MMM - Scenario Planner

Updated by Eddie Chou

Overview

The Rockerbox Scenario Planner allows you to explore different marketing scenarios, where you may be looking to expand, contract or reallocate budget subject to different constraints. The goal of scenario planning use the Marketing Mix Model to understand the impact of different budget allocations on your marketing performance.

The Scenario Planner takes the user through several steps:

  • Setup Forecast Period.
  • Setup Budget Period
  • Establish Baseline
  • Choose Scenario
  • Setup Constraints

1. Setup Forecast Period

Select the timeframe you want to use for forecasting. Select daily, weekly, monthly, quarterly, or a custom timeframe. Adjusting the projection period helps in different phases of budget planning and optimization.

2. Setup Budget Period

Defines the granularity of budget allocation (daily, weekly, monthly).

We recommend setting the budget period that best matches how you plan budgets. For example, if you plan budgets on a monthly horizon, then select monthly.

3. Establish Baseline

This setting serves two purposes:

  • It establishes the starting point for budget constraints.
  • For the purposes of forecasting "baseline" impact for all non-paid marketing (ex: Promotions, Discounts), it will use the average from the selected baseline period.

We recommend selecting a Baseline that most closely matches the period you plan to forecast against. For some customers, this is the most recent period. For others, the previous year's period is more appropriate.

Note: you can only select Baseline Periods from the model Training Period.

4. Choose Scenario

Scenario Planner offers four scenarios:

  • Optimize channel mix to maximize revenue.
    • Select this option to maximize your revenue.
    • Scenario Planner will propose a channel mix based on your budget in order to maximize revenue.
  • Reduce [Daily/Weekly/Monthly] Budget
    • If your goal is to reduce overall budget, select this option to see which channels to cut and/or reduce spend.
    • This option will reflect the selected budget period from step 2.
  • Expand [Daily/Weekly/Monthly] Budget
    • If your goal is to increase overall budget, select this option to see which channels to increase spend.
    • This option will reflect the selected budget period from step 2.
  • Optimize to ROAS Target
    • Select this option to target a specific ROAS for your channel mix.
    • Scenario Planner will propose a channel mix based on your budget in order to hit your specific target. In other words, it will aim to reduce spend for channels that underperform while increasing spend that overperform.

5. Setup Constraints

In this section, the user can select a broad set of constraints that can then be further refined per channel. The goal of these constraints is to limit the model's budget change proposals based on your tolerance for changes. The setting defaults to a 30% change for all channels.

You can adjust the broad set of constraints from 15%, 30%, 50%, and 100%. Additionally, you can override this constraint at the channel-level by selecting + Refine Constraints per Channel.

The model will limit proposed budgets to these constraints.

6. Generate Forecast

Once all settings have been configured, the final step is to click Generate Forecast to review the model's proposed budget allocation and its projected impact on key metrics. The output will include:

  • Monthly Budget: The proposed budget for each channel compared to the baseline.
  • Projected Monthly Revenue: The estimated revenue for each channel based on the new budget.
  • ROAS (Return on Ad Spend): The efficiency of each channel, showing both the overall and marginal ROAS.

The results are displayed in a table, highlighting the percentage change in budget and revenue for each channel. This allows you to evaluate which channels are expected to perform better or worse based on the proposed changes. Additionally, it offers insights into seasonal effects on revenue by comparing baseline and predicted marketing revenue to estimate the total predicted revenue.

The goal is to provide a clear view of how the adjustments will impact overall marketing performance, enabling data-driven decision-making for budget reallocation.


How did we do?