ad-teams-2023
INMA (International News Media Association) article covering four data platforms (CDPs and reader networks) helping digital publishers monetize audiences in the post-third-party-cookie era. Addresses how publishers evolve advertising strategies and improve yields using first-party data platforms. Snippet-only — full content unavailable from this environment.
source.docs-snowflake-com.en-user-guide-data-load-overview-2026
Official Snowflake documentation overview of all data loading methods: COPY INTO bulk loading (PUT + COPY), Snowpipe micro-batch continuous loading (serverless), and Snowpipe Streaming (sub-minute row-level ingestion). Covers staging locations (S3, GCS, Azure, internal stages), INFER_SCHEMA schema detection, and DML error logging for fault-tolerant loads.
operational-task.ingest-batch-data-via-source-connector
Upload a structured file (CSV, JSON, Parquet) to a cloud staging area or use a UI workflow to map source columns to XDM fields and load records into a platform dataset as a completed batch.
archetype.media-publisher-first-party-data-cdp-evaluator
News and digital media publishers evaluating CDPs to build first-party audience data infrastructure for subscription lifecycle management, consent-aware programmatic ad activation, and paywall optimization. Characterized by high anonymous-traffic ratios (80%+), IAB TCF 2.2 compliance requirements, and the need to bridge editorial engagement data with advertiser-facing audience segments.
source.growthloop-com.product-composable-ai-decisioning-2026
Official GrowthLoop product page for Composable AI Decisioning. Named features: AI Studio (marketer command center), Decisioning Node (in Universal Journeys), Always-On Measurement, Agentic Context Graph, Data Manager, Universal Journeys. Supports BigQuery, Snowflake, Databricks, and Amazon Redshift. Allegro case study: 2x ROAS, 60%+ GMV increase, 4x CTR on Meta. No publication date on page.
source.cmswire-com.customer-data-platforms-why-readiness-cdp-success-2026
CMSWire practitioner article (April 2026) arguing CDP success is determined by organizational readiness rather than technology selection. Reports customer data duplication rates of 10–30%, governance challenges nearly doubled year-over-year, and teams spending more time on data preparation than on activation. Cites Amperity CEO Tony Owens. Identifies lean cross-functional team capacity as the primary barrier to CDP value realization.
org-dim.operational-profile.engineering-led-cdp
An organization where data engineering or IT owns and operates the CDP activation layer — defining audiences in SQL or dbt, configuring activation via reverse-ETL jobs, and managing segment pipelines programmatically. Marketing consumes prebuilt segments and reports rather than directly configuring the activation layer. This operational profile is the primary selection signal for SQL-first composable CDPs (Hightouch, Fivetran Activations) and data-engineering-accessible packaged CDPs (Segment, mParticle, Adobe Experience Platform) over marketer-self-service tools.
org-dim.industry.media-publishing
Organizations in digital media (streaming, broadcast, digital news) and publishing (subscription content, magazine, audio/podcast) with CDP use cases centered on subscriber lifecycle management, content personalization, consent-aware audience monetization for advertisers, and reader/viewer churn prediction.
source.news-adobe-com.news-2026-04-adobe-unveils-cx-enterprise-coworker-2026
Adobe announced two new AI capabilities on April 20, 2026: (1) CX Enterprise Coworker — an agentic AI solution that monitors signals, recommends next-best actions, and executes experiences across channels in real time based on defined goals, built on MCP and Agent2Agent open standards with integrations to AWS, Anthropic, Google Cloud, Microsoft, OpenAI, and NVIDIA; (2) Adobe Engagement Intelligence — a decisioning engine optimized for customer lifetime value that powers personalization at scale. Both integrate with Adobe Real-Time CDP to unify structured and unstructured customer data. Human oversight is maintained throughout the CX Enterprise Coworker execution model.
source.martechsquare-substack-com.p-the-gartner-cdp-quadrant-is-out-but-2026
Independent analyst commentary on Gartner Magic Quadrant for CDPs 2026: Salesforce and Hightouch confirmed as Leaders; 'decisioning' appears 35x and 'orchestration' 39x in the MQ report, displacing 'single customer view' framing; thesis is that the CDP is transitioning from data aggregation toward decision-ready context for AI agents.
source.fastly-com.products-edge-compute-2026
Fastly Compute: serverless edge computing platform running WebAssembly at CDN PoPs globally with no cold starts; supports JavaScript, Rust, Go, and C++; sub-microsecond execution; integrates Key Value Store and Fanout for real-time personalization, A/B testing, and API acceleration at the edge.
source.segment-com.docs-unify-profile-api-2025
Segment documentation for the Profile API — programmatic access to unified user- and account-level profiles including external_ids, traits, and event history. Supports up to 200 traits per query via ?limit parameter. snippet-only — full fetch unavailable.
source.segment-com.docs-protocols-tracking-plan-create-2025
Segment documentation for creating Protocols Tracking Plans — JSON Schema-based event schema definition and governance for Segment sources. Supports adding events, properties, data types, and permitted values. snippet-only — full fetch unavailable from this environment.
source.segment-com.docs-protocols-enforce-schema-violations-2025
Segment documentation for Protocols schema enforcement — how violations are detected against Tracking Plan definitions, violation streams for monitoring non-conforming events, and blocking behavior. snippet-only — full fetch unavailable from this environment.
source.segment-com.docs-engage-audiences-sql-traits-2025
Segment documentation for SQL Traits — warehouse-backed audience attributes imported into Unify/Engage via custom SQL queries. Note: SQL Traits entered End of Sale as of March 31, 2024; existing customers retain access but no new customer availability. snippet-only.
source.segment-com.docs-engage-audiences-computed-traits-2025
Segment documentation for Computed Traits — auto-calculated user attributes (event counters, percentiles, session windows) for audience building and destination personalization. Sent to Event destinations via Identify calls. snippet-only — full fetch unavailable.
source.segment-com.docs-engage-audiences-2025
Segment documentation for Engage Audiences — real-time audience builder using identity-resolved profiles, computed traits, SQL traits, and audience memberships. snippet-only — full fetch unavailable from this environment.
source.law-cornell-edu.uscode-text-47-227-2026
The Telephone Consumer Protection Act statutory text hosted on Cornell LII, covering restrictions on autodialed calls and texts, prior express written consent requirements for marketing SMS, opt-out/revocation provisions, and penalty structure.
source.iabeurope-eu.tcf-2-0-2019
IAB Europe's Transparency and Consent Framework (TCF) standardizes how publishers, CMPs, and ad tech vendors communicate consent and legitimate interest signals across the programmatic advertising ecosystem. The framework defines roles for Publishers, Consent Management Platforms (CMPs), Vendors, and Advertisers and Agencies. It includes TCF Supporting Resources, TCF Governance, TCF Compliance Programmes, and CMP and Vendor Notification mechanisms. TCF v2.0 published 2019; updated to v2.2 in 2023. IAB Europe page copyright 2026.
source.hightouch-com.docs-syncs-overview-2024
Official Hightouch documentation for Syncs — the mechanism delivering model query results to destinations on configurable schedules (interval, cron, dbt Cloud or Fivetran integration). Tracks changed data between runs; sync modes include upsert, insert, update.
source.hightouch-com.docs-destinations-overview-2025
Official Hightouch documentation for the Destinations catalog — 400+ integrations (CRMs, marketing platforms, ad networks, analytics) activated via reverse ETL pipeline using TLS 1.2+ secured APIs.
source.hightouch-com.docs-audiences-overview-2026
Official Hightouch documentation for Customer Studio — visual workspace enabling marketers to build audiences from warehouse data, orchestrate multi-channel campaigns via journeys and experiments, and sync real-time to ad platforms, CRMs, and messaging tools without SQL.
source.gdpr-info-eu.art-17-gdpr-2016
Art. 17 GDPR requires data controllers to erase personal data without undue delay when any of six grounds apply: data no longer needed for original purpose; consent withdrawn with no alternative legal basis; data subject objects under Art. 21 and no overriding legitimate interest exists; processing was unlawful; a legal obligation requires deletion; or data was collected from a child under Art. 8. Controllers who made data public must take reasonable technical steps to notify other processors of the erasure request. Five exceptions allow continued processing: freedom of expression, legal compliance obligations, public health, archiving/research/statistics, and establishing or defending legal claims. CDP architectures must implement automated consent-withdrawal triggers, cross-vendor deletion APIs, audit trails documenting erasure attempts, and exception-mapping logic before deletion processing.
source.gartner-com.magic-quadrant-customer-data-platforms-2026
Gartner Magic Quadrant for Customer Data Platforms, published 26 January 2026 (ID G00836173). Leader: Salesforce. Challenger: Treasure Data. Visionaries: Adobe, Oracle. Niche Players: Amperity, BlueConic, Tealium. Authors: Lizzy Foo Kune, Rachel Dooley, and 3 more. Core thesis: the CDP market is bifurcating into platformization and agentification — buyers should assess needs for orchestration vs autonomy, prioritize warehouse-native composable architectures, and evaluate cost-to-value alongside agentic AI investment.
source.fivetran-com.docs-activations-2026
Fivetran Activations is the reverse ETL product acquired from Census in May 2025 and integrated into the Fivetran platform. Enables no-code, automated reverse ETL pipelines that sync data from a source data warehouse or data lake to destinations such as CRMs, marketing platforms, and support tools. Fivetran's combined connector library (ingestion + activation) exceeds 700. Billing uses a MAR-based consumption model with separate meters for connection MAR, activation MAR, and transformation MMR. One of the first platforms to cover both ELT ingestion and reverse ETL activation in a single product.
source.docs-tealium-com.iq-tag-management-getting-started-2026
Official Tealium documentation for iQ Tag Management getting started — covers Profile setup, data layer configuration, and tag container deployment using the Universal Tag architecture.
source.docs-snowflake-com.en-user-guide-ui-snowsight-worksheets-2026
Official Snowflake documentation for Snowsight worksheets — integrated SQL and Python execution environment with 24-hour result caching, versioning, and sharing. Legacy worksheets deprecate June 22, 2026 in favor of Workspaces.
source.docs-snowflake-com.en-user-guide-tasks-intro-2026
Official Snowflake documentation introducing Tasks for automated pipeline scheduling — supports fixed intervals, CRON expressions with timezone, and stream-data event triggers.
source.docs-snowflake-com.en-user-guide-snowflake-cortex-cortex-analyst-semantic-model-spec-2026
Official Snowflake documentation for semantic views — schema-level objects that define business metrics, dimensions, and facts as a semantic layer, enabling consistent metric definitions and enhancing Cortex Analyst natural language query accuracy.
source.docs-snowflake-com.en-user-guide-querying-semistructured-2026
Official Snowflake documentation for querying VARIANT columns — dot/bracket notation for traversal, FLATTEN/LATERAL FLATTEN for array expansion, and higher-order functions (FILTER, REDUCE, TRANSFORM) for complex nested data like GA4 nested JSON.
source.docs-snowflake-com.en-user-guide-querying-joins-2026
Official Snowflake documentation for SQL join types used to combine rows across tables — inner, outer, natural joins — key for identity stitching and multi-dataset merge patterns.
source.docs-snowflake-com.en-user-guide-ml-functions-classification-2026
Official Snowflake documentation for the ML Classification function (GA) — gradient-boosting binary and multi-class classifier for propensity scoring, churn prediction, and fraud detection. Operates as a schema-level object with train, predict, and evaluate APIs.
source.docs-snowflake-com.en-user-guide-kafka-connector-overview-2026
Official Snowflake documentation for the Kafka Connector v3 (Classic) — buffers Kafka topic messages, stages files, and ingests via Snowpipe. Note: v3 scheduled for deprecation mid-2026; Snowflake recommends Kafka Connector v4 for new implementations.
source.docs-snowflake-com.en-user-guide-kafka-connector-install-2026
Official Snowflake documentation for Kafka Connector installation — JAR file (Confluent or Apache), RSA key-pair authentication (passwords not supported), and JSON/properties configuration. Supported versions: Apache Kafka 2.8.2/3.7.2, Confluent 6.2.15/7.8.2.
source.docs-snowflake-com.en-user-guide-dynamic-tables-intro-2026
Official Snowflake documentation for Dynamic Tables — self-refreshing materialized tables that auto-update based on a query definition and TARGET_LAG freshness parameter, replacing manual scheduling.
source.docs-snowflake-com.en-user-guide-data-load-snowpipe-streaming-overview-2026
Official Snowflake documentation for Snowpipe Streaming: sub-second row-level ingestion with up to 10 GB/s throughput, as-low-as 5-second end-to-end latency, and exactly-once delivery via Java, Python, Node.js SDKs or REST API.
source.docs-snowflake-com.en-user-guide-data-load-snowpipe-intro-2026
Official Snowflake documentation for Snowpipe continuous micro-batch data loading from cloud storage using event notifications or REST API with Snowflake-managed compute.
source.docs-snowflake-com.en-sql-reference-sql-create-table-2026
Official Snowflake SQL reference for the CREATE TABLE command: column definitions, data types, constraints, and variants including CTAS, CLONE, and LIKE for data warehouse table provisioning.
source.docs-snowflake-com.en-sql-reference-sql-create-dynamic-table-2026
Official Snowflake SQL reference for CREATE DYNAMIC TABLE — key parameters include TARGET_LAG (freshness window, minimum 60s), WAREHOUSE, REFRESH_MODE (AUTO/FULL/INCREMENTAL), and DOWNSTREAM mode for refresh-on-demand.
source.docs-snowflake-com.en-sql-reference-sql-copy-into-table-2026
Official Snowflake reference for the COPY INTO command for bulk loading staged data files into tables from internal/external stages or cloud storage, with support for multiple file formats and data transformation during ingestion.
source.docs-snowflake-com.en-sql-reference-functions-parse-json-2026
Official Snowflake reference for PARSE_JSON — converts a JSON-formatted string to a VARIANT value, enabling semi-structured data access via dot notation. Returns NULL for empty/whitespace input rather than raising an error.
source.docs-snowflake-com.en-guides-overview-ml-functions-2026
Official Snowflake documentation for ML Functions — automated prediction and pattern detection without ML expertise. Supported types include Forecasting, Anomaly Detection, Classification, and Top Insights with separate storage and compute costs.
source.docs-mparticle-com.docs-2026
mParticle product documentation hub covering IDSync identity resolution, Data Master data quality enforcement, Audience segmentation, Custom Rules, User Activity View, Privacy compliance tools (GDPR/CCPA), and real-time APIs (Events API, Profile API, Firehose). SDKs available for Android, iOS, and Web. Cortex predictive AI is referenced as a distinct product offering. Profile API described as a real-time API to drive user personalization. Page footer timestamp: 2026-05-12.
source.docs-getdbt-com.docs-deploy-deploy-jobs-2026
Official dbt documentation for deploy jobs — scheduled or event-triggered dbt command execution with cron syntax (including L operator for last-day-of-month), interval scheduling, and job chaining. All times in UTC.
source.docs-getdbt-com.docs-build-seeds-2026
Official dbt documentation for seeds — CSV files in the seeds/ directory loaded into the data warehouse via `dbt seed`. Ideal for static reference data (country codes, test exclusion lists); referenced downstream via ref() function.
source.docs-getdbt-com.docs-build-models-2026
Official dbt documentation for models — SQL SELECT statements saved as .sql files that transform warehouse data in-place via `dbt run`. Models reference each other using ref() to build complex nested transformation DAGs for ELT workflows.
source.docs-getdbt-com.docs-build-metrics-overview-2026
Official dbt documentation for MetricFlow metrics — five metric types (simple, cumulative, derived, ratio, conversion) defined in YAML. The dbt Semantic Layer centralizes metric definitions queryable across Tableau, Excel, Hex, and other BI tools.
source.docs-getdbt-com.docs-build-incremental-models-2026
Official dbt documentation for incremental models — transforms only new/updated rows on subsequent runs, reducing compute cost versus full rebuilds. Supports on_schema_change config (sync_all_columns) and unique key definitions for upsert semantics.
source.docs-confluent-io.platform-current-connect-transforms-overview-2026
Official Confluent documentation for Kafka Connect SMTs — middleware transforms modifying records in-flight. 25+ built-in transforms including ExtractField, ValueToKey, ReplaceField, Cast, EventRouter. SMTs operate between source production and Kafka write, or Kafka read and sink delivery.
source.cppa-ca-gov.regulations-consumer-privacy-act-2023
California Privacy Protection Agency official regulatory text for the California Consumer Privacy Act as amended by CPRA. Final regulations approved March 29, 2023, operationalizing consumer rights introduced by CPRA: right to know, right to delete, right to opt-out of sale, and right to correct inaccurate data. CPPA is the enforcement authority. Regulations consolidate and clarify legal requirements for businesses handling consumer data, including CDP marketing pipelines activating California resident profiles.
source.adexchanger-com.data-privacy-roundup-freshpaint-recovering-cdp-healthcare-2023
Independent trade media profile (AdExchanger, Nov. 7, 2023) on Freshpaint's pivot from general CDP to healthcare privacy platform. Key capabilities documented: HIPAA identifier blocking, user ID deidentification with anonymous masking, video embedding without IP sharing, and third-party tracker scanning. Safety-by-default architecture restricts sensitive health data automatically until marketers explicitly opt-in per destination. Pivoted after FTC's 2022 crackdown on digital health companies sharing patient data. Not a vendor source; AdExchanger is independent trade editorial.
operational-task.validate-unified-profile-via-api
Use the Platform Profile API with a namespace and identity value to retrieve a merged profile record programmatically, confirming that attributes, events, and segment memberships from all source datasets are correctly unified and accessible to external applications.
operational-task.validate-real-time-audience-segment-delivery
Trigger a segment qualification event and confirm that the correct payload — with appropriate membership status — arrives at the configured downstream streaming endpoint within the expected latency window.
operational-task.validate-and-monitor-streaming-ingestion-pipeline
Produce a test event through a streaming pipeline, confirm end-to-end delivery and schema compliance at the ingestion endpoint, and verify the event appears on the target profile or dataset within the expected latency window.
operational-task.schedule-and-export-query-results-as-datasets
Convert an ad-hoc analytical query into a scheduled, persisted CTAS query that writes results to a named derived dataset, then connect a BI tool or downstream system to that dataset for repeatable reporting or activation.
operational-task.run-ad-hoc-sql-queries-against-datasets
Connect to the platform's SQL-accessible query layer, enumerate available dataset tables, write standard SQL using XDM dot-notation field paths to retrieve and aggregate event or profile data, and interpret the results for analytical decision-making.
operational-task.route-server-side-events-to-third-party-platforms
Extend a server-side forwarding property to deliver filtered event data from the edge network to multiple external destinations such as analytics services, cloud functions, and advertising platforms.
operational-task.route-journey-actions-to-external-channels
Define a custom action endpoint that allows a customer journey to deliver personalized messages or trigger workflows in external systems such as SMS providers, Slack, or custom webhooks.
operational-task.integrate-external-data-sources-into-journey
Configure a real-time REST API connection that enriches a customer journey with contextual data — such as weather, inventory, or loyalty balance — fetched at the moment of journey execution.
operational-task.ingest-streaming-events-via-sdk-or-api
Configure an XDM-mapped streaming endpoint or SDK launch rule to deliver behavioral events to the platform in real time, verifying arrival through dataset monitoring.
operational-task.design-and-publish-multi-step-customer-journey
Compose a stateful, event-triggered sequence of personalized actions — messages, waits, conditions — that a customer traverses in real time based on their behavior.
operational-task.define-xdm-schema-for-event-capture
Identify which field groups are required for a given event type, add the mandatory SDK mixin, configure identity fields, enable the schema for the real-time profile store, and validate payload shape using a browser debugger.
operational-task.define-calculated-metrics-and-attribution-models-for-journey-reporting
Compose reusable calculated metrics from base XDM event fields, select attribution models and lookback windows, and apply them inside workspace analyses to measure the contribution of each journey touchpoint.
operational-task.connect-bigquery-data-source-and-ingest-ga4-event-data
Authenticate a cloud data warehouse via OAuth, map its event table fields to a target schema, configure a delta-based ingestion schedule, and launch the dataflow to land GA4 behavioral data in the platform.
operational-task.configure-xdm-schemas-and-profile-datasets
Design XDM schemas for both profile and event data by selecting appropriate field groups, defining primary and secondary identity fields with namespace assignments, enabling schemas for profile ingestion, and creating linked datasets that hydrate the real-time customer profile.
operational-task.configure-tag-management-and-edge-routing
Set up a tag management property with a client-side SDK extension, define data elements and rules, wire them to an edge routing configuration, and publish the library to a deployment environment.
operational-task.configure-server-side-event-forwarding-rules
Create a server-side data collection property with data elements and rules that intercept edge network traffic and forward selected XDM event fields to external endpoints without client-side code execution.
operational-task.configure-propensity-scoring-model
Set up a machine-learning model that predicts the likelihood of a customer performing a target action, using historical behavioral event data.
operational-task.configure-offer-decisioning-catalog-and-eligibility-rules
Build a catalog of personalized and fallback offers, define profile-based eligibility rules, assemble those offers into decision scopes, and publish the decision for real-time delivery.
operational-task.configure-kafka-source-connector-for-streaming-ingestion
Install and configure a Kafka Connect sink connector that continuously reads messages from Kafka topics and delivers them as XDM experience events to a streaming ingestion endpoint in real time.
operational-task.configure-journey-entry-events-and-guardrails
Define the XDM event schema and payload that triggers a customer journey, and set the frequency-capping, re-entry, and suppression rules that govern journey execution safety.
operational-task.configure-event-triggered-journey-entry-and-real-time-action-routing
Define streaming entry events with schema-matched payloads, assemble a journey canvas that routes profiles through conditional actions, and publish the journey for real-time execution.
operational-task.configure-deterministic-identity-stitching
Define identity namespaces, designate primary and secondary identity fields on XDM schemas, and validate that the identity graph correctly links profile fragments from multiple datasets and channels into a single unified profile.
operational-task.build-rule-based-audience-segment
Define a reusable audience by composing attribute and behavioral rules against unified profile data, then publish it for downstream activation.
operational-task.build-cross-channel-analysis-workspace
Stitch behavioral data from web, voice, and call-center channels into a unified person-level connection, define a data view with dimensions and metrics, and surface journey funnels and flow visualizations in an analysis workspace.
operational-task.build-and-publish-audience-segment
Author a segment definition combining profile attributes and behavioral event criteria, preview the estimated audience size, save the definition for reuse, and optionally export segment membership to a dataset or downstream activation destination.
operational-task.analyze-ingested-ga4-data-using-platform-query-tools
Join a GA4-sourced event dataset with a loyalty profile dataset on a shared key, build an analysis workspace with attribution-aware freeform tables and funnel visualizations, and derive actionable conversion insights from the combined data.
operational-task.activate-audience-to-external-system
Ingest externally defined audiences into the CDP and create actionable segments from them, or export internally defined audiences to external platforms for bidirectional sharing.
source.prnewswire-com.news-releases-growthloop-composable-ai-decisioning-302742327-2026
GrowthLoop announced Composable AI Decisioning on April 15, 2026. The platform adds three capabilities: (1) Decisioning Node — real-time channel and offer allocation within marketer-defined journeys, automatically adjusting toward options that drive stronger outcomes; (2) Always-On Lift Measurement — continuous incremental impact tracking across marketing activities; (3) Agentic Context Graph — a causal knowledge base that learns which marketing decisions actually improve key outcomes like revenue or lifetime value, compounding over time. Unlike correlation-based AI decisioning, it uses causal measurement. Runs natively on BigQuery and Snowflake without copying or moving data. Named enterprise customers: Costco, Albertsons, Ford. Founded by former Google executives. G2 Momentum Leader.
source.martech-org.building-a-center-of-excellence-for-cdps-in-2024-2024
Independent MarTech editorial (February 22, 2024) on CDP center-of-excellence team composition. Key finding: IT organizations now own CDP decisions due to regulatory requirements (GDPR, CCPA), with CMO as influencer rather than primary owner. CDP platform choice directly determines team structure: decision-oriented CDPs (Blueshift, BlueConic, Simon, Lytics, ActionIQ) attract marketing-operational ownership; data/product platforms (Adobe, mParticle, Segment) require IT and engineering leadership with marketers as secondary users. Composable CDP model requires a product owner, customer insights team, and ML engineering professionals. Traditional CDP model requires marketing ops, data engineers, front-end engineers, and analysts. No headcount benchmarks provided. Note: published pre-2025 agentic AI wave.
source.globenewswire-com.news-release-3290373-tealium-ai-edge-2026
Tealium May 2026 product announcement covering four new capabilities: (1) AI at the Edge — high-performance native mobile SDK enabling on-device transformations and edge AI inference without app store releases; (2) AI Decisioning — runs real-time decisioning on live event streams (churn scores, product affinities) with IYOM (Invoke Your Own Model) flexibility; (3) Configuration Agent — MCP-powered agent bridging Claude, Gemini, and OpenAI to Tealium configuration via natural-language prompts; (4) AI Partner Ecosystem — bi-directional Amazon Bedrock connectors plus Pinecone (vector retrieval) and LangChain (agent orchestration) integrations.
source.businesswire-com.news-home-20260113029326-tjc-growthloop-2026
BusinessWire press release (January 13, 2026) announcing TJC's deepened financial partnership with GrowthLoop as enterprises modernize marketing with agentic AI and composable data infrastructure. GrowthLoop operates as an agentic composable CDP running natively on Snowflake, BigQuery, and Databricks with no data ingestion or vendor-managed profile store. Customers named include Costco, Albertsons, Ford, Indeed, Priceline, and Allegro.
vendor.adobe-experience-platform
Adobe's packaged Customer Data Platform. Strong identity stitching, profile store, and activation ecosystem; bounded by activation quotas and query-time caps that distinguish it from composable CDW-centric stacks. April 2026 capability expansion adds CX Enterprise Coworker (agentic AI for CX orchestration) and Engagement Intelligence (CLV-optimized decisioning) built on MCP + Agent2Agent open standards.
source.experienceleague-adobe-com.en-docs-experience-platform-query-guardrails-2026
Adobe's official documentation for Query Service guardrails in Adobe Experience Platform — concurrent session limits, batch query execution caps, and add-on entitlement tiers.
source.docs-snowflake-com.en-user-guide-cost-understanding-data-transfer-2026
Official Snowflake documentation covering data transfer and egress pricing — same-region vs cross-region costs, what triggers charges, and the Egress Cost Optimizer introduced April 2025.
source.databricks-com.blog-reverse-etl-lakebase-activate-your-lakehouse-data-operational-analytics-2026
Databricks blog announcing Lakebase — a Postgres database natively integrated with the Databricks lakehouse, enabling sub-10ms reverse ETL from Unity Catalog gold-layer tables into operational applications.
source.cxtoday-com.customer-analytics-intelligence-gartner-magic-quadrant-cdp-2026-2026
CX Today's analysis of the 2026 Gartner Magic Quadrant for Customer Data Platforms, covering vendor placements, new entrants, notable drops, and the shifting definition of CDPs as enterprise data-strategy infrastructure.
source.activeprospect-com.blog-tcpa-text-messages-2026
ActiveProspect's compliance guide on TCPA rules for SMS marketing in 2026 — covering the one-to-one consent rule effective January 26 2026, opt-out requirements, penalties, and FCC/state regulatory landscape.
tech-dim.dev-team.full-stack-services
Engineers who own the application services layer — APIs, orchestration services, webhook handlers, and glue code that binds CDP capabilities into the live product experience.
tech-dim.dev-team.frontend-mobile
The team responsible for native iOS and Android instrumentation, mobile SDK integration, push-credential management, and in-app experience delivery. Enables the mobile-app modality at Intra-page and Inter-page latency tiers.
tech-dim.dev-team.backend-email-delivery
The team responsible for ESP integration, transactional email pipelines, suppression-list synchronization, and deliverability monitoring. Required for any email-modality activation.
tech-dim.dev-team.backend-data-warehouse
The team responsible for warehouse administration, compute-credit governance, access control, query optimization, and cost attribution on the CDW. Distinct from data engineering — this team keeps the platform healthy so data engineering and analytics can use it effectively.
org-dim.industry.travel-hospitality
Airlines, hotels, rental cars, cruise lines, and OTAs. Dominant use cases involve loyalty program personalization, booking-flow abandonment recovery, ancillary upsell, and post-stay win-back.
org-dim.industry.financial-services
Banks, credit unions, investment firms, payment networks, and fintechs. High regulatory complexity (PCI-DSS, SOX, GLBA). Dominant use cases involve product cross-sell, onboarding sequence optimization, and fraud-signal-aware activation suppression.
capability.identity-resolution
The process of stitching disparate identifiers — cookie, device ID, email address, phone number, loyalty ID — into a single, persistent customer profile. The foundational prerequisite for any cross-modality or cross-session activation.
source.theaudiencers-com.wan-ifra-trends-outlook-2026-subscription-model-2026
Independent media-industry analysis (January 2026) synthesizing WAN-IFRA Trends Outlook 2026 findings on the state of digital subscriptions and audience monetization for news publishers. Key data: 58.5% of news organizations globally now offer digital subscriptions (72.9% in developed markets); digital circulation = 14% of publisher revenue (up from 9% in 2021). Describes five subscription strategies: bundling, dynamic paywalls, flexible retention, membership, and diversified acquisition channels.
source.martech360-com.tech-analytics-multi-touch-attribution-vs-media-mix-modeling-2026
Independent trade-press analysis (May 2026) arguing that treating MTA vs. MMM as a binary choice is obsolete; the practitioner norm is Unified Marketing Measurement running both models in parallel — MTA as the tactical microscope for day-to-day digital optimization, MMM as the strategic telescope for full-mix budget allocation. Covers cookie-deprecation impact on MTA and the shift to consent-driven, first-party-data-anchored attribution.
capability.multi-touch-attribution
Statistical modeling technique distributing conversion credit across marketing touchpoints encountered before a conversion event. Core methods: rules-based (first-touch, last-touch, linear, time-decay, position-based) and algorithmic (Markov chain, Shapley value). In composable CDP and CDW stacks, implemented as dbt pipelines over event-stream tables or via Snowflake Model Explainability for feature-level Shapley attribution within trained conversion-prediction models. In 2026, standalone MTA is a minority pattern; the practitioner norm is Unified Marketing Measurement (UMM) — MTA as one input in a hybrid system reconciled with Media Mix Modeling (MMM).
source.fastly-com.blog-fastly-ranked-leader-2026-forrester-wave-development-2026
Fastly named one of only two Leaders in The Forrester Wave: Edge Development Platforms Q1 2026; received perfect 5/5 scores across eight criteria including Innovation, Developer Experience, Performance, and Observability; the only evaluated vendor to receive above-average customer feedback from Forrester's survey.
org-dim.industry
The vertical the customer operates in. Drives use-case prioritization, regulatory regime, identity model, and acceptable activation cadence.
vendor.fastly
Edge cloud platform; commonly cited in CDP architectures for Inter-page edge decisioning where the activation needs to happen within milliseconds of the request.
capability.edge-decisioning
The capability to execute personalization or routing logic at CDN edge nodes (geographically distributed PoPs) rather than in a central data center, achieving sub-100ms signal-to-activation latency. Two implementation tiers: (1) rule-based edge compute — WebAssembly runtimes (vendor.fastly Compute@Edge, Cloudflare Workers, Lambda@Edge) for pre-computed segment flag lookups and A/B test assignments at sub-microsecond to sub-10ms speed; (2) edge AI inference (vendor.akamai Inference Cloud) for transformer-class ML model execution at CDN PoPs when decisioning logic cannot be pre-encoded as rules.
source.techtarget-com.searchbusinessanalytics-feature-how-to-structure-and-manage-a-data-science-team-2025
TechTarget practitioner guide covering nine core data science and ML team roles (data scientist, ML engineer, data engineer, data architect, data analyst, analytics translator, product owner, team manager, data governance lead) and three organizational structure models (centralized, decentralized, hybrid). Published August 5, 2025.
archetype.aep-heavy-enterprise-evaluating-composable
A large enterprise with AEP as the incumbent CDP, a growing data warehouse footprint, and escalating friction from AEP's export and query limits. Evaluating whether to extend, supplement, or replace AEP with composable alternatives.
constraint.aep-simultaneous-activation-per-destination-50
Adobe Experience Platform performance guardrail recommending a maximum of 50 simultaneous (concurrent) activation operations per destination instance. Not system-enforced — AEP will not hard-block above 50 — but performance degradation is expected above this concurrency threshold. Distinct from constraint.aep-batch-destination-per-destination-audience-250, which governs the total number of audiences mapped to a destination; this governs how many can be simultaneously activating (running export jobs) at one time.
constraint.aep-mapped-attributes-per-destination-50
Adobe Experience Platform performance guardrail recommending a maximum of 50 mapped profile attributes per destination instance. Not system-enforced — AEP does not hard-block at 50 mapped attributes — but performance degradation is expected above this threshold. Applies during the field-mapping step in the Destinations workflow: each profile attribute (first name, email, loyalty tier, propensity score, product affinity, etc.) mapped to a destination field counts against this limit. Highly enriched destination payloads (personalization platforms requiring 40–60 profile attributes for content rendering) approach this guardrail and require attribute prioritization.
constraint.aep-batch-file-max-records-5m
Adobe Experience Platform enforces a system limit of 5 million records per exported file for batch (file-based) destinations, including cloud-storage destinations (Amazon S3, Azure Blob, Google Cloud Storage, SFTP) and any file-based batch destination. Exports of audiences exceeding 5 million records must be split across multiple files or multiple export jobs. For enterprise and healthcare organizations with large activated segments, this limit dictates file-partitioning strategy and downstream ingestion tooling requirements at the receiving platform.
constraint.aep-batch-audiences-per-hour-100
Adobe Experience Platform performance guardrail recommending a maximum of 100 audiences per batch export hour across all batch (file-based) destinations. Not system-enforced — AEP does not hard-block above 100 — but throughput degradation is expected when more than 100 audiences are scheduled to export within any single hour. This hourly-rate guardrail governs export throughput density: it limits how many audiences can be simultaneously scheduled to export within any given hour, not the total number of audiences configured across all destinations. Organizations with more than 100 active batch audiences should stagger export schedules to avoid clustering exports in the same hour.
constraint.aep-adhoc-concurrent-job-per-audience-1
Adobe Experience Platform enforces a system limit of 1 concurrent ad-hoc activation job per audience. If an ad-hoc export job is already running for a given audience, a second ad-hoc job for the same audience cannot start until the first completes. This serialization constraint affects time-sensitive multi-wave campaign launches where the same audience must be activated to multiple destinations on-demand in rapid succession — each destination activation for the same audience queues behind the previous one.
constraint.aep-adhoc-activation-max-audiences-80
Adobe Experience Platform enforces a system limit of 80 audiences per ad-hoc (on-demand) activation job. Ad-hoc activation allows practitioners to immediately export audiences outside the regular scheduled export cadence — used for time-sensitive campaign launches, emergency suppressions, or re-activation triggers. When a campaign launch requires immediate activation of more than 80 audiences, multiple sequential ad-hoc jobs must be queued. This limit interacts with constraint.aep-adhoc-concurrent-job-per-audience-1: only 1 concurrent ad-hoc job per audience is allowed, serializing rapid multi-wave activations for any single audience.
source.github-com.tasmananalytics-tasman-dbt-mta-2026
Open-source dbt package (GPL-3.0) implementing a multi-touch attribution engine with boolean-algebra rule-based configuration. Supports last-touch, first-touch, U-shaped, and custom positional/time-based attribution models across Snowflake and BigQuery. Latest release v1.1.2, April 30, 2026.
source.docs-snowflake-com.en-user-guide-kafka-connector-classic-2026
Official Snowflake documentation for the v3 Classic Kafka Connector deprecation schedule and migration path to v4. Formal deprecation planned mid-2026 with 18-month migration window (end-of-life approximately late 2027). No immediate changes required for existing workloads.
source.braze-com.product-overview-2026
Official Braze product overview page documenting the platform's channel support (email, SMS/RCS, push, WhatsApp, web, LINE, KakaoTalk, media audience sync), BrazeAI capabilities (Decisioning Studio, Operator, Agent Console, Creative Studio), and 2025 scale metrics (4.5 trillion messages, 10.2 trillion API calls, 17.9B AI-powered decisions).
operational-task.configure-multi-touch-attribution-model
Define and run an algorithmic model that assigns fractional revenue or conversion credit across all touchpoints in a customer journey.
org-dim.raci-role
The communication architecture between stakeholders. Identifies who is Responsible, Accountable, Consulted, and Informed for each decision the CDP touches.
org-dim.raci-role.committee
Organizations where CDP decision authority is distributed across a standing committee (e.g., a Data Governance Council, a Digital Transformation Steering Committee, or a joint IT/Marketing/Finance committee) rather than vested in a single executive. Consensus-building and explicit decision criteria are required before any architectural path can advance. Recommendations for this governance pattern should be structured as a decision matrix — not a single recommendation — with clear TCO comparisons, risk-tier ratings, and reversibility assessments that each committee function can evaluate against its own priorities.
capability.streaming-ingestion
The ability to continuously ingest event streams or change-data-capture feeds into a cloud data warehouse with sub-minute end-to-end latency. In the composable CDP stack, streaming ingestion is the mechanism that keeps CDW profile models fresh enough to power near-real-time activation: events (page views, purchases, support interactions) flow from production systems to Snowflake within seconds to minutes, rather than waiting for the next scheduled batch load. Primary implementations: Snowflake Kafka Connector (Kafka-based; v3 Classic formal deprecation announcement planned mid-2026 — existing workloads fully supported, 18-month migration window starts upon announcement, EoL approximately late 2027; v4 current), Snowpipe Streaming SDK (programmatic, no Kafka required), and Fivetran connectors with append-mode scheduling (1–15 minute cadence; not open-connection streaming). Latency target: sub-minute to 5 minutes. Distinct from AEP Edge Network real-time event processing (sub-100ms, intra-request) — streaming ingestion feeds the CDW profile store, not the edge decisioning plane.
source.eagleeye-com.blog-beginners-guide-cdps-loyalty-platforms-2025
Eagle Eye guide covering CDP and loyalty platform integration challenges, real-time synchronization patterns, segmentation use cases, and CDC/event-driven architecture for retail, ecommerce, and travel loyalty programs.
source.cdp-com.articles-what-is-twilio-segment-2026
CDP.com analysis of Twilio Segment: 700+ connectors, MTU-based pricing, developer-first positioning, IDC MarketScape CDP Leader 2024-2025, composable CDP roadmap, and B2C scaling cost considerations.
source.cdp-com.articles-what-is-mparticle-2026
CDP.com analysis of mParticle: mobile-first CDP with 300+ integrations, Niche Player in Gartner MQ 2024, Strong Performer in Forrester Wave 2024, mParticle Credits pricing, Hybrid CDP on Snowflake (2025), Match Boost (2026).
source.wiley-law.alert-update-11th-circuit-vacates-fccs-one-to-one-tcpa-consent-rule-2025
Independent legal analysis from Wiley Rein (telecommunications law firm) on the January 24, 2025 Eleventh Circuit decision vacating the FCC's one-to-one TCPA consent rule in Insurance Marketing Coalition v. FCC.
source.tealium-com.products-audiencestream-cdp-2026
Tealium's official product page for AudienceStream CDP — a server-side platform for identity resolution, real-time visitor profiling, audience segmentation, and activation across 1,400+ integrations.
source.salesforcedevops-net.gartner-crowns-salesforce-data-360-as-cdp-leader-2026
Independent analysis of the January 2026 Gartner Magic Quadrant for Customer Data Platforms, covering vendor quadrant placements, market bifurcation (platformization vs. agentification), and the significance of Salesforce Data 360 as sole top Leader.
source.salesforceben-com.salesforce-data-cloud-renamed-to-data-360-agentforce-360-2025
Independent Salesforce community analysis of the October 2025 rename of Salesforce Data Cloud to Data 360, its strategic context within Agentforce 360, and the product's history and CDP capabilities.
source.rockhealth-com.insights-enabling-hipaa-safe-performance-marketing-ours-privacy-2025
Rock Health's investment thesis for Ours Privacy, the specialized HIPAA-compliant CDP for healthcare marketing. Provides an independent investor perspective on the product differentiation and healthcare marketing data compliance gap that Ours Privacy addresses.
source.martechtherapy-com.cdp-split-official-magic-quadrant-2026-review-2026
Independent marketing technology commentary on the 2026 Gartner Magic Quadrant for CDPs, analyzing the market bifurcation between platformization and agentification, vendor placements, and the significance of composable CDP entrants.
source.fcc-gov.document-fcc-removes-one-one-consent-rule-nullified-court-decision-2025
Official FCC page documenting the removal of the one-to-one TCPA consent rule after the 11th Circuit vacated it in January 2025. The FCC formally deleted the vacated rule language and reinstated the prior PEWC standard on August 29, 2025.
source.akamai-com.products-akamai-inference-cloud-platform-2025
Akamai's product page for Akamai Inference Cloud, launched October 2025 with NVIDIA. An edge-native AI inference platform delivering sub-millisecond response latency for fraud detection, financial decisioning, and real-time personalization at the edge — infrastructure layer, not a CDP-layer product.
source.iabeurope-eu.tcf-2-2-2023
IAB Europe's official TCF 2.2 specification page, covering the May 2023 update to the Transparency and Consent Framework. Key changes from TCF 2.0: legitimate interest legal basis removed for advertising purposes 3-6 (consent is now the only permissible basis at registration level); Global Vendor List (GVL) incremented to version 3 with per-purpose data retention periods and category-of-data declarations; CMPs must disclose total vendor count on first UI layer; plain-language user-facing descriptions replace legal text.
source.docs-tealium-com.server-side-getting-started-audiencestream-cdp-2026
Tealium's official developer documentation for AudienceStream CDP — covering introduction, architecture, visitor profile model, audience rules, connectors, and release notes through 2026.
vendor.treasure-ai
Enterprise CDP vendor rebranded from Treasure Data to Treasure AI in April 2026 (domain: treasure.ai). Now positioned as an 'Agentic Experience Platform' rather than a CDP, retaining core Intelligent CDP capabilities: 27+ built-in AI models (propensity scoring, lookalike generation, real-time next-best-action), 50+ prebuilt intelligent attributes, and AI agents for customer journey orchestration. 2026 Gartner MQ Challenger (pricing cited at nearly 2× the next-highest vendor). Forrester Wave B2C CDP Leader (Q3 2025). Primary verticals: retail, automotive, consumer goods, telecommunications.
vendor.tealium
Enterprise CDP offering AudienceStream (server-side identity resolution, real-time segmentation, 1,400+ integrations), iQ Tag Management, and — as of May 2026 — a native AI capability suite: AI at the Edge (on-device edge inference via mobile SDK), AI Decisioning with IYOM (Invoke Your Own Model), MCP-powered Configuration Agent, and AI Partner Ecosystem (Amazon Bedrock, Pinecone, LangChain). 2026 Gartner Magic Quadrant Challenger.
constraint.iab-tcf-2-2-eu-programmatic-advertising
IAB Europe Transparency and Consent Framework version 2.2 (effective April 2023, GVL v3) removes legitimate interest (LI) as a valid legal basis for advertising purposes 3–6 in the EU: ad delivery, ad personalization, ad measurement, audience research, and product development at the advertising layer. Consent is now the only permissible basis at CMP registration level for these purposes. CDPs exporting audiences to EU programmatic destinations (DV360, The Trade Desk, Amazon DSP) must enforce per-vendor, per-purpose consent filtering at the audience-export stage — before segment delivery, not after.
concept.consent-management
The operational practice of obtaining, recording, and machine-enforcing customer permissions for marketing communications across modalities. Consent state must be authoritative, near-real-time, and propagated to every activation point before any send.
source.hipaavault-com.resources-2026-hipaa-changes-2026
Overview of the 2026 HIPAA Security Rule overhaul: final rule published early 2026, effective approximately 60 days after Federal Register publication, compliance deadline approximately 180 days from effective date (~September 2026). New mandatory technical requirements beyond the 2003 rule: annual penetration testing, biannual vulnerability scanning, 72-hour critical-system recovery time objective, and annual written vendor verification of technical safeguard implementation. The addressable-vs-required two-tier standard is eliminated — all safeguards are now mandatory. Marketing rule and PHI marketing authorization requirements are unchanged. MFA, encryption at rest and in transit, asset inventories, and breach-notification timelines also confirmed as mandatory.
source.eur-lex-europa-eu.legal-content-en-txt-celex-32016r0679-2016
Primary text of Regulation (EU) 2016/679 of the European Parliament and of the Council on the protection of natural persons with regard to processing of personal data and free movement of such data (GDPR). Published in the Official Journal of the EU: 4 May 2016, L 119/1–88. Establishes core GDPR principles (transparency, lawfulness, data minimization), individual rights (access, rectification, erasure, objection), and controller/processor obligations including DPIAs and security measures. Authoritative primary regulatory source for Art. 6 lawful bases, Art. 7 consent requirements, Art. 17 right to erasure, and Art. 22 automated decision-making restrictions. Tier-1 fetch returned preamble and recitals through Recital 84; full 99-article text in the Official Journal.
recommendation.aep-locked-healthcare-cx-deepen-or-supplement
For a healthcare organization (health system, regional payer, specialty care network) running AEP as the incumbent CDP with Snowflake CDW, VP Engineering as RACI-Accountable, and HIPAA data-residency constraints, the primary evaluation question is: stay-and-optimize on AEP (Path 1), supplement with Hightouch activation on Snowflake (Path 2), or evaluate Amperity as AEP replacement (Path 3). The choice is driven by whether the gap is a configuration gap (Path 1), an activation-speed/flexibility gap (Path 2), or a data-model/identity resolution gap (Path 3). Switching cost is real — AEP licensing depth makes Path 1 the default unless a clear gap type is established.
org-dim.industry.healthcare
Highly regulated; PHI handling, HIPAA, often segregated infrastructure. Constrains both the architecture (data residency, encryption-at-rest) and the activation surface (consented vs non-consented uses).
constraint.hipaa-security-rule-2026
The HIPAA Security Rule underwent its first major overhaul since 2003. Final rule published early 2026; compliance deadline approximately September 2026 (180 days from effective date). Mandatory requirements: MFA for all ePHI access; encryption at rest and in transit (elevated from addressable to required); comprehensive asset inventories; documented systematic risk analyses; annual penetration testing; biannual vulnerability scanning; 72-hour critical-system RTO; annual written vendor verification of technical safeguard implementation; shortened breach-reporting windows. The addressable-vs-required safeguard distinction is eliminated. No changes to the HIPAA marketing rule or PHI marketing authorization requirements.
constraint.gdpr-right-to-erasure
GDPR Article 17 requires data controllers to erase personal data without undue delay on six grounds: original purpose expired; consent withdrawn; Art. 21 objection upheld; unlawful processing; legal obligation; child data (Art. 8). Controllers who made data public must notify other processors of erasure requests using reasonable technical measures. For CDP architectures, erasure must cascade across CDW tables, rETL destination syncs, ad platform custom audiences, CRM records, and all downstream copies. Five exceptions: freedom of expression; legal compliance; public health; archiving/research/statistics; legal claims.
source.martechtherapy-com.cdp-cep-midmarket-predictions-2026-2025
Independent practitioner analysis by Matthew Niederberger predicting mid-market CDP/CEP adoption in 2026 — composable and warehouse-native architectures gaining share, marketing-led mid-market firms as the primary growth segment, and the blurring boundary between CDP and CEP categories.
source.digitalbydefault-ai.blog-segment-cdp-review-2026-2026
Independent product review of Twilio Segment CDP published March 2026. Covers developer-first architecture (instrument-once philosophy), 400+ integrations (analytics, marketing automation, CDWs, CRMs), identity resolution (deterministic and probabilistic), Linked Profiles for account/household data, Protocols schema enforcement, Twilio Engage multichannel orchestration, and MTU-based pricing tiers.
source.cdp-com.articles-cdp-for-financial-services-2026
CDP.com article covering how CDPs enable compliant personalization, cross-sell optimization, and unified customer intelligence for banking, insurance, and wealth management. Highlights GLBA Safeguards Rule, GDPR, and CCPA compliance requirements; notes 20-30% cross-sell rate improvements.
source.cdp-com.articles-cdp-for-retail-2026
Six primary CDP use cases for retail: omnichannel customer identification, clienteling and associate-assisted selling, loyalty program optimization (behavior-based loyalty personalization improves active member rates 15-25%, reduces point liability via targeted redemption campaigns), retail media network activation (CDP-built audiences command 2-5x higher CPMs), seasonal campaign optimization, and store performance/inventory insights. Published April 12, 2026.
operational-task.configure-cloud-streaming-activation-destination
Define and authenticate a real-time streaming endpoint in a CDP platform so that audience segment qualifications can be delivered continuously to downstream consumers.
use-case.loyalty-program-personalization
CDP-powered loyalty program operation: unified member profiles across purchase channels, real-time tier and points balance calculation, behavior-based segment targeting, and triggered offer delivery when loyalty events occur (tier upgrade, milestone, redemption). Primary CDP use case for retail, ecommerce, and travel-hospitality organizations with loyalty programs operating at scale.
source.opusresearch-net.uniphore-acquires-actioniq-zero-data-2024
Independent analyst note (December 6, 2024) by Ian Jacobs of Opus Research covering Uniphore's acquisition of ActionIQ and Infoworks, announced December 5, 2024. Describes the Zero Data AI Cloud platform — ActionIQ's composable zero-copy data platform enables AI access to enterprise data without ETL replication. ActionIQ rebranded as CDP Agent within Uniphore. Acquisition price not disclosed.
source.chesshealthsolutions-com.2026-hipaa-rule-updates-2025
Healthcare consultant overview (November 6, 2025) of 2026 HIPAA compliance changes: NPP revision deadline February 16 2026; Security Rule modernization mandating MFA, encryption at rest and in transit, 24-hour breach reporting for BAs; Privacy Rule reproductive health protections (finalized April 2024); PHI marketing restrictions require individual authorization for PHI disclosures to third parties.
source.bounteous-com.boost-value-cdp-adobe-healthcare-shield-2024
Adobe implementation partner analysis (June 10, 2024) by Abby Matchett of Bounteous describing Adobe Healthcare Shield capabilities on Real-Time CDP: HIPAA BAA available for covered entities, customer-managed encryption keys, automated consent enforcement with healthcare-specific PHI data labels, field-level and role-based access controls. No specific Epic or Salesforce Health Cloud integration details mentioned.
operational-task.inspect-unified-customer-profile
Verify that behavioral events, profile attributes, and segment memberships are correctly unified and accessible through the platform UI, a REST API client, and a browser-based event debugger.
operational-task.ingest-cross-channel-behavioral-events
Generate and observe real behavioral events — anonymous browsing, account registration, mobile interactions — flowing into a unified profile with cross-device identity stitching.
operational-task.configure-data-collection-environment
Provision the SDK-based collection pipeline — tag properties, datastreams, and edge routing — needed to capture web and mobile events and forward them to a customer data store.
vendor.uniphore
Enterprise CDP and conversational AI platform, named a Leader in the 2026 Gartner Magic Quadrant for CDPs. Uniphore Marketing AI (formerly ActionIQ; acquisition closed late 2024) is the CDP product, built on a warehouse-native enterprise data fabric with zero-copy querying across all major CDWs. Strengths: LLM-based prompt segmentation, bidirectional CRM/ERP/contact-center integration, hybrid composable model spanning structured/semistructured/unstructured data. Post-acquisition, ActionIQ is branded CDP Agent within Uniphore's Zero Data AI Cloud. The platform's zero-copy data architecture enables enterprise AI deployments by connecting to data wherever it resides, eliminating ETL replication. Strategic-direction uncertainty noted by Gartner remains: product-management alignment between CDP Agent and Uniphore's broader conversational-AI and knowledge-processing platform is still in progress.
source.helpx-adobe-com.legal-product-descriptions-aep-activation-2026
Adobe's official product description for Adobe Experience Platform Activation, governing contractual export quotas. States that customers may batch export an average of up to 200 KB of Second-generation Data per Profile per year through the Data Access API. Separate from the operational guardrails documented in Experience League.
source.creately-com.blog-diagrams-types-of-organizational-charts-2024
Creately blog post covering types of organizational structures and their chart representations, including functional, divisional, matrix, flat/horizontal, and hierarchical org charts. Used as a reference taxonomy for org chart types in the KG.
source.businesswire-com.news-home-20260423931363-braze-agentic-ai-2026
Braze press release announcing agentic AI product launches in April 2026, including BrazeAI Operator (generally available), BrazeAI Agent Console, and Braze Creative Studio. BrazeAI Decisioning Studio is GA with EU hosting. These additions extend Braze's real-time decisioning capabilities into agent-orchestrated multi-step workflows.
source.businesswire-com.news-home-20250501266140-fivetran-acquires-census-2025
BusinessWire press release announcing Fivetran's signed agreement to acquire Census on May 1, 2025. Census is integrated into Fivetran as Fivetran Activations, bringing total connector count to 900+ and adding reverse ETL and data activation to Fivetran's ingestion-first platform. Census co-founder Boris Jabes joins to lead data activation strategy.
vendor.census
Reverse ETL platform connecting cloud data warehouses (Snowflake, BigQuery, Databricks, Redshift) to 200+ operational tools. Focuses on CRM-first and sales-tool activation alongside marketing channels. Peer to Hightouch in the composable CDP activation layer.
constraint.aep-second-gen-data-access-api-200kb
Adobe Experience Platform limits batch export of Second-generation Data via the Data Access API or any non-Destination-SDK means to 200 KB per profile per year — more restrictive than the first-generation 500 KB threshold, effectively closing the agnostic-export path for second-gen profile data and reinforcing the architectural push toward AEP-as-edge-node for composable CDP integrations.
capability.reverse-etl
The operational pattern of syncing computed segments, scores, or attributes from a cloud data warehouse directly to downstream activation destinations — email platforms, ad networks, CRMs, notification services — without routing through a packaged CDP.
source.scandiweb-com.blog-ecommerce-cdp-what-is-salesforce-data-cloud-2026
Independent e-commerce agency analysis of Salesforce Data 360 capabilities published March 2026: four-stage pipeline (ingest → standardize/identity-resolve → analyze → activate), deterministic and probabilistic identity matching (email high-confidence, phone medium, name combinations low), activation to Marketing Cloud journeys, Commerce Cloud personalization, external ad platforms, and service-agent context layers. Enterprise pricing not published — typically six figures annually plus 1.5–3x implementation cost.
vendor.salesforce-data-360
Salesforce's customer data platform — renamed from Data Cloud to Data 360 at Dreamforce (October 2025). The real-time data unification and activation engine underlying the Agentforce 360 AI-agent platform. Sole top Leader in the 2026 Gartner Magic Quadrant for CDPs.
source.real-time-clarity-md
User-authored taxonomy of Signal-to-Activation Time tiers. Foundational source for the latency-tier and many use-case nodes.
source.packaged-vs-composable-md
User-authored discussion of AEP's activation guardrails and how they constrain a composable strategy. Includes the comparison table that became the seed Tradeoff nodes.
source.overview-md
User-authored enumeration of the Technical and Organization considerations the agent must elicit. Foundational source for the OrgDimension and TechDimension trees and for the seed Question catalog.
source.cdp-recommendation-agent-md
User-authored project-motivation note. Articulates the agent's goal of providing transparent, unbiased CDP guidance and surfaces the data-liquidity friction point.
tradeoff.query-flexibility
How freely data engineers can run complex queries. Packaged CDPs cap compute (e.g. AEP 10-min timeout); composable stacks scale elastically.
tradeoff.integration-philosophy
Whether destinations come from the vendor's ecosystem (curated, governed, sometimes incomplete) or from agnostic reverse-ETL tools (broader coverage, customer-managed).
tradeoff.data-egress
How freely data can leave the CDP/CDW. Packaged CDPs govern egress by license terms; composable stacks govern only by raw infrastructure cost.
tradeoff.cost-predictability
Profile/volume-based licenses (predictable, capped) versus consumption-based pricing (pay for what you query, less predictable).
tradeoff.center-of-gravity
Where the canonical version of customer data lives — inside the CDP's profile store, or in the cloud data warehouse.
tech-dim.tech-ops
The disciplines that keep deployed systems running — issue management, build/release, observability, customer support, content delivery, documentation. Often under-staffed in organizations that focus all hiring on Dev.
tech-dim.marketing-ops
The intersection of marketing strategy and technical execution. Owns process mapping, campaign operations, data governance, sales-marketing alignment, and budget oversight.
tech-dim.dev-team
The engineering disciplines required to build, deploy, and operate a modern marketing data stack. Each sub-discipline maps to specific Capabilities the architecture must support.
tech-dim.dev-team.frontend-tags
The team responsible for tag deployment, tracking, and event collection on web properties. Owns the integrity of the upstream signal from the user's browser.
tech-dim.dev-team.backend-stream-processing
The team responsible for low-latency event processing — Kafka/Pulsar/Pub-Sub, real-time feature stores, decisioning services. Required for any activation at Intra-session latency or below.
tech-dim.dev-team.backend-data-engineering
The team responsible for ELT pipelines, data modeling, and data quality in the warehouse. The load-bearing discipline for any composable architecture.
tech-dim.dev-team.agent-development
The team responsible for building and operating AI agents that augment marketing workflows. A new discipline that recasts which capabilities are buy-vs-build.
org-dim.marketing-goal.retention
Reducing churn and re-engaging lapsed customers. Demands churn-prediction models, win-back orchestration, and tight integration with product-usage telemetry.
org-dim.marketing-goal.new-product
Driving adoption of a newly launched product. Heavy on segmentation by existing-customer affinity, A/B testing infrastructure, and early-signal analytics.
org-dim.marketing-goal
Top-level taxonomy of marketing objectives the organization is optimizing for. Each goal shapes which use cases, latency tiers, and architectural patterns are appropriate.
org-dim.marketing-goal.engagement-advocacy
Deepening the relationship with active customers and turning them into advocates (referrals, reviews, social amplification).
org-dim.marketing-goal.efficiency-data
Reducing waste in marketing spend, improving attribution, and increasing decision-speed through data. The "data-as-internal-product" stance.
org-dim.marketing-goal.customer-lifetime-value
Optimizing the long-term value of existing customers. Demands rich profile state, predictive scoring, and cross-channel orchestration over months/years.
org-dim.marketing-goal.customer-experience
Optimizing the felt quality of the customer's interactions across all touchpoints. The most cross-modality goal — every modality contributes to the experience.
org-dim.marketing-goal.brand-awareness
Building familiarity and positive sentiment among broad audiences. Often the *least* CDP-intensive goal — heavy on paid media, light on personalization.
org-dim.marketing-goal.acquisition
Optimizing for new-customer growth. Implies investment in identity resolution from anonymous traffic, attribution infrastructure, and look-alike audience capabilities.
org-dim.industry.telecommunications
Subscription-based with high churn sensitivity, usage-driven retention models, and complex household/account hierarchies.
org-dim.industry.retail
Brick-and-mortar (and hybrid) retail businesses. Dominant use cases involve in-store/online experience consistency, loyalty programs, and POS-side activation.
org-dim.industry.ecommerce
Online-first transactional businesses. Dominant use cases include cart recovery, product recommendations, post-purchase upsell. High clickstream volume and rich first-party data.
vendor.bigquery
Google Cloud's serverless data warehouse. Strong fit for organizations already on Google Marketing Platform / GA4 / Google Ads stack.
vendor.adobe-audience-manager
Adobe's historical DMP/audience platform. Played the role of trait-state-change tracker for journey progression — a function now largely subsumed by AEP's profile and segmentation features.
modality.web
Browser-based delivery — page renders, banners, on-page personalization. Native fit for Intra-page (client-side) and Inter-page (edge-decisioned) latency tiers.
modality.sms
Text-message delivery. Highly regulated (TCPA in the US) and cost-per-message non-trivial; reserved for high-value activations.
modality.push
Mobile/web push delivery via APNs, FCM, and web-push protocols. Permission-revocable in one tap — fatigue is the dominant operational risk.
modality.pos
In-store transaction terminals. Typically batch-updated (Inter-session or slower) which creates latency-mismatch challenges with web/mobile activations expecting Inter-page freshness.
modality.mobile-app
Native iOS/Android app delivery. Strong identity (device ID + login), supports push notifications, often the channel with the deepest user-state visibility.
modality.email
Outbound message delivery via SMTP/ESPs. Tolerates Intra-day to Weekly latency tiers; transactional cases (Inter-page) are the exception.
modality.connected-device
Smart TVs, IoT sensors, wearables. Highly variable latency profiles; identity is often device-only (no user login).
latency-tier.weekly
Lapsed-user re-engagement and loyalty-program offers.
latency-tier.temporal
Signals driven by external temporal conditions rather than user activity — birthdays, holidays, seasons.
latency-tier.intra-session
Streaming evaluation of the current event combined with the user's recent history; supports stochastic predictions across multiple interactions rather than a single materialized attribute lookup.
latency-tier.intra-page
Decisioning happens entirely inside the browser/app context with no network round-trip; data is already local (IndexedDB, OPFS, WASM, in-memory).
latency-tier.intra-day
Transactional follow-ups tied to a single-day journey; classic batch-processing window.
latency-tier.inter-session
Audience segment membership refresh on downstream activation systems; common floor for high-throughput programmatic systems where streaming cache updates are uneconomic.
latency-tier.inter-page
Edge-decisioned next-page personalization. The challenge is propagation across modalities, since the edge response may not yet be reflected in batch systems.
latency-tier.inter-day
Cross-day onboarding and lifecycle activations.
latency-tier.daily
Multi-day follow-ups driven by content engagement or upsell signals.
use-case.transactional-email
User-action-triggered emails with an explicit functional purpose; the email *is* part of the workflow, not a marketing add-on.
use-case.next-best-action
Stochastic recommendation of the user's likely next action, evaluated across multiple events and historical context. Lives at the Intra-session latency tier.
use-case.coupon-mobile-pos-sync
Cross-modality activation where the same offer surfaces in mobile (real-time) and at point-of-sale (batch). The latency mismatch is an operational concern, not a bug.
use-case.abandoned-cart
Reminding a user that they have items left in their cart. High-ROI on average due to direct association with expressed user intent — but ROI can invert if the reminder is too frequent or annoying.
pattern.micro-batching-workaround
When a CDP enforces query-time caps (e.g. AEP's 10-minute timeout), engineers break complex queries into time-windowed slices that complete individually and merge results downstream.
pattern.fail-fast-within-compliance
Combine an experimentation culture (one experiment beats a thousand expert opinions) with the regulatory and policy infrastructure required for the industry. "Fail-fast" never implies non-compliance.
pattern.aep-as-edge-node
For organizations heavily invested in AEP but pivoting toward a composable stack, treat AEP as a downstream edge node rather than the single source of truth. Heavy lifting happens in the CDW; AEP receives only aggregated profile traits and pre-computed audiences.
constraint.aep-adhoc-query-timeout-10min
AEP's ad-hoc query service caps execution at 10 minutes, blocking complex JOINs across large fact tables and forcing engineers into micro-batching or CTAS workarounds.
concept.signal-to-activation-time
The interval between when a signal is generated (e.g. the timestamp on an HTTP GET) and when a marketing activation derived from that signal is ready for delivery via mobile app, web page, SMS, etc. The single most clarifying axis when discussing "real-time."
concept.roi-uncertainty
The degree of confidence in the revenue impact of a particular architectural decision. Even use cases that seem obviously good (e.g. abandoned-cart reminders) can produce negative net impact in specific contexts.
concept.modality-complexity
The architectural cost imposed by supporting multiple delivery channels in parallel. Each modality adds its own latency profile, identity context, and governance regime; cross-modality consistency is the hidden cost driver.
concept.data-liquidity
How freely data flows to where compute is most efficient. In a truly composable architecture, liquidity is high; in a packaged CDP it is governed by license terms.
source.rockhealth-com.insights-healthcare-innovation-turn-of-2026-2025
Rock Health's December 2025 research report applying an Innovation Maturity Curve framework to categorize digital health innovations across six priority segments: wearables, clinical analytics, environmental health, preventive care, mental health technology, and benefits infrastructure. Provides a measurable taxonomy of digital health company types scored by three dimensions — research activity (PubMed volume), venture funding (Rock Health database), and partnership activity.
source.hightouch-com.docs-match-booster-implementation-2026
Hightouch Match Booster implementation documentation covering identity graph enrichment, PII hashing (SHA256/MD5) for emails and phone numbers, match key types (email, phone, MAIDs, postal addresses, IP addresses), household-level matching, CTV ID translation, and secure mapping of hashed primary keys against third-party identity graph.
source.braze-com.product-2026
Braze product landing page documenting channel support (email, SMS/RCS, WhatsApp, push notifications, mobile app, web, LINE, KakaoTalk, media audience sync), BrazeAI decisioning studio for 1:1 personalization, BrazeAI Agents for autonomous engagement optimization, predictive intelligence, 170+ turnkey partner integrations, and 2025 Gartner Magic Quadrant for Multichannel Marketing Hubs recognition.
archetype.aep-locked-healthcare-cx-evaluator
A healthcare organization (health system, regional payer, or specialty care network) running Adobe Experience Platform as the incumbent CDP with Snowflake as CDW, seeking CX improvement across digital patient/member touchpoints (web, mobile, email), constrained by HIPAA data residency requirements, with VP Engineering as the RACI-Accountable. Distinguishing signals: CX primary goal (not acquisition); AEP already licensed and in use (switching cost is real); data residency true (HIPAA BAA must cover every layer); medium data engineering staffing; intra-session latency requirement; VP Engineering driving the evaluation.
vendor.braze
Customer engagement platform (CEP) with AI-powered real-time decisioning (BrazeAI Decisioning Studio). Cross-channel orchestration across email, push, SMS, in-app messaging, and web. Receives audience data and activation signals from a CDP or CDW and manages delivery execution and experimentation. Not a CDP — Braze does not maintain a canonical profile or run segmentation independently of an upstream data source.
source.tech-training-getting-started-ex8
Exercise 0.8 from the AEP comprehensive technical tutorial. Demonstrates retrieving a unified customer profile in real time via the CX App (call center simulation) using a phone number identity lookup — showing cross-channel customer context available at the point of service.
source.tech-training-getting-started-ex7
Exercise 0.7 from the AEP comprehensive technical tutorial. Covers navigating the AEP Profile UI to inspect a specific customer profile by identity namespace and value, viewing Profile Attributes, Experience Events timeline, and Segment Membership — illustrating the unified identity graph across email, ECID, and phone namespaces.
source.tech-training-getting-started-ex6
Exercise 0.6 from the AEP comprehensive technical tutorial. Demonstrates mobile app event ingestion using the same authenticated email identity, cross-device event stitching in the Real-time Customer Profile, and cross-channel profile continuity visible from both mobile and desktop surfaces.
source.tech-training-getting-started-ex5
Exercise 0.5 from the AEP comprehensive technical tutorial. Demonstrates anonymous behavioral event capture (product views via Web SDK), account registration merging ECID to email/phone identifiers, and the resulting unified Real-time Customer Profile visible in the Profile Viewer panel.
source.tech-training-getting-started-ex2
Exercise 0.2 from the AEP comprehensive technical tutorial. Covers creating Datastreams for web and mobile channels, adding the Adobe Experience Platform service with Event and Profile Dataset targets, linking tag properties to the Datastream via the Web SDK and Edge Network extensions, and publishing to the Development environment.
source.tech-training-getting-started-ex1
Exercise 0.1 from the AEP comprehensive technical tutorial. Covers using Adobe Demo System Next Quick Setup to provision web and mobile Data Collection tag properties and associated projects for an AEP sandbox.
source.salesforceben-com.salesforce-marketing-cloud-ai-rebuild-2026
Salesforce Marketing Cloud is being rebranded to Agentforce Marketing, representing a strategic shift to AI-driven marketing. Five pillars of the rebuilt platform: (1) Two-Way Messaging — bidirectional email and SMS with AI agent handling; (2) Unified Data Customer Journeys — powered by Data 360 as a single intelligent customer brain; (3) Adaptive Experiences — dynamic personalization adjusting content, timing, and channel based on real-time behavioral signals; (4) AI-Speed Campaign Creation — compresses planning from 7-8 weeks via Campaign Creation Agents, Orchestration Agents, and Campaign Activation Agents; (5) Insights That Drive Real Impact — embedded predictive analytics and automated narrative reporting. Channels: email, SMS, push, web, and ads.
source.prnewswire-com.news-releases-rokt-and-mparticle-merge-2025
Rokt announced a US$300 million investment in mParticle on January 16, 2025, taking a 100% stake. mParticle founders remain in leadership — Michael Katz as mParticle CEO, Andrew Katz as CTO of Rokt, Jason Lynn as mParticle CPO. Rokt powers 6.5 billion annual ecommerce transactions for 400 million customers including PayPal, Uber, and Hulu. Strategic rationale: combining Rokt ecommerce-moment relevance with mParticle real-time CDP delivers up to 50% better consumer and business outcomes for joint clients.
source.hipaajournal-com.hipaa-marketing-rules-2025
HIPAA defines marketing as any communication about a product or service that encourages recipients to purchase or use it. Authorization from the individual (or representative) is mandatory for using or disclosing PHI in marketing communications, with two narrow exceptions: face-to-face communications and promotional gifts of nominal value. Covered entities cannot disclose PHI in exchange for remuneration to a third party for that party to market its own products. Exceptions to authorization exist for refill reminders, alternative treatment recommendations, and case management without remuneration beyond communication costs. Digital-age obligations include no PHI in email subject lines, no PHI in social media posts, HIPAA-compliant form channels.
source.cdp-com.articles-cdp-for-media-2026
Six primary CDP use cases for media and publishing: paywall optimization (15-30% subscription conversion improvement), audience segmentation for first-party programmatic (2-4x higher CPMs vs open programmatic), content personalization via real-time profile activation (20-35% pages-per-session increase), churn prevention via predictive models (10-20% cancellation reduction), ad yield optimization via enriched audience data (25-40% CPM lifts), and cross-platform unification revealing 30-50% fewer unique audience members than siloed analytics. Over 80% of publisher traffic is anonymous, requiring progressive identity resolution. Privacy capabilities include GDPR/CCPA adherence, consent-aware activation, data residency controls, and purpose limitation.
vendor.salesforce-marketing-cloud
Multi-channel customer engagement platform (CEP) from Salesforce, also known as Agentforce Marketing (Q1 2026 official rebrand). Orchestrates email, SMS, push, social, and paid-media campaigns using profiles and segments from Salesforce Data 360 as the upstream data layer. Next-gen Marketing Cloud (2025) adds Agentforce-powered conversational email and AI-driven journey orchestration. Not a CDP — receives decisions from Data 360; executes and measures delivery.
vendor.mparticle
Mobile-first CDP acquired by Rokt in January 2025 (US$300M, 100% stake). Rokt powers ecommerce transaction-moment relevance; mParticle provides the upstream profile unification and governance layer. Core: IDSync identity resolution, Data Master schema enforcement, Audiences, Custom Rules, Events/Profile/Firehose APIs. Cortex AI predictive models — sold separately. 'Hundreds of vendors' integration footprint. Strongest in mobile-first B2C digital brands. Not included in Gartner MQ 2026 — did not meet inclusion criteria for CDP customers and contract value.
vendor.akamai
Edge cloud and CDN platform with Akamai Inference Cloud (launched October 2025, NVIDIA-backed), which runs transformer-class AI models at CDN Points of Presence (PoPs) with sub-millisecond inference latency. In CDP architectures, Akamai provides the edge-AI inference tier for intra-page (<100 ms) personalization decisions where ML model complexity is required — use cases that exceed what rule-based edge compute (Fastly, Cloudflare Workers) can support natively.
source.docs-tealium-com.server-side-getting-started-eventstream-api-hub-connectors-2024
Tealium documentation for EventStream connectors: integrations between Tealium and external vendors that transmit data via event feeds and actions. Covers connector credential storage, actions (vendor API methods), event feed triggers, and data mappings that translate event attributes into vendor-required parameters.
source.chiefmartec-com.2025-12-heres-your-copy-of-our-martech-for-2026-report-free-and-ungated-2025
Scott Brinker's 127-page annual Martech for 2026 report covering SaaS and AI realities in martech, AI agents, context engineering, composable architectures, and mid-market marketing trends. Composable CDP share projected to double from 5% to 12% between 2024 and 2026; mid-market identified as dominant growth vector.
recommendation.platform-bundled-ai-decisioning-for-stream-processing-gap
When intra-session latency personalization is needed (10 s – 5 min tier) but no backend stream-processing team is present, prefer platform-bundled AI decisioning (Layer 2) over building a custom streaming infrastructure. Building stream processing without dedicated staff is high-risk; platform-bundled decisioning absorbs the infrastructure layer at the cost of vendor lock-in.
source.martech-org.oracle-enhances-unity-cdp-offering-2024
MarTech.org editorial coverage of Oracle Unity CDP September 2024 enhancements including Account Profile Explorer, buying group and opportunity scoring with 27+ AI models, industry onboarding accelerators for 8 verticals, and Oracle Analytics Cloud integration. First published September 2024.
source.consumerfinanceinsights-com.fcc-final-rule-eliminating-one-to-one-consent-requirement-2025
Goodwin Procter law firm analysis of the FCC's September 2025 Final Rule that formally eliminated the one-to-one TCPA consent requirement following the Eleventh Circuit vacatur in Insurance Marketing Coalition Ltd. v. FCC. Confirms standard PEWC requirements remain; only the single-seller restriction is gone.
source.burr-com.fcc-delays-tcpa-revoke-all-rule-2027-2026
Burr & Forman law firm analysis of the FCC January 2026 order extending the TCPA Revoke-All rule (47 CFR 64.1200(a)(10)) compliance deadline to January 31, 2027. Rule requires callers to treat an opt-out for any one call type as a universal opt-out from all future calls. Delay granted while FCC reviews whether to allow more granular consumer control.
vendor.oracle
Enterprise packaged CDP from Oracle, named a Leader in the 2026 Gartner Magic Quadrant for CDPs. Unifies online and offline customer data within the Oracle ecosystem; applies AI/ML predictive scoring, real-time segmentation, and next-best-action recommendations; activates to Oracle CX suite and third-party destinations. Best fit for organizations already invested in Oracle Fusion ERP or Oracle CX Cloud.
constraint.tcpa-revoke-all-universal-opt-out
FCC rule (47 CFR § 64.1200(a)(10)) requiring any opt-out communication from a consumer — regardless of channel — to revoke all consent for robocalls and robotexts from that caller. Originally adopted 2024, compliance deadline extended by FCC order to January 31, 2027. CDP and composable activation stacks routing messages across SMS, email, and push must implement cross-channel suppression propagation before the deadline: one STOP message to any channel must suppress the profile across all downstream activation destinations.
constraint.tcpa-prior-express-written-consent-sms
Federal requirement under 47 U.S.C. § 227 (TCPA) that marketers obtain Prior Express Written Consent (PEWC) before sending marketing SMS messages. Consent must be written (digital opt-in accepted), clearly authorize the specific sender and channel, and be revocable at any time (STOP keyword must propagate to suppression immediately). Statutory damages $500–$1,500 per unconsented message.
source.treasure-ai.product-intelligent-cdp-2026
Treasure Data's product page for its Intelligent CDP. Treasure Data positions its platform as an AI-first CDP with embedded AI models, AI agents for marketers, customer journey orchestration, predictive segmentation, and built-in ML. 2026 Gartner MQ Challenger (dropped from higher position). Gartner cited pricing as a concern: Treasure Data's sample-scenario response was nearly 2x the second-highest vendor and 4x the average across all vendors. Core capabilities: data integration, identity resolution, audience segmentation, AI-powered journeys, and 27+ built-in AI models.
source.salesforce-com.news-stories-marketing-cloud-next-announcement-2025
Salesforce's official press announcement (June 2025) for Marketing Cloud Next, consolidating nine previous Salesforce marketing acquisitions into a single platform powered by Agentforce AI agents. Built natively on Data 360 (formerly Data Cloud). Supports both B2B and B2C marketing from one platform. Key agentic capabilities: Agentforce Campaign Creation, Agentforce Personalization Decisioning, Agentforce Web Curation, Agentforce Lead Generation and Management. Channels: email, SMS, and web. Conversational two-way email rolled out Spring 2026.
source.martech-health.articles-freshpaint-healthcare-privacy-platform-2024
Independent healthcare martech trade media article confirming that HHS updated its March 2024 guidance on online tracking technologies to explicitly name CDPs like Freshpaint as viable alternatives to non-BAA tracking technologies. Freshpaint's Healthcare Privacy Platform replaces risky web trackers (Google, Facebook pixels) with a BAA-supported CDP using ID masking, anonymization, and cryptographic hashing to protect PHI. Enables HIPAA-compliant performance marketing for healthcare organizations.
source.businesswire-com.news-home-20251202496602-braze-decisioning-studio-gcp-2025
BusinessWire announcement (December 2025) confirming BrazeAI Decisioning Studio is available on Google Cloud Marketplace, enabling enterprise procurement through committed GCP cloud spend. BrazeAI Decisioning Studio uses reinforcement learning to autonomously experiment and determine the optimal channel, message, offer, timing, and frequency for each individual customer. Offered in two tiers: Go and Pro. Pro tier supports Braze, Salesforce Marketing Cloud, and Klaviyo as underlying CEPs. Google Cloud Marketplace availability confirms GA status and enterprise distribution.
org-dim.raci-role.responsible
The practitioner who builds, configures, and operates the CDP stack. Typically a data engineer, marketing operations engineer, or IT implementer. As the most likely primary user of the recommendation agent, the Responsible role needs tactically specific guidance: which vendors support which integration patterns, what engineering team capabilities are required (e.g., dev-team.backend-stream-processing for real-time segmentation), and what the operational complexity of each architectural choice looks like day-to-day.
org-dim.raci-role.accountable
The executive sponsor who signs off on CDP decisions and is ultimately answerable for the outcome. Typically a VP Marketing Ops, CDO, CMO, or CTO. Recommendations for this role should be strategically framed: what are the business outcomes and risk tradeoffs, which architectural path is defensible to peer executives with different priorities (IT, Finance, Marketing), and what does the total cost of ownership look like over a 3-year horizon.
org-dim.org-chart-type
How decisions are made and authority flows through the organization. Materially affects how a CDP rollout will land and which stakeholders the agent must address.
vendor.treasure-data
Enterprise packaged CDP positioned as an 'Intelligent CDP' with embedded AI agents, AI-powered customer journey orchestration, predictive segmentation (27+ built-in AI models including propensity scoring, lookalike generation, and real-time recommendations), and 50+ prebuilt intelligent attributes. 2026 Gartner MQ Challenger (pricing noted by Gartner as nearly 2× the next-highest vendor). Forrester Wave B2C CDP Leader (Q3 2025). Primarily serves large enterprise retail, automotive, consumer goods, and telecommunications organizations.
vendor.freshpaint
Healthcare Privacy Platform that automatically detects and blocks PHI before forwarding behavioral event data to downstream analytics and marketing tools. HHS guidance (March 2024) explicitly named CDPs like Freshpaint as viable alternatives to non-BAA tracking technologies. Uses server-side cryptographic hashing (irreversible) and ID masking to de-identify data at the tracking layer — enabling HIPAA-compliant performance marketing by replacing non-BAA tracking pixels (Google, Facebook) with a BAA-supported CDP layer. Not a general-purpose CDP — Freshpaint is a healthcare-vertical specialist focused on the tracking-layer PHI interception problem.
modality.paid-media
Digital paid advertising channels — programmatic display, paid social (Meta, Google, LinkedIn, TikTok), CTV/OTT, and paid search — activated via CDP audience segments. The primary governance dimension is identity: paid media requires a deterministic match signal (hashed email, IDFA/GAID, LiveRamp RampID) or probabilistic look-alike modeling, and cannot use PHI in HIPAA-covered contexts. Typical latency tier: inter-day (custom audience refresh) to weekly (look-alike model refresh). CCPA right-to-opt-out-of-sale and GDPR legitimate-interest balancing apply to targeting; HIPAA prohibits routing PHI to non-BAA ad platforms.
constraint.aep-second-gen-export-1500kb
AEP allows export of up to 1500 KB of Second-Generation Data per profile per year — but only via pre-built or Destination-SDK destinations, forcing reliance on AEP's ecosystem rather than agnostic reverse-ETL.
constraint.aep-first-gen-export-500kb
Adobe Experience Platform caps batch export of First-Generation Data via the Data Access API at an average of 500 KB per profile per year — a binding limit on using AEP as an upstream source for an enterprise data warehouse.
capability.audience-segmentation
The process of computing cohort membership from customer profile attributes — behavioral events, computed scores, demographic data, and consent flags. Packaged CDPs offer UI-based, no-code Segment Builder interfaces; composable CDPs implement segmentation as SQL or dbt models running in the CDW. The central architectural question is whether the segmentation engine runs inside the CDP (packaged) or inside the data warehouse (composable). Segment freshness — how quickly a new customer lands in the correct segment — is the primary latency and cost driver.
concept.activation-tax
The vendor-imposed structural cost of moving data out of a packaged CDP to external systems — expressed as byte export quotas, concurrent activation limits, audience slot ceilings, and ecosystem-lock incentives. Distinct from infrastructure egress costs (which scale predictably with volume), the activation tax is an architectural constraint enforced by commercial product decisions. High activation tax correlates with low data liquidity in the external direction; the `pattern.aep-as-edge-node` and composable CDP architectures are the primary structural mitigations.
source.tasman-ai.news-open-source-dbt-package-for-advanced-attribution-2026
Official Tasman Analytics blog post announcing the open-source tasman-dbt-mta dbt package for configurable multi-touch attribution (MTA), supporting last-touch, first-touch, and U-shaped models with parameterized attribution windows and deterministic rules-based attribution on Snowflake and BigQuery.
archetype.engineering-led-mid-market-composable-evaluator
A mid-market organization ($25M–$500M revenue, 50K–1M customer profiles) where data engineering or IT is RACI-Accountable for the CDP selection and operates the activation layer via SQL, dbt, and reverse-ETL — without relying on marketing self-service tools. Evaluating SQL-first composable CDPs (Hightouch, Fivetran Activations/Census) or engineering-accessible packaged CDPs (Segment, mParticle) over marketer-self-service tools. Distinguished from enterprise engineering-led archetypes (AEP, Salesforce ecosystem) by budget ceiling (~$400K/year), absence of an incumbent enterprise CDP, and an existing CDW that data engineering already owns.
vendor.twilio-segment
Niche Player in the 2026 Gartner Magic Quadrant for CDPs. Developer-first CDP with 700+ connectors, MTU-based pricing, and deterministic identity resolution. Module architecture: Twilio Segment Connections (data collection + routing), Twilio Unify (profile unification), Twilio Engage Foundations (basic orchestration). IDC MarketScape CDP Leader 2024-2025. Best for engineering-led teams with high integration breadth and privacy governance requirements. Owned by Twilio since 2020 ($3.2B acquisition); placed in Twilio 'Data & Applications' unit following June 2025 restructure. Twilio Engage Premier sunset June 2025. $285.7M intangible asset impairment recorded Q4 2023.
source.rockhealth-com.insights-2025-year-end-digital-health-funding-overview-2026
Rock Health annual digital health market report for 2025; US digital health startups raised $14.2B (35% YoY increase); 50% of deals involved AI-enabled companies capturing 54% of total funding; M&A surged 61% to 195 deals; defines funding categories and investment trends across the digital health industry.
source.cmswire-com.customer-data-platforms-see-growth-lack-new-players-2025
CMSWire independent trade analysis of the 2025 CDP market; distinguishes engineering-led composable CDP buyers from marketing-led business users; composable CDPs represent less than 5% of market by size despite 12.9% annual growth; 71% of CDP vendors are delivery/campaign-focused.
org-dim.operational-profile.marketing-led-cdp
An organization where marketing owns and directly operates the CDP activation layer — building audiences, configuring journeys, and deploying AI-powered decisioning — without requiring data engineering participation in day-to-day CDP operation. Data engineering provides the warehouse foundation (BigQuery, Snowflake, or Databricks) and data contracts; marketing operates on top. This operational profile is the primary selection signal for marketer-accessible composable CDPs (GrowthLoop, BrazeAI Decisioning Studio) over data-engineering-first tools (Hightouch, Fivetran Activations).
source.prweb-com.releases-ours-privacy-next-gen-analytics-suite-302638126-2025
Ours Privacy December 2025 press release announcing expansion from pixel-replacement CDP to full-stack healthcare marketing platform — adding MTA, web analytics, session replay, and user funnels, all HIPAA-compliant.
source.hightouch-com.docs-syncs-warehouse-sync-logs-2026
Hightouch Warehouse Sync Logs writes sync metadata (per-row status and errors) back to the data warehouse, enabling SQL-based analysis of audience delivery, error categorization, and sync history. Requires Lightning engine and Business tier. Supports Redshift, Databricks, BigQuery, Greenplum, PostgreSQL, Snowflake.
source.hightouch-com.docs-match-booster-overview-2026
Hightouch Match Booster enriches audiences and conversion events with supplementary identifiers (alternate emails, phone numbers, device IDs) sent to ad platforms. Warehouse-native — data never persisted on Hightouch infrastructure beyond a hash of the model primary key.
source.experienceleague-adobe-com.en-docs-experience-platform-destinations-guardrails-2026
Adobe's official documentation for data activation export guardrails in Adobe Experience Platform — per-destination audience caps (250 performance guardrail, 20 hard limit for external audiences), batch frequency limits, streaming limits, dataset export limits, and ad-hoc activation limits. Page last updated May 23, 2026.
source.developers-facebook-com.docs-whatsapp-overview-2026
Official WhatsApp Business Platform developer documentation covering the Cloud API, messaging categories (service, marketing, authentication, utility), opt-in consent requirements, 24-hour customer care messaging window, and template-based marketing message obligations.
modality.messaging-apps
Conversational messaging channels delivered via over-the-top (OTT) mobile applications and rich carrier protocols — including WhatsApp, LINE (Japan/Southeast Asia), KakaoTalk (South Korea), WeChat (China), and RCS (Rich Communication Services, a carrier-level SMS successor available via Google Messages and Samsung Messages). Distinct from modality.sms (carrier-based text; international, no-app-required) and modality.push (device OS-level notifications; app install required but not app-specific messaging platform). OTT messaging channels require platform-specific API credentials and template approvals (WhatsApp Business API requires Meta approval; LINE and KakaoTalk require regional business account registration). CDP activation to messaging-app channels typically routes through a customer engagement platform (Braze, Salesforce Marketing Cloud, Vonage, Infobip) that holds platform API credentials and handles session threading, message template compliance, and opt-in verification. High opt-in friction and geography-specific market penetration make this modality most relevant for D2C, retail, and financial services organizations operating in EMEA, Japan, Southeast Asia, or South Korea.
constraint.whatsapp-24h-messaging-window
Within 24 hours of the most recent inbound message from a user, WhatsApp Business API permits free-form outbound service messages (text, images, interactive buttons, carousels). Outside this window, only pre-approved message templates may be sent — including all marketing re-engagement messages. Marketing templates additionally require explicit consumer opt-in consent separate from the 24-hour window. This platform-enforced policy (not a regulation) materially constrains CDP activation planning for win-back, abandoned-cart, and lapse-reactivation use cases, which by definition target users who are not currently active. CDP practitioners must build and pre-approve template libraries before activating these use cases over WhatsApp, adding compliance lead time compared to email or SMS.
source.prnewswire-com.news-releases-blueshift-next-era-customer-ai-302641676-2025
Blueshift press release announcing Compass AI agent (identifies untapped revenue by scanning customer behavior and surfacing under-engaged audiences) and Launchpad AI agent (converts marketing ideas to cross-channel campaigns via conversational workflows). Early Optimizer Agent users achieved ~10x more experiments with 36% avg goal metric lift. GA planned Spring 2026.
source.openloyalty-io.insider-loyalty-software-trends-2026
Annual loyalty software trends report (March 2, 2026) covering personalization investment priorities, predictive segmentation adoption, and CDP integration patterns. Key stats: 59.7% cite personalization as top investment area in 2026; 41.5% cite predictive segmentation as major impact driver (up from 21.3% in 2021); 35% of brands explicitly prioritize ROI proof. Composable architectures gaining ground; loyalty software must integrate with CDPs rather than replace them.
source.blueshift-com.customer-engagement-platform-overview-2026
Blueshift vendor product page for their AI Customer Engagement Platform combining native CDP, agentic AI (Customer AI Agents), predictive AI, and cross-channel orchestration across email, SMS, push, in-app, web, and paid media.
vendor.blueconic
Niche Player in the 2026 Gartner Magic Quadrant for CDPs. Marketing-first CDP specializing in NBA (next-best-action) decisioning and upstream consent enforcement. 'Growth plays' are preconfigured, outcome-driven solution bundles (lead generation, cart abandonment, product discovery). North America/Europe focused; B2C retail, ecommerce, media, and publishing verticals. Zero-copy data sharing with Snowflake only — narrowest CDW integration of any 2026 Gartner MQ vendor.
vendor.amperity
Niche Player in the 2026 Gartner Magic Quadrant for CDPs. AI-powered identity resolution platform with Lakehouse CDP architecture. Amperity's patented Stitch technology applies up to 45 ML models to resolve customer identity without requiring rigid schemas. 'Bring your own architecture' model separates compute and storage; zero-copy data sharing with Snowflake and Databricks. North America focused; strongest in retail, travel, and financial services.
pattern.platformization
Strategic CDP architecture in which the CDP is positioned as the foundational layer of a broad, integrated enterprise application ecosystem. Goal is cross-functional orchestration — unified commercial motions across marketing, sales, service, commerce, and analytics — with the CDP governing data flow between front-office and back-office systems. Canonical vendor: Salesforce Data 360. Contrasts with pattern.agentification.
pattern.agentification
Strategic CDP architecture in which the CDP is positioned as a lightweight, composable data foundation for autonomous AI-agent execution. Operational orchestration shifts from human campaign managers to specialized AI agents pursuing marketer-defined goals. Contrasts with pattern.platformization. Canonical vendors: Hightouch (AI Decisioning), Treasure Data (AI Agent Foundry, Marketing Super Agent roadmap).
source.salesforce-com.blog-next-gen-marketing-cloud-details-2026
Salesforce's official blog post detailing Next-Gen Marketing Cloud, positioned as a complete agentic marketing solution for personalizing customer moments across the lifecycle. Key capabilities include multi-channel engagement (email, mobile SMS/push, social, ads), Conversational Email powered by Agentforce, journey orchestration (Journey Builder), and deep integration with Salesforce Data 360 as the unified customer data foundation.
source.pcisecuritystandards-org.standards-2026
PCI Security Standards Council's standards page for PCI DSS 4.0.1, the current global benchmark for securing cardholder data. Applies to any entity that stores, processes, or transmits payment card data. Key scope-reduction controls include tokenization, point-to-point encryption (P2PE), hosted payment pages, and network segmentation to isolate the cardholder data environment (CDE).
source.oracle-com.cx-marketing-gartner-mq-customer-data-platforms-2026
Oracle's landing page for its 2026 Gartner Magic Quadrant for Customer Data Platforms Leader recognition. Oracle Unity Customer Data Platform is positioned as an enterprise CDP that unifies customer data from online and offline sources, applies AI/ML for predictive analytics and next-best-action, and activates audiences across Oracle CX applications and third-party destinations.
source.ftc-gov.business-guidance-privacy-security-gramm-leach-bliley-act-2026
FTC guidance hub for the Gramm-Leach-Bliley Act (GLBA), covering the Financial Privacy Rule, Safeguards Rule, and Pretexting Provisions. Financial institutions must provide privacy notices, allow opt-out of NPI sharing with non-affiliated third parties, and maintain a written information security program. 2023 amendment added mandatory data-breach reporting to the FTC, effective May 2024.
source.businesswire-com.news-home-20260128149689-uniphore-cdp-leader-2026
BusinessWire press release announcing Uniphore's inclusion as a Leader in the 2026 Gartner Magic Quadrant for Customer Data Platforms, published January 28, 2026. Uniphore is described as a Business AI company specializing in enterprise conversation intelligence, multimodal AI, and real-time customer interaction data.
org-dim.org-chart-type.matrixed-enterprise
Organizations where marketing, IT/data, and business units report through separate functional chains with shared accountability on cross-functional initiatives. CDP decisions require concurrent alignment from at least two leadership tracks (typically CMO/CDO and CTO/CIO) with different priorities, success metrics, and vendor preferences.
org-dim.operational-profile.contact-center-heavy
An organization whose customer interactions are substantially driven by high-volume inbound contact center operations: voice calls, chat, and video interactions at a scale that produces significant conversational data. Indicative threshold: >500 agents or >1M calls/month. This operational characteristic enables conversation-intelligence CDP use cases unavailable to organizations without material contact-center operations.
vendor.snowflake
Cloud data warehouse commonly used as the center of gravity in composable CDP stacks. Elastic compute, consumption-based pricing, no vendor-imposed data egress quotas.
vendor.databricks
Cloud data and AI platform combining lakehouse storage, Spark-native compute, and ML/AI tooling. Often the composable-stack center for organizations with heavy ML workloads.
use-case.win-back
Identifying customers who have churned or lapsed and delivering targeted re-engagement campaigns. Distinct from churn prevention — the customer has already defected; the signal is the cancellation or inactivity event itself.
constraint.snowflake-egress-cross-region
Snowflake charges for cross-region data transfer at approximately $20/TB in US regions and up to $140/TB in some Asia Pacific regions. Same-region transfers are free. The Snowflake Egress Cost Optimizer (ECO, launched April 2025) can reduce cross-region replication costs by up to 96% via intelligent caching.
constraint.pci-dss-scope-payment-card-data
Organizations that store, process, or transmit payment card cardholder data (CHD) — or whose systems could impact the security of CHD — enter the PCI DSS scope and must meet 12 requirements (PCI DSS 4.0.1). CDPs and CDWs that receive raw PANs, CVVs, or cardholder data are in-scope. Scope-reduction mechanisms (tokenization, P2PE, hosted payment pages, network segmentation) can remove the CDP/CDW from PCI scope by ensuring CHD never reaches them.
constraint.glba-nonpublic-personal-information-financial-services
Under the Gramm-Leach-Bliley Act (15 U.S.C. §§ 6801–6827), financial institutions must protect the confidentiality and security of customers' Nonpublic Personal Information (NPI) and provide opt-out rights before sharing NPI with non-affiliated third parties. CDPs used by financial institutions must classify data fields as NPI vs. non-NPI, gate third-party activation on NPI opt-out status, and implement a security program meeting the Safeguards Rule (amended 2021/2023).
constraint.aep-query-service-concurrent-sessions-10
Adobe Experience Platform Query Service limits each organization to 10 concurrent interactive (ad-hoc) query sessions. Additional sessions are rejected with a "Session Limit Reached" error. Scalable via the Ad Hoc Query Users add-on (+5 sessions per add-on pack).
constraint.aep-activation-external-audience-limit-20
Adobe Experience Platform limits the number of external audiences (Federated Audience Composition, custom uploads, Audience Composition) that can be activated to any single destination instance to 20. Applies per destination instance, not per destination type.
concept.composable-cdp
A customer data platform assembled from best-of-breed components — a cloud data warehouse as the data backbone, identity resolution tooling, reverse ETL or native connectors for activation, and a customer engagement platform for execution — rather than procured as a single packaged vendor system. Contrasts with packaged CDPs that maintain proprietary profile stores and managed activation.
source.macbach-com.insights-healthcare-marketing-benchmarks-2026-2026
Practitioner benchmark report on HIPAA compliance in healthcare marketing stacks, based on live platform data from 30-40 Macbach client accounts (Jan 2025–Mar 2026). Key findings: only 38% of audited stacks were fully compliant end-to-end; 84% lacked server-side Conversion API bridges; 71% had tracking pixels firing on post-form-fill pages containing patient data. Published April 2026.
source.experienceleague-adobe-com.en-docs-experience-platform-destinations-catalog-warehouse-snowflake-batch-2026
Official Adobe Experience League documentation for the AEP Snowflake Batch connection, covering the March 2026 table structure change from column-per-audience format (ups_<audience-id> columns) to a column-per-attribute row model with AUDIENCE_STATUS. Old format deprecated at end of June 2026. Includes V2 naming convention during transition, 7-day TTL, and Azure Private Link incompatibility. Page last updated May 28, 2026.
operational-task.configure-streaming-activation-destination
Connect a CDP to an external marketing or advertising platform so that audience membership changes are delivered in real time or on a scheduled basis.
archetype.healthcare-provider-hipaa-performance-marketing
A healthcare provider or health plan that needs paid-media performance marketing (patient acquisition, service-line promotion) but is blocked by HIPAA compliance concerns about routing behavioral event data — which may contain PHI — to non-BAA ad platforms (Google, Meta). HHS March 2024 tracking-technology guidance explicitly named this as a recognized architectural problem. The recommended path: replace standard non-BAA tracking pixels with a tracking-layer PHI suppression platform that de-identifies behavioral events before they reach any ad platform — enabling HIPAA-compliant paid-media analytics and activation.
recommendation.hipaa-tracking-layer-phi-suppression
For healthcare organizations (health systems, health plans, digital health companies handling PHI) blocked from standard performance marketing by HIPAA compliance concerns about routing behavioral event data to non-BAA ad platforms: deploy a tracking-layer PHI suppression platform between the website/app event stream and downstream ad destinations. The platform de-identifies behavioral events server-side before forwarding to Google Analytics, Meta Ads, and programmatic platforms — enabling paid-media acquisition measurement without routing PHI to non-compliant destinations. Vendor-neutral architectural description: pattern.fail-fast-within-compliance. Vendor implementations: vendor.freshpaint.
org-dim.industry.digital-health
Digital health companies — telehealth platforms, consumer health apps, wearables, health information technology startups, and clinically-integrated digital care tools — that apply technology to healthcare delivery or wellness. Distinct from traditional healthcare providers (health systems, hospitals, specialty practices) in organizational profile: technology-first structure, startup-speed iteration, mobile-first data collection, engineering-led culture. CDP architecture varies significantly based on whether the organization handles PHI and is subject to HIPAA.
use-case.hipaa-safe-performance-marketing
A healthcare organization running paid-media performance marketing (patient or member acquisition, service-line promotion) with HIPAA-compliant tracking infrastructure that prevents PHI from reaching non-BAA ad platforms (Google, Meta). The architectural pattern: replace standard non-BAA tracking pixels with a tracking-layer PHI suppression platform that de-identifies behavioral events server-side before forwarding to ad destinations — enabling full paid-media analytics without routing PHI to non-compliant destinations.