MARKETING & STRATEGY
Marketing ROI simulation: choose the audience size before you spend
Marketing ROI simulation with TIMi forecasts the return of a campaign for different audience sizes, by combining the lift of a propensity model with your costs and margins, so you decide how many customers to contact before spending the budget.
ROI by audience size
See profit change as the contacted audience grows.
Costs and margins included
Cost per contact and margin per response enter the calculation.
Lift you can read
The ranking behind the simulation is explained variable by variable.
HOW IT WORKS
How marketing ROI simulation works.
Three steps, in one engine, on a standard server with no cluster.
1. Prepare the data
Anatella gathers past campaigns, responses and customer data, then structures them into variables.
2. Rank the customers
Modeler builds a propensity model and measures its lift on a test set: how many responders are found in the top 10%, 20%, and so on.
3. Simulate the return
Anatella combines the lift curve with the cost per contact and the margin per response, to compute the expected profit for each audience size.
Anatella
ModelerGETTING IT RIGHT
Choices that shape an ROI simulation.
The curve is a forecast on past behavior, not a promise.
The simulation is only as good as the ranking
It rests on the lift measured on a test set. A weak ranking gives a flat curve and a low ROI, which is useful to know before launching.
Costs and margins come from your teams
TIMi uses them as inputs and shows how the best audience size moves when they change.
Compare with actual results
Reality will differ from the curve. Comparing it with the results of each campaign is how the estimates get calibrated.
EXPLAINABILITY
A forecast you can challenge.
The audience size you choose depends on the lift of the model. A white-box model shows which characteristics put customers at the top of the ranking, so teams can check that the simulated response is plausible before committing the budget.
For marketing teams
Understand who sits at the top of the ranking and why.
For finance
Costs, margins and response rates are explicit inputs of the ROI calculation.
For model owners
Modeler’s built-in Drift Report shows when a model loses accuracy, so you update it when needed instead of on a fixed calendar.
DATA
What data a campaign simulation uses.
Anatella connects to more than 80 sources, so this data can be gathered without prior custom development.
Past campaigns
Audiences contacted and measured responses.
Cost per contact
Channel cost for each contact.
Margin per response
Margin generated by each purchase.
Customer data
Profile and purchase history.
PROOF
From ranking to net euros.
We do not have a published case study dedicated to campaign ROI simulation. The reference below is indirect: it applies the same logic of turning a model ranking into a financial result.
Recovery costs halved, +€1.7M net
Not a marketing campaign. A predictive ranking applied to recovery, with the result measured in net euros. It illustrates the logic of converting the lift of a model into a financial outcome.
FAQ
Common questions.
Is this a forecast of actual sales?
It is an estimate of the return by audience size, based on the lift of the model on past data. Actual results depend on the offer, the timing and the market, and should be compared with the simulation after each campaign.
Can we simulate several channels?
Yes, if each channel has its own cost per contact and measured response. Each channel gets its own curve, and the curves can be compared.
Do we need a propensity model first?
Yes. The simulation reuses the ranking of a propensity model, such as lead scoring.
Who sets the cost and margin values?
Your business teams. TIMi uses them as inputs and shows how the best audience size changes when they change.
RELATED USE CASES
Keep exploring.
Marketing Mix Optimization
Find how to split a marketing budget across actions to maximize expected profit.
Lead Scoring
Rank leads by their likelihood to convert, with the reasons behind each score.
Cross-Sell Prediction
Find the customers most likely to buy a complementary product, and why.
SEE FOR YOURSELF
Test it on your own data.
Run TIMi on your own data with a free trial, or talk to us about your case.
