Research previewTechnologyConnectorsResults
TITAN-REC 3.0
TITAN-HORIZON 2.5
TITAN-CORE 2.4
TITAN-TREND 1.7
TITAN-LINK 2.2
TITAN-FIRSTSOON
TITAN-ARCSOON
FR/EN
Back to deployments
TITAN-REC · April 2026

Titane deploys its personalized recommendation model across tens of thousands of customers

Titane deploys its personalized recommendation model across tens of thousands of customers

Master Outillage is a national distributor of professional tools. Its e-commerce store lists 100,000 product SKUs, and asked us to work on a base of several tens of thousands of active customers. Titane Intelligence deployed its personalized product recommendation model TITAN-REC there in one week.

The context

Before this deployment, Master Outillage sent identical newsletters to its entire base. The same email, the same products, for every profile — whether an electrician or an independent carpenter.

The problem is familiar to every e-commerce manager: a broad catalog, varied customer profiles, and a personalization capability limited by the tools in place. Email performance stagnates because the content is not relevant to the majority of recipients.

What was deployed

Titane connected Master Outillage's infrastructure — PrestaShop and Mailchimp — to its TITAN-REC recommendation model, through proprietary connectors developed in-house. No intermediary, no third-party platform.

Deployment architecture
PrestaShopMailchimp
Data
Customers, products & activity
TITAN-REC
Recommendation model
Mailchimp
Mailchimp
Email activation

Every customer now receives product recommendations computed individually, injected dynamically into Mailchimp campaigns. Priority was given to loyal customers, regular buyers, and the profiles identified by Titane as inactive.

Example of a personalized product recommendation email
Example of a personalized product recommendation email.

One week

From the first data connection to the first personalized campaign sent in production: one week.

Titane handled the entire process: data ingestion, building and shipping the model to production, full deployment on Mailchimp. No migration. No rebuild of the existing setup. The personalized recommendations were ready to use.

The results

The first recommendation campaigns were measured against the same benchmark: Master Outillage's standard newsletter.

Open rate
Standard newsletter
Titane recommendation
×3.4
versus the standard newsletter
Click rate
Standard newsletter
Titane recommendation
×2.7
versus the standard newsletter