TABLE OF CONTENTS
1. Overview
This release introduces enhancements to the Simulator and Optimiser to improve long‑term modelling accuracy and simplify constraint configuration. Users can now include historical carryover effects in Simulator baselines and apply constraints consistently across multiple sub‑dimensions with a single action. In addition, several issues across Optimisation, Media Insights, and Activities have been resolved to improve reliability, clarity, and data correctness.
2. Historical Carryover Option in Simulator Baseline
What it is
The Historical Carryover option enables historical media spend effects to influence Simulator outputs. When enabled, the Simulator baseline and scenario results account for the lingering impact of past spends, rather than relying only on the spends defined within the simulation window.
This feature applies to long-term models and affects both baseline and scenario simulations.
Why it exists
Previously, Simulator outputs were calculated using only the spends defined for the simulation period, which could result in outputs that did not fully reflect real-world conditions. Media effects often persist beyond the original spend period, and excluding these effects could lead to unrealistic or understated results.
This feature ensures Simulator outputs are:
- More realistic and pragmatic
- Better aligned with expected real-world performance
- Considerate of historical investments already made
What it does
When the Historical Carryover toggle is enabled:
- The system appends historical spend data to the simulation input before running calculations.
- Effects generated by these historical spends are carried forward into the simulation time window.
- The Simulator input remains unchanged from a user perspective, except for a flag that enables or disables historical carryover.
Handling gaps between modelling and simulation periods
If there is a time gap between the end of the modelling period and the start of the simulation window, this gap is automatically populated using historical spend data from the previous year.
Example:
If a model’s modelling period ends in H2 2024 and a simulation is run for H2 2025, the input data is constructed as follows:
- Historical spends: H1 2024, H2 2024
- Gap period: H1 2025 (populated using H1 2024 spends)
- Simulation spends: H2 2025
Resulting input structure:
[Historical spends] – [Gap period] – [Simulation spends]
[H1 2024, H2 2024] – [H1 2024] – [H2 2025]
This ensures continuity of effects even when simulation periods do not immediately follow the modelling window.
Why this matters
- Produces more credible and context-aware Simulator baselines
- Improves interpretation of simulated outcomes
- Better reflects the cumulative impact of historical media activity
Limitations / Caveats
- For gap periods, historical spends are populated using year-back data.
- Results may differ from earlier simulations where only the simulation-period spends were considered.
Users can now choose whether to include historical carryover effects when viewing Simulator baselines for long‑term models.
When enabled, the Simulator baseline incorporates long‑term historical effects generated by AI, providing additional context for interpreting simulation outcomes.
How to Access it:
Please check with the Admin to enable it for you
What this means for you
- More realistic Simulator baselines for long‑term models
- Clear visibility into historical effects influencing current simulations
- Improved interpretation of long‑term outcomes
Applies to: Long‑Term models only and Non‑long‑term models are not affected.
What it is
The Historical Carryover option enables historical media spend effects to influence Simulator outputs. When enabled, the Simulator baseline and scenario results account for the lingering impact of past spends, rather than relying only on the spends defined within the simulation window.
This feature applies to long-term models and affects both baseline and scenario simulations.
Why it exists
Previously, Simulator outputs were calculated using only the spends defined for the simulation period, which could result in outputs that did not fully reflect real-world conditions. Media effects often persist beyond the original spend period, and excluding these effects could lead to unrealistic or understated results.
This feature ensures Simulator outputs are:
- More realistic and pragmatic
- Better aligned with expected real-world performance
- Considerate of historical investments already made
What it does
When the Historical Carryover toggle is enabled:
- The system appends historical spend data to the simulation input before running calculations.
- Effects generated by these historical spends are carried forward into the simulation time window.
- The Simulator input remains unchanged from a user perspective, except for a flag that enables or disables historical carryover.
Handling gaps between modelling and simulation periods
If there is a time gap between the end of the modelling period and the start of the simulation window, this gap is automatically populated using historical spend data from the previous year.
Example:
If a model’s modelling period ends in H2 2024 and a simulation is run for H2 2025, the input data is constructed as follows:
- Historical spends: H1 2024, H2 2024
- Gap period: H1 2025 (populated using H1 2024 spends)
- Simulation spends: H2 2025
Resulting input structure:
[Historical spends] – [Gap period] – [Simulation spends]
[H1 2024, H2 2024] – [H1 2024] – [H2 2025]
This ensures continuity of effects even when simulation periods do not immediately follow the modelling window.
Why this matters
- Produces more credible and context-aware Simulator baselines
- Improves interpretation of simulated outcomes
- Better reflects the cumulative impact of historical media activity
Limitations / Caveats
- For gap periods, historical spends are populated using year-back data.
- Results may differ from earlier simulations where only the simulation-period spends were considered.