# Customer Data Canonical URL: https://babadook.tech/sub-modules/customer-data Entity: sub-modules Content hash: 30e309db83e32bf0e1004613565d046e8648ced61cd86116649d44d0372dbd51 Unified customer profiles with identity resolution, segmentation, and recommendation engine. Customer Data Platform capabilities including identity resolution, segmentation, recommendations, search, and loyalty programs for unified customer intelligence. Customer Data creates a single, actionable view of every customer across all touchpoints. ## One Customer, One Profile, Every Channel subtitle: Unify fragmented customer data into actionable profiles with real-time identity resolution across web, mobile, email, and in-store. ### Identity Resolution desc: Deterministic and probabilistic matching across devices, channels, and sessions. ### Real-Time Segmentation desc: Dynamic audience segments that update in real-time as customer behavior changes. ### Recommendation Engine desc: ML-powered content, product, and experience recommendations personalized per visitor. ### Unified Search desc: Semantic search across all customer data, content, and interactions. ### Profile resolution value: <200ms unit: ms ### Per profile value: 50+ unit: attributes ### Segment updates value: Real-time ## unified-customer-intelligence Unified Customer Intelligence Identity resolution combines deterministic matching (email, phone, login) with probabilistic signals (device fingerprints, behavioral patterns, IP clustering) to build a single customer profile across all touchpoints. Profiles are resolved in under 200ms at request time. Segments are computed in real-time as customer behavior changes. A visitor who adds an item to cart immediately enters the 'active-shopper' segment. When they abandon, they transition to 'cart-abandoner' with the associated re-engagement campaigns triggered automatically. The recommendation engine uses collaborative filtering, content-based filtering, and contextual bandits to surface the right content, product, or experience for each visitor. Recommendations improve over time as the models learn from implicit and explicit feedback signals. ### data Viz ## Customer Data Platform Specifications - **Profile Resolution:** <200ms deterministic + probabilistic matching - **Profile Attributes:** 50+ attributes per profile with custom extensions - **Segment Updates:** Real-time with sub-second propagation - **Identity Graph:** Cross-device, cross-channel, cross-session - **Recommendation Models:** Collaborative filtering, content-based, contextual bandits - **Search:** Semantic search with typo tolerance and faceted filtering - **Data Sources:** Web, mobile, email, POS, CRM, and custom event streams ## faq section ### Section 1 How does identity resolution handle anonymous visitors? Anonymous visitors get a temporary profile that merges with their identified profile when they log in or provide identifying information. All behavioral data from the anonymous session transfers to the identified profile seamlessly. ### Section 2 What happens when profiles conflict? When two profiles are identified as the same person, DXP CDP merges them using configurable conflict resolution rules. You define which data source wins for each attribute type — most recent, most trusted source, or manual review. ### Section 3 How are segments different from static lists? Segments are dynamic — they update in real-time as customer behavior changes. A customer enters and exits segments automatically based on rules you define. Static lists are snapshots that don't change unless manually updated. ### Section 4 Can the recommendation engine work without historical data? Yes. New deployments use popularity-based and content-based recommendations until sufficient behavioral data is collected. The system transitions to collaborative filtering and contextual bandits as the data volume grows. ## features search: true loyalty: true identity: true segmentation: true recommendations: true ## Relationships ### products - DXP — https://babadook.tech/products/dxp ### modules - Customer Search — https://babadook.tech/modules/customer-search - Identity Resolution — https://babadook.tech/modules/identity-resolution - Loyalty & Rewards — https://babadook.tech/modules/loyalty-rewards - Recommendation Engine — https://babadook.tech/modules/recommendation-engine - Segmentation Engine — https://babadook.tech/modules/segmentation-engine ### deployment types - Cloud (SaaS) - Self-Hosted