Customer Uptake & Channel
Report type key: recommender_customer_markdown
Configuration type: Interactions
Group: Interactions
Pseudonymous customer uptake, cohorts, repeat behavior, channel preferences, switching, journeys, and experiments.
How to get it
- Create and save an Interactions configuration.
- Enable: Interactions family customers.
- Run the job and wait until completed.
- Open Reports (
/reports), select that job, and choose this type.
Markdown viewer: /markdown/JOB_ID/recommender_customer_markdown
API: GET /reports/JOB_ID/recommender_customer_markdown
The customers family was off or identity coverage was too low. Check job logs and Analysis & LLM.
What it contains
- Tracked vs visible vs suppressed pseudonymous customers (
minimum_cell_size, default 10). - Offer/response timeline and ratios for identified customers only.
- Sample journeys (offers presented → responded → accepted, with channels).
- Daily interaction movement.
- Behavioral cohorts and repeat behavior (prefer exact
$groupcounts when present). - Channel preference scorecards, cross-channel journeys, switching, preference evidence.
Customer identifiers are hashed. Do not expect raw PII.
How to read
Use this for cohort mix and channel journeys, not ledger ARPU. Low-evidence rows disappear by design.
Companion artifacts
| Key | Artifact |
|---|---|
recommender_customer_markdown | Markdown report |
recommender_customer_notebook | Jupyter notebook |
recommender_customer_results | JSON results |
recommender_customer_graphs | Cohort / channel graphs |
Markdown, notebook, and graphs for the same family usually land together. JSON feeds Analysis Insights.
Related
Last updated on