Skip to content
Sebastian.Gebhardt

Andrea

AI Consumer Simulation for Marketing

The simulation marketing was missing: a 100,000-person synthetic society, creative auditing and Bayesian attribution in one system.

In production · 8 brandsuseandrea.com
100,000 synthetic respondents · a focus group convenes in seconds
Synthetic respondents
100K
Brands in production
8
Modules
30+

The problem

Marketing decides on intuition what engineering already learned to decide with data. Focus groups are slow and expensive, research arrives late, and traditional attribution can't survive a board asking why.

How it works

Andrea runs a 100,000-respondent synthetic society built for Latin America. On top of it: focus groups with real group physics, an auditor that scores creative before money is spent, and Bayesian attribution over actual sales. Research, production, intelligence and performance in one system.

Inside the system.

My perspective

Andrea is my technical project. I started coding after my MBA and kept building: consumer simulation, creative evaluation and marketing measurement in one product. The question behind it comes from the operating side of my work: what can we learn before committing the budget?

The design decision

Bring research, creative evaluation and measurement into the same workflow. A simulation can help explore an idea; an auditor can help revise a piece; attribution over actual sales asks what happened after spending. Each answers a different question. The useful system connects those decisions.

What the result shows

The published deployment covers eight brands. In one creative iteration, the auditor’s score moved from 42 to 72. That is a result inside the evaluation system: it shows how the feedback can guide a revision. The population and module counts describe the product’s scope.

How to read the evidence

An improved creative score is not a measured increase in sales, and synthetic respondents are not observed customers. Assessing market performance requires comparison with real behavior. This case reports product scope and the creative iteration; it does not claim a causal revenue lift from either.

The path of a decision

  1. 01

    Explore a consumer question with the synthetic society and its focus groups.

  2. 02

    Evaluate a creative piece and use the feedback to revise it before media spend.

  3. 03

    Use attribution over actual sales to examine the performance of the marketing investment.

The thinking behind the work, in my essays:

What it does

  1. 01

    Synthetic society

    100,000 synthetic respondents for consumer research on demand, without recruiting a panel.

  2. 02

    Creative auditor

    Scores creative before media spend. One piece jumped from 42 to 72 points in a single iteration.

  3. 03

    Bayesian attribution

    Attribution over actual sales, not clicks — built to answer board-level questions.

If you recognize this problem in your business, let’s talk. For an implementation with my team, the advisory work lives at Menlo & Oak.