E EREACE
Back
Improve margin and mixDTC Brand / Cosmetics - Marketing and Advertising

Optimising 40 AI-Generated Ad Creatives

A direct-to-consumer cosmetics brand uses AI to generate 40 ad variants every week (image, headline, call to action). A classic A/B test with 40 variants would take weeks and spend most of the budget on poor creatives. In addition, creatives 'burn out' in 10-14 days. The CMO wants to learn fast and not waste budget.

Estimated annual impact on EBITDA
+€2.10M
DTC brand with €6M of annual ad investment. Illustrative example with indicative figures: replace them with your own company figures.

Where the money comes from

Less budget on losing creatives
+€1.50M
More sales with the same budget (margin)
+€800k
Cost of the system and AI production
-€200k
Annual EBITDA impact
+€2.10M
Adds profitReduces profitSystem cost

What to do now

  1. 1Define the bandit's reward as the margin of the sale, not the click.
  2. 2Set brand guardrails and pre-review of the creatives.
  3. 3Launch with 10% exploration and compare with a classic A/B test.

For your committee

One page with the decision, the impact, the risks, the steps and the KPI to track.

The print dialog opens: choose "Save as PDF".

Watch out: more volume is not always more profit

The creative that converts the most may be that of a low-margin product or an aggressive offer. Optimise the bandit by margin per impression, not by clicks or sales.

KPI for the committee

Margin per 1,000 impressions.

Detail for your data team

Proposed technique: Multi-Armed Bandit

Why: The goal is not to predict, but to decide while learning. The bandit balances exploration (trying variants) and exploitation (showing the best one), shifting traffic towards the winners in hours instead of weeks. Ideal when there are many variants and a short life.

Cost of getting it wrong: Exploring too much spends budget on poor creatives; exploiting too early locks in a false winner and fails to discover the best one.

Data needed: Impressions, clicks and purchases per creative in real time (ad platform APIs).; Cost per impression by channel and time slot.; Metadata for each creative: format, tone, product, colours.

The Ereace system

Test decisions with synthetic people before you test them with real ones.

This tool helps you decide with numbers. Before you invest, understand, test and rehearse the decision with the people involved: your customers, your team or your committee.

Understand
QualiSynth

Before you invest, find out what your customers think, need or reject about the initiative.

Test
StrataSynth

Put the offer, the message or the journey in front of synthetic people who do not always follow the happy path.

Train
ArenaSynth

Rehearse the committee presentation or the vendor negotiation against synthetic decision-makers.

See all Ereace products

About this tool. It is free and made by Ereace, an AI-native product studio and human intelligence ecosystem: it connects real human signals, synthetic people and the systems built around them, so teams can understand, test and act on human behavior before decisions reach the real world. About Ereace

Comments or suggestions? info@ereace.es