MMM Attributions Table (BETA)

mmm_attributions_table

The MMM Attributions table tracks predicted KPI contribution by modeled group over time. One row per model run per date per group per set type. Rows split the model prediction by group.

📋
Table Guide
What You Can Answer With This Table

How much KPI did the model attribute to each modeled group over time? — use date, group_name, kpi

How did attributed KPI compare with paid spend by modeled group? — use group_name, spend, kpi

How did non-paid controller inputs contribute? — use group_name, controller_value, kpi

Before You Query

This table uses date, not event_date, for time-based queries.

For multiplatform models, the data is broken out by sales platform: each date and set type has one row per sales platform plus an additional _all row that aggregates across all of them.

For multiplatform models, always filter platform = _all when reading spend or ROAS. Per-platform rows have spend = 0, and the _all row already holds the channel total kpi (the sum across platforms), so summing _all plus every per-platform row double-counts kpi.

Key Relationships

Table Join Key Use This Join To
MMM Models model_id Get the configuration and accuracy scores for the model run
MMM Marginal Lifts model_id, group_id Add each modeled group's marginal return alongside its attributed contribution
MMM Campaign Groups model_id, group_id See which campaigns make up the modeled group and map it back to Custom Categories

When to Use a Different Table

Use MMM Predictions when you need total predicted and actual KPI across all modeled groups. MMM Attributions breaks the model prediction down by group.

Use MMM Marginal Lifts when you need each modeled group's marginal return on the next spend increment. MMM Attributions stores total attributed contribution by group.

View the full Triple Whale Data Ontology →

Table	Join Key	Use This Join To
MMM Models	`model_id`	Get the configuration and accuracy scores for the model run
MMM Marginal Lifts	`model_id`, `group_id`	Add each modeled group's marginal return alongside its attributed contribution
MMM Campaign Groups	`model_id`, `group_id`	See which campaigns make up the modeled group and map it back to Custom Categories

Dimensions

Dimensions are immutable properties that can be used for grouping data.

Title	ID	Type	Description
Attribution Created Date	`created_at`	timestamp	The time at which the attribution row was written to the database. Formatted according to the ISO 8601 international standard. Example value: `2024-01-22 06:15:00`
Date	`date`	date	The date the attribution applies to, at the model's `time_unit` granularity (daily, weekly, or monthly). Example value: `2024-01-15`
Group ID	`group_id`	string	The unique identifier of the modeled group. When MMM uses Custom Categories, this corresponds to the subcategory ID used for model grouping. Joins to MMM Campaign Groups to see the underlying campaigns. Example value: `c41d8e2f-7a90-4b3c-9d12-5e6f0a1b2c34`
Group Name	`group_name`	string	The display name of the modeled group. When MMM uses Custom Categories, this is the subcategory name. Example values: `Meta Prospecting`, `Google PMAX`, `Email`, `Discounts`
Model ID	`model_id`	string	The unique identifier of the model run that produced the attribution. Joins to MMM Models. Example value: `9f2c1a7e-3b44-4c9a-8e21-7d5f0b2a1c33`
Platform	`platform`	string	The sales platform associated with the attribution. `_all` represents the aggregate values across all sales platforms, while platform-specific values represent attribution broken down by individual platform. Example values: `_all`, `shopify`, `amazon`, `tiktok-shops`
Set Type	`set_type`	string	The dataset split the row belongs to: training data (in-sample) or test data (out-of-sample, held-out). Possible values: `test`, `train`
Settings ID	`settings_id`	string	The unique identifier of the model configuration. All runs of the same model share one `settings_id`. Example value: `3a7b2c9d-1e44-4f8a-9c12-6b0d8e2f4a51`

Measures

Measures are numeric fields that can be aggregated and/or combined to calculate new metrics.

Title	ID	Type	Description
Controller Value	`controller_value`	numeric	For non-paid (controller) channels, the driver's input value, such as email count or discount percentage. `0` for paid media channels.
Attributed KPI	`kpi`	numeric	The KPI value the model attributes to the modeled group for the date.
Attributed KPI Lower Bound	`lower_ci`	numeric	The lower bound of the attribution's confidence interval (85% HDI). `NULL` when no interval value is available.
Spend	`spend`	numeric	The paid media spend for the modeled group on the date. `0` for non-paid (controller) channels. By default shown in the shop's currency.
Attributed KPI Upper Bound	`upper_ci`	numeric	The upper bound of the attribution's confidence interval (85% HDI). `NULL` when no interval value is available.