Predict

Here's how to use Predict.

Predict is an agent designed to help you anticipate the future results of a campaign for a specific Impact KPI. To generate these predictions, the tool analyzes the parameters you provide (country of diffusion, media type, advertising tone, etc.) and compares them with data from the Happydemics Brand Lift benchmark.

The first step is to build your basis prediction.

This is the reference configuration that represents your campaign as you currently envision it. It serves as the foundation for your main prediction, as well as the comparison point for evaluating alternative scenarios.

Choosing the KPI and creating the basis prediction

Start by selecting the Impact KPI for which you want to obtain a prediction (e.g., Consideration, Brand Image…).

No predictions are available for creative diagnostic KPIs, as it is unreliable to generate these forecasts without taking the creative into account.

Next, provide the key elements that define your campaign:

• Country of diffusion

• Industry

• Media used

• Allocated budget

• Expected reach

• Campaign objective (DTS or Branding)

• Advertising tone

• Type of claim included in the ad

This information forms the calculation basis on which Predict builds your initial projection.

Adding variables

Once your basis prediction is set, you can add variables, which are alternative values for certain criteria.

These variables represent potential optimization levers (changing the media, adjusting the tone, modifying the budget, etc.).

By combining them, Predict generates different scenarios, allowing you to explore several possible paths to improve your campaign’s performance.

• You can add up to 5 variables.

• This step is optional: you can first generate your basis prediction, then enrich your analysis later by adding new variables.

Reading the results

Once your prediction is generated, Predict presents the results in two distinct parts:

Basis prediction

This is the base prediction, calculated solely from the parameters you initially defined.

Scenarios

Predict then calculates the results produced by combining all the variables you added. Each scenario shows the potential impact of modifying your criteria and is automatically compared with your basis prediction to make interpretation easier.

• Grey characteristics show what remains identical to the basis prediction.

• Blue characteristics highlight what changes.

Thanks to this comparative visualization, you can quickly identify the configuration that generates the best uplift and make informed decisions to optimize your campaign.

👋 If you don't find your answer, feel free to reach out to [email protected] for further consideration !