What is the Difference Between Optimiser and Simulator in the Prediction Module?

Modified on Tue, 9 Sep at 8:54 AM

TABLE OF CONTENTS

Overview
Why This Matters?
Detailed Comparison
When to Use Optimiser?
When to Use Simulator?
Best Practices
Summary

Overview

The Prediction module in the Insights Platform is designed to help marketers and planners make data-driven investment decisions. It offers two powerful capabilities:

Optimiser – For finding the most efficient allocation of budgets to maximise returns or achieve specific targets.
Simulator – For exploring “what-if” scenarios by adjusting media and non-media drivers to understand potential outcomes.

Although both tools support planning, they serve different purposes and are used in different contexts.

Why This Matters?

Choosing the right tool ensures you:

Allocate budgets effectively.
Test strategies before committing spend.
Balance precision with flexibility in planning.

Detailed Comparison

Aspect	Optimiser	Simulator
Primary Purpose	Determine the best allocation of spend to maximise ROI or meet a target KPI.	Explore hypothetical scenarios to understand the impact of changes.
Typical Use Case	“How should I allocate my £500,000 budget for maximum sales?”	“What happens if I increase Paid Search by 10% and reduce TV by 5%?”
Inputs Required	- Budget amount (for investment optimisation) - Target KPI (for target optimisation) - Optional constraints (e.g., minimum spend per channel)	- Flexible adjustments to media drivers (e.g., spend by channel) - Non-media drivers (e.g., pricing, promotions)
Outputs Delivered	- Optimal budget allocation - ROI, profit, and sales forecasts (with confidence intervals)	- Scenario outcomes showing impact on sales, profit, and other KPIs
Approach	Algorithm-driven: uses advanced models to find the most efficient allocation within defined constraints.	User-driven: you create and test scenarios manually.
Constraints	Supports detailed constraints at media, channel, or publisher level.	No hard constraints; full flexibility to adjust inputs.
Granularity	Portfolio or model-level optimisation with drill-down capability.	Weekly evolution and campaign/channel-level scenario editing.
Save & Activate	- Save optimisations. - Activate and track performance in Media Insights.	- Save simulations for planning iterations (detailed results only).

When to Use Optimiser?

Choose Optimiser when:

You have a fixed budget and want the system to allocate it for maximum ROI, sales, or profit.
You have a specific KPI target and need to know the minimum spend required to achieve it.
You require constraint-based optimisation, such as:
- Spend at least £1,000 on Paid TV.
- Cap Paid Search at £2,000.
You want system-suggested alternatives alongside your plan.
You plan to activate recommendations and later compare actual vs predicted performance in Media Insights.

Think of Optimiser as your “best plan generator” under defined rules.

When to Use Simulator?

Choose Simulator when:

You want to explore “what-if” scenarios without rigid constraints.
You need to adjust both media and non-media drivers (e.g., pricing, promotions).
You require weekly-level insights or custom scenario breakdowns.
You want to iterate quickly with high-level results before requesting detailed outputs.

Think of Simulator as your “strategy lab” for experimentation.

Best Practices

Choose your starting point based on certainty
- Start with Optimiser when a budget/target is fixed and you need the most efficient plan under constraints.
- Start with Simulator when assumptions are still fluid, or when non‑media factors (price, promos, competitor moves) need testing.
Pair them for resilience: Run Simulator to stress‑test any optimised plan against plausible “what‑ifs” and confirm it holds up across scenarios (e.g., +10% competitor spend, -5% TV availability).
Reflect real‑world constraints: Encode minimums/caps in Optimiser; use Simulator for changes you cannot express as constraints (timing shifts, weekly profiles, external shocks).
Iterate fast, then go deep: Use Simulator’s high‑level results for quick iteration; only generate detailed results when a scenario merits further analysis.
Measure and learn: Activate the chosen optimisation and compare predicted vs actuals in Media Insights to refine future plans.

Summary

Optimiser = Best allocation under defined rules.
Simulator = Flexible experimentation for future planning.

Both tools complement each other and, when used together, provide a robust framework for strategic media planning.