Models
The model functionality in Mix Modeler allows you to configure, train, and score AI/ML models specific to your business objectives and supported by AI-driven transfer learning between multitouch attribution and marketing mix modeling.
The models are based on the harmonized data that you create as part of the Mix Modeler application workflow.
A model in Mix Modeler is a machine learning model employed to measure and / or predict a specified outcome based on a marketer’s investments. Marketing touchpoints and summary-level data can be used as an input. Mix Modeler allows you to create variants of models based on different sets of variables, dimensions, and outcomes, such as revenues, units sold, leads.
A model requires:
- one conversion,
- one or more marketing touchpoints (channels) comprised of summary-level data, marketing touchpoint data (event data) or both,
- a configurable lookback window for
- a configurable training window.
A model can optionally include:
- external factors,
- internal factors,
- so-called ‘priors’ (probability distribution representing knowledge or uncertainty of data prior or before observing that data), which indexes prior conversions by channel,
- spend share, which uses relative spend share as a proxy when marketing data is sparse.
Create a model
To create a model, use the Mix Modeler step-by-step guided model configuration flow available when you select Open model canvas. See Create a model for more details.
Manage models
To view a table of your current models, in the Mix Modeler interface:
-
Select
-
You see a table of the current models.
The table columns specify details about the model.
table 0-row-2 1-row-2 2-row-2 3-row-2 4-row-2 5-row-2 6-row-2 layout-auto Column name Details Name Name of the model Description Description of the model Conversion event The conversion you have selected for the model. Run frequency The running frequency of training the model. Last run The date and time of the last training of the model. Status The status of the last run of the training of the model.
● Success
● Training issue
● Awaiting training
● Failed
● _ (when a last run is in progeress) -
To change the columns displayed for the list, select
View details of a model
To view more details of a model:
- Select
Model insights
To view insights of a model, in the Mix Modeler interface:
-
Select
-
Select the name of a model with a Last run status of ● Success from the Models table. Model insights is only available on successfully trained models.
-
From the context menu, select Model Insights. You are redirected to Model Insights.
Re-score
To re-score a model, in the Mix Modeler interface:
-
Select
-
Select the name of a model with a Last run status of ● Success from the Models table. Re-score is only available on successfully trained models.
-
From the context menu, select Re-score. It may take a few minutes to show an updated status for the model.
Delete a model
To delete a model:
-
Select the name of the model that you want to delete.
-
From the context menu, select Delete to delete the model.
note warning WARNING The model is deleted immediately.