The Impact of ISO Certifications, Machine Translation (MT), & Large Language Models (LLMs) in the Quality of English into Spanish Translations

Abstract

[EN]Over the last few years, ISO (International Organization for Standardization) certifications have become a requirement in nearly every industry. They have grown in importance to such an extent that specific certifications have been created for industries that, until recently, had no formal standards to follow. This is particularly true for the flourishing translation industry, which has expanded rapidly in this era of globalization. The ISO 17100:2015 certification has become a worldwide reference, signifying that a business adheres to ISO Quality Assurance Standards. From a client’s perspective, this certification assures that the services provided meet rigorous quality criteria. However, the internal benefits of this certification for small and medium-sized businesses are still not well understood. The standard that is universally recognized across industries is ISO 9001:2015, which outlines the requirements for achieving the highest quality in any given service. More recently, ISO 17100 was established to address the specific needs of the translation and localization industry. This standard has become a seal of approval for clients, ensuring that organizations adhere to a series of guidelines that lead to the delivery of high-quality services. However, questions remain about how this standard truly impacts the quality of translations and the worklife balance of both translators and project managers. This research aims to provide an in-depth internal perspective on the adaptation process of a Language Service Provider (LSP) to the requirements of ISO 17100, and to assess whether this certification enhances the quality of English-to-Spanish translations. Furthermore, it seeks to evaluate the certification’s impact on the work environment, particularly in terms of job satisfaction and motivation among translators and project managers, with the goal of addressing the industry’s well-documented high turnover rates (Deci & Ryan, 2017). Additionally, it investigates the effect of ISO 17100 certification on client satisfaction. The study also explores the integration of Machine Translation (MT) technologies in the context of ISO 18597:2017, which sets guidelines for the post-editing of machine-generated translations. While this standard has the potential to raise translation quality by aligning machine outputs with professional human standards, the present study limits its analysis to the effects of this certification on project managers’ and translators’ job satisfaction, as the MT service was only provided to clients in 2024, resulting in an insufficient sample size to evaluate translation quality comprehensively. In light of recent technological advancements, the research provides a foundation for future exploration of Large Language Models (LLMs), such as GPT-4, which have significantly influenced the translation industry by offering advanced natural language processing capabilities. Although LLMs were not implemented or evaluated within the scope of this study, the research establishes a theoretical framework to guide subsequent investigations. To conduct this research, a multidisciplinary approach was employed. This includes a qualitative analysis of written sources related to ISO standards (Meriam, 2009), as well as a comparative analysis of English into Spanish translations produced before and after ISO certification (Rabadán, Labrador & Ramón, 2009). Furthermore, the study delves into the personal experiences of translators and project managers through structured questionnaires (Pickering, 2008) to provide a deeper understanding of how ISO certifications, in combination with MT and LLM technologies, influence translation quality and professional satisfaction. The research will first explore the challenges faced by a medium-sized company’s Translation & Localization department, such as inconsistencies in translations and the frustration experienced by both in-house teams and freelance translators. These challenges, exacerbated by the lack of standardization within the profession (European Commission, 2012), have led to high turnover rates. To mitigate these issues, the company pursued ISO 8 17100 certification, which outlines the requirements for the competence and qualifications of translation service providers to ensure consistency, accuracy, and quality in translation processes. Additionally, the company sought ISO 18587 certification to standardize the postediting of Machine Translation outputs, recognizing the industry’s shift towards more automated forms of translation. As the use of MT and LLM technologies continues to grow, this research will explore how these advancements, when aligned with ISO standards, contribute to maintaining and enhancing the quality of translation services. Both, the complete documentation created by the company in order to achieve the certification along with a detailed analysis of the timeframe in order to accomplish each objective will be presented. Last, a thorough analysis of the improvement, elaborated taking three methods into account: comparison between translations done before and after achieving the certification, a questionnaire for translators and a questionnaire for project managers to evaluate how their quality of life has improved and if the turnover has decreased.