The Effect of LLM Automation on Performance Marketing Scale: Evidence on CAC, ROMI, and Creative Cycle Velocity

Oleksandr Rodkin
Additional contact information
Oleksandr Rodkin: Marketing Director, CreativeByte, Alicante, Spain

Post-Print from HAL

Abstract: Performance marketing operations in technology-intensive firms have increasingly incorporated large language model (LLM) automation as a mechanism for addressing scalability constraints. This paper examines the quantitative relationship between the degree of LLM automation and three performance indicators: customer acquisition cost (CAC), return on marketing investment (ROMI), and the duration of the creative production cycle. A systematic review of peer-reviewed literature was combined with a comparative case analysis of five digital-native organizations that implemented LLM-based automation stacks between 2022 and 2024. The findings indicate that full LLM automation is associated with a CAC reduction of approximately 39 % relative to unaided baselines, a ROMI increase of 68 %age points, and a compression of the creative cycle from an average of 17.5 days to approximately 2.5 days. Furthermore, A/B test velocity increased by a factor of approximately eight, and the number of active audience segments expanded from fewer than ten to more than ninety. These results suggest that LLM automation constitutes a structural shift in performance marketing capacity rather than an incremental efficiency gain. The findings are of interest to marketing technology practitioners, digital product managers, and researchers studying AI-driven organizational transformation.

Keywords: generative AI; omnichannel marketing; Go-to-Market strategy; unit economics; marketing scalability; creative automation; return on marketing investment; customer acquisition cost; performance marketing; large language models (search for similar items in EconPapers)
Date: 2025-10-01
References: Add references at CitEc
Citations:

Published in European Journal of Management, Economics and Business, 2025, 2 (5), pp.186-196. ⟨10.59324/ejmeb.2025.2(5).14⟩

There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:hal:journl:hal-05712184

DOI: 10.59324/ejmeb.2025.2(5).14

Access Statistics for this paper

More papers in Post-Print from HAL
Bibliographic data for series maintained by CCSD ().