Marketing Mix (Adstock & Saturation)
Report type key: recommender_marketing_mix_markdown
Configuration type: Interactions
Group: Interactions
Adstock and saturation transforms on campaign/channel exposure versus response.
How to get it
- Create and save an Interactions configuration.
- Enable: Interactions family marketing_mix (off by default).
- Run the job and wait until completed.
- Open Reports (
/reports), select that job, and choose this type.
Markdown viewer: /markdown/JOB_ID/recommender_marketing_mix_markdown
API: GET /reports/JOB_ID/recommender_marketing_mix_markdown
Enable marketing_mix on Analysis & LLM (off by default). If still missing, exposure/response series were too short or the step was skipped.
What it contains
Diagnostic adstock (geometric or Weibull) and saturation (Hill or Michaelis–Menten) transforms of observed campaign/channel exposure versus attributed acceptance.
Reports carryover half-life, peak lag, and observed elasticity along the curve. Optional overlay from a Financial job (overlay_financial_job_id) joins spend series — it still does not invent incremental ROAS.
How to read
These are not a causal media-mix model and do not replace declared A/B uplift. Use Campaign & predictor for verified uptake rates and Financial campaigns for mapped ROI.
Companion artifacts
| Key | Artifact |
|---|---|
recommender_marketing_mix_markdown | Markdown report |
recommender_marketing_mix_notebook | Jupyter notebook |
recommender_marketing_mix_results | JSON results |
recommender_marketing_mix_graphs | Graphs |
Markdown, notebook, and graphs for the same family usually land together. JSON feeds Analysis Insights.