 Artificial Intelligence in the Generation of Product Description on the Conversion-RateAndré Almeida () and
Darius Zumstein ()
A chapter in Digital Management and Artificial Intelligence, 2025, pp 412-426 from Springer Abstract: Abstract The efficient creation and adaptation of product description on e-commerce platforms, especially with extensive product catalogues, poses a significant challenge. High-quality product descriptions are crucial for customer engagement and purchase decisions. However, manual text creation is time and resource-intensive, leading to increased costs and delays. The introduction of large language models, such as GPT-3 in late November 2022, has expanded the boundaries of what is possible in terms of text quality through machine processing, offering new opportunities for personalized and dynamic product texts. This contribution explores to what extent AI-supported text generation models like ChatGPT can optimize the creation of product texts for e-commerce platforms and influence the conversion rate. Using Do it + Garden Migros as an example, the practical applicability and potential impacts of increased investment in this technology are demonstrated. The aim of the work is to develop a cost-effective and scalable solution that can be applied both within the Migros Group and beyond. The use of RPA enhances the innovative approach of this work. The focus is on the impact of AI on the conversion rate, while legal and ethical aspects such as copyright are consciously delimited. Additionally, the influence of AI within the customer journey. At the core of the work, the current process at Do it + Garden Migros is examined, where new product texts are created only when necessary and by an external text agency, costing CHF 70 per text. For the entire assortment of 40.000 products, the costs would amount to approximately CHF 2.8 million, and the text creation would take 13 to 14 weeks. In an experimental approach, 703 articles were selected to test the efficiency of AI-generated texts. The GPT-3.5-Turbo model was used, resulting in a significant cost reduction from CHF 14 $$7^{\prime}$$ 7 ′ 630 to just $0.46 for all selected articles. Moreover, the AI was able to generate texts in three languages in just two hours and 20 min, representing a considerable time saving compared to manual text creation. The research results demonstrate not only the efficiency and cost-effectiveness of AI-generated texts but also the necessity of continuous monitoring and adjustment of AI models. Overall, the results of the experiment confirm that the use of AI in product text generation has a positive effect on the conversion rate (up to a 23.7% increase). In conclusion, a potential new process was outlined, proposed for the future. In summary, this thesis confirms that the formulated goals and hypotheses were achieved and offers solutions that are relevant not only for Do it + Garden Migros but also for other e-commerce platforms, thus making a significant contribution to solving one of the challenges in the e-commerce sector. Keywords: AI-generated product description; conversion rate optimization; content creation; e-commerce; product management; ChatGPT (search for similar items in EconPapers) There are no downloads for this item, see the EconPapers FAQ for hints about obtaining it. Related works: Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text Persistent link: https://EconPapers.repec.org/RePEc:spr:prbchp:978-3-031-88052-0_34 Ordering information: This item can be ordered from DOI: 10.1007/978-3-031-88052-0_34 Access Statistics for this chapter More chapters in Springer Proceedings in Business and Economics from Springer |
Is your work missing from RePEc? Here is how to contribute.
Questions or problems? Check the EconPapers FAQ or send mail to .
EconPapers is hosted by the
School of Business at Örebro University.