Customer Journey Analytics (CJA) by Adobe brings a powerful way to unify customer data across channels. But to make the most of it, you need to master three key components: datasets, connections, and data views.
🔹 Dataset: the raw foundation
A dataset is a structured data table ingested into Adobe Experience Platform. You can have:
- event datasets (page views, purchases),
- profile datasets (CRM info),
- lookup datasets (products, campaigns, etc.).
Each dataset follows a specific schema (XDM – Experience Data Model) and has its own identifiers and fields.
🔹 Connection: merging the sources
A connection allows you to join multiple datasets into one unified model. It’s where you define:
- which datasets are included,
- identity stitching across sources,
- retention periods and data types (event, lookup, profile).
💡 You can merge web, app, CRM, and even offline sources — building a truly cross-channel customer view.
🔹 Data View: the business-ready view
Once your connection is set, you can create one or more data views, each tailored for a specific purpose:
- filter by region, device, product line, etc.
- define attribution model and analysis window,
- expose only relevant dimensions and metrics,
- choose your reporting granularity (user, session, event).
Examples:
- A marketing view for mobile traffic in France,
- A global data view focused on post-purchase journeys,
- A paid media campaign analysis view.
🎯 Why it matters
This 3-layer model gives you clean separation between:
- collection (datasets),
- modeling (connection),
- and analysis (data views).
It provides clarity and flexibility for analysts, marketers, and product teams alike — without compromising data integrity.
👉 TL;DR: CJA pushes you to think before you analyze. That’s not a limitation — that’s good data hygiene.