Model Tab
Apoorva Wate avatar
Written by Apoorva Wate
Updated over a week ago

The model tab provides a crucial interface for assessing the efficacy of your marketing strategies through sophisticated analytics and comparative model evaluations.

NRMSE: The Normalized Root Mean Square Error is also referred to as the prediction error. It measures the square root of the average squared differences between predicted and actual outcomes.

MAPE: The Mean Absolute Percentage Error for experiments is activated when calibrating and is referred to as the calibration error.

Actual Vs Predicted

This chart shows the actual KPI (e.g. revenue) vs. the predicted value from the model. You can analyze the trend line here to see how accurately the model predicted the revenue.

Saturation & Ad Stock Curves

Ad Stock chart represents the carryover effect of each media variable (paid & organic). A common hypothesis of adstock is that offline channels have stronger the carryover.

Saturation Curve -The saturation curve shows how the incremental impact of a marketing effort decreases as the level of that effort increases. Saturation points represent the point at which additional investment in a particular marketing activity, such as advertising or promotions, will no longer result in a proportional increase in sales or revenue.

Immediate Vs Carryover

This chart breaks down the impact of media spending on revenue into two distinct categories: immediate and carryover contributions. It allows you to discern when the majority of a campaign's effects are realized.

  1. Immediate Contribution: This refers to the percentage of direct effect of media spending within the same period it is spent. For Lifesight it the the same week. For instance, if an advertisement runs in week 1 of June, the immediate contribution is the revenue directly generated from this advertisement in week one of June itself.

  2. Carryover Contribution: This involves the percentage residual effects of past media spending that continue to influence revenue beyond the initial period of the spend. This phenomenon is also known as adstock. Conceptually, carryover effects are similar to brand equity metrics such as ad recall or campaign

Did this answer your question?