What is AEP (Adobe Experience Platform)?
AEP captures the customer journey. Data from many sources is stitched together through a schema, which builds an identity graph unique to each customer. AEP has a data lake into which data from the various sources is streamed; that data is used to build each customer's profile.
In short, AEP is real-time customer profiles plus AI and machine learning plus an open ecosystem.
Which data sets does AEP define?
- Attributes — characteristics such as the customer's name, email and gender
- Identities — unique identifiers such as the ECID (Experience Cloud ID), email, membership ID and phone number
- Segments — categories such as online shoppers, gender or location. Segments can, for example, be exported for use in an email campaign.
- Behaviors — actions such as logging in to the website, installing an application or adding an item to the cart
What does AEP solve?
- Disconnected identities
- Slow and vulnerable data transfer
- AI and ML working in silos, where extracting data is hard
- Data governance (CCPA, GDPR and so on) that is usually not enforced
- Features spread across many places instead of one
Major AEP capabilities
- Actionable, real-time, intelligent customer profiles
- Enriched data and deeper insight through AI, ML and SQL queries
- Open, composable components, such as open APIs
- Better delivery
- Privacy and data protection: a privacy framework, consent and security
Integration with other Adobe applications
Adobe Experience Cloud applications — the marketing, analytics, advertising and commerce clouds — integrate easily with AEP, and customer attributes flow into AEP from each of them. For example:
- Adobe Analytics sends data as soon as each data point is captured.
- Adobe Target sends decisions made and content presented.
- Audience Manager sends traits and audiences.
- Adobe Campaign sends profile and event data.
AEP uses Launch and the Web SDK to bring data in directly from these applications.
How do AEM and AEM Forms use AEP?
AEM can use AEP data to personalize content on pages and forms. See also personalization in AEM with ContextHub.
Implementation phases and roles
- Plan — leads and an enterprise architect plan against the business goals and document KPIs.
- Implement — data architects and engineers build the data lake (models, schemas), make it available, and check data integrity with the query service.
- Use — marketers, data analysts, data scientists (through the query service) and application developers work in the UI and integrate with other Adobe applications such as Campaign, Target and Analytics.
- Grow — results are reported back to the initial team (enterprise architects, company leads).
Basic architecture
Data is ingested into the AEP data lake — from third-party ETL, ERP or sales systems, or from Adobe applications through Launch. It lands as batches and files, and anything streamed in also goes through the Experience Platform pipeline. Ingested data quickly reaches the identity graph and the profile store.
Controls native to AEP
- Access control — permissions on data and users
- Data governance — keeps data use compliant
- Experience Data Model (XDM) — a common data model that can be extended as needed
- Query service — SQL access to the data, with connectors so other SQL tools can use it
- Data Science Workspace — lets data scientists build, train and deploy models
- Intelligent services — prebuilt models such as Attribution AI and Customer AI, configured to run on your data
- Segmentation — for categorization, on both streaming and batch data
Application services
- Customer Journey Analytics — combines data from every channel, with Analysis Workspace on top of AEP to visualize and explore everything in the data lake
- Real-Time Customer Data Platform (CDP) — rich real-time profiles, actionable insight, data governance and segmentation
- Journey Orchestration — triggered interactions such as registration confirmations and location-based messages
- Intelligent Services — AI and ML to predict customer behavior
- Offer Decisioning — build offers, apply decisioning rules and deliver the right offer
Use cases
- Real-time customer data platform — stitch known and unknown data together to activate customer profiles.
- Customer journey intelligence — data-driven methods, AI and ML for real-time decisions and actions.
- Delivery and cross-channel experience — consistent, personalized experiences on every channel by combining the platform with Experience Cloud products.
- Customer experience application development — an open, extensible platform with low-latency access to profiles, decisions and insights for building new applications.