The person you measure is not the person you activate
The identity graph decides whom to activate now. Analytics stitching rebuilds who it was. They are two answers, and each is designed on its own.
Generic diagram. It does not reproduce a client architecture.
Experience Platform and Customer Journey Analytics are often expected to count the same people. There is no reason they should. The identity graph joins identifiers to activate someone right now, with rules that protect the profile. Analytics stitching reassigns events to a person to measure what already happened, with its own person ID.
Both fail in the same way when the identifier does not represent a person: a shared account, a household email or a placeholder value. The architecture decision is not which tool to use. It is which identifier is a person in each one, and how that is checked.
Production signals
- The people in reports look nothing like the active customers.
- A household or company account ID acts as the person ID.
- Each dataset in the connection uses a different identifier.
- Analytics is expected to match the profile, and nobody explains what each number is for.
Recommended architecture
Choose as person ID an identifier that represents one person, the same across every dataset in the connection, and validate the people against real customers.
Explain in writing which question each tool answers and why their people counts do not match.
When not to apply it
- Do not use the account as a person when the account is what you analyse: it belongs in a dimension.
- Do not force the numbers to match at the cost of joining different people.
Try it in a simulator
Interactive scenarios where this pattern fails and gets fixed.
- CJA
How many people visited us?
How many people and what conversion rate does CJA report with each stitching method?
Open the simulator → - CJA
Who is one person to CJA?
Who is "one person" to CJA, depending on the field you choose as person ID?
Open the simulator → - AEP · CJA
Graph or stitching
Who is the customer for AEP when activating, and who was she for CJA when measuring?
Open the simulator →
Explore this decision
- Data governance reaches analysis too
What data usage labels do in Customer Journey Analytics, what enabled policies block, and what happens to data views and exports.
- Fix the data or fix the report?
A misclassified channel can be fixed at source, during ingestion or in CJA. Compare who receives the correction and what happens to historical data.
- The purchase is in AEP. Where does it go missing in CJA?
Trace a purchase missing from CJA through the dataset, connection, historical import, data view and filters before loading the data again.