← Selected work

LG Electronics Applied LLMs

LLM email personalization

An email personalization engine built around 11 attitude-based audience segments.

Large language modelsAudience segmentationPersonalization
LLM email personalization
Audience attitudes
11 segments
Personalized copy
Pilot evaluation
From a business question to a working system.Workflow overview
pilot conversion lift
+58%
copywriting time
−75%

The problem

Email copy needed to reflect differences in customer attitudes while making the copywriting process more efficient.

What I built

Used large language models to personalize email content for 11 attitude-based segments.

Evaluated the approach in a pilot, tracking conversion, click-through rate, and copywriting time.

The outcome

The pilot increased conversion by 58% and click-through rate by 78%, while reducing copywriting time by 75%. These are pilot results.

A summary of the work described on my public profile.

View on LinkedIn
Explore another projectSemantic search at Amazon