Work

AI-Powered Personalization: Exploring the Role of Large Language Models in Driving Email Marketing Conversions

Year: 2025
Type: dissertation
Abstract: Email marketing faces a growing challenge: generic content and static segmentation techniques are losing effectiveness in a saturated digital landscape. As consumer expectations for relevance increase, marketers struggle to personalize at scale without sacrificing quality or brand voice. This thesis investigates whether Large Language Models (LLMs) can offer a solution by generating behaviorally informed, personalized email content that drives measurable improvements in marketing performance. A series of four A/B tests were conducted using a real-world dataset from a high-end cosmetics brand. Group A received traditionally written emails; Group B received LLM-generated content derived from dynamic customer behavioral insights. While the results were not statistically conclusive in favor of LLMs across all campaigns, several noteworthy outcomes emerged. LLM-generated emails consistently produced higher click-through rates and, in some cases, significantly higher average order value—even when conversion rates remained unchanged. These findings suggest that LLMs can enhance the economic quality of conversions and serve as valuable tools for scaling personalized content creation. However, their effectiveness is context-dependent, and human oversight remains crucial. This study concludes that LLMs are best viewed not as replacements for human creativity, but as scalable partners in the personalization process—particularly when aiming to balance productivity with contextual relevance.
Author: Milojevikj, Ana Marija
Language: English
Subfield: Marketing
Open Access status: green
OpenAlex ID: W7112452000
Cites: 14
Cited by:
Related to: 10