Multilingual Communication in a Global Tech Support Center: A Case Study

Thesis Abstract

This study investigates multilingual communication practices within a global tech support center to understand how language diversity shapes service quality, customer satisfaction, and operational efficiency. The problem addressed concerns rising customer expectations for effective communication across languages in high-volume support environments, where language barriers can lead to miscommunication, longer call durations, and decreased customer trust. The aim is to examine how language proficiency, linguistic repertoire, and communication strategies influence issue resolution, customer experience, and agent performance. Specific objectives include (1) identifying the linguistic profiles of frontline agents and the languages most frequently used in cross-language interactions; (2) analyzing the effectiveness of current multilingual communication strategies and training interventions; (3) assessing the impact of language concordance between agents and customers on call duration, first-contact resolution, and customer satisfaction; (4) exploring the role of culture and pragmatics in multilingual exchanges; and (5) developing a model of best practices for multilingual support in tech service contexts. The study adopts a mixed-methods design anchored in social interactionist perspectives and speech accommodation theory, complemented by Networked Language Theory and sociolinguistic alignment concepts to capture dynamic language use in authentic support encounters. The population comprises frontline technical support agents (n = 120) and customers from five major language groups who interacted with the center over a nine-month period. A stratified random sample yields 60 recorded support interactions per language group (total n ? 300) for quantitative analysis, alongside purposive subsamples of 40 interactions for qualitative examination. Data collection instruments include (i) a standardized multilingual proficiency assessment for agents, (ii) an instrument capturing linguistic resources and repertoire breadth, (iii) structured call metrics from the customer relationship management system (call duration, first-contact resolution rate, escalation rate, sentiment scores), (iv) customer satisfaction surveys post-interaction, and (v) semi-structured interview protocols with 20 agents and 15 team leads to elicit perceptions of training efficacy, cultural factors, and perceived barriers. For analytical rigor, the study employs regression analysis to quantify the relationship between language variables and performance metrics, ANOVA to test differences across language groups, and thematic analysis of interview transcripts to extract patterns of pragmatic strategies and cultural nuance. Interactional coding of a subset of 100 recorded calls using Conversation Analysis (CA) methods will illuminate real-time alignment, code-switching, and repair sequences. Validity and reliability are enhanced through triangulation across metrics, expert panel validation of coding schemes, and inter-coder reliability checks (Cohen’s kappa > 0.80). Expected findings indicate that higher linguistic repertoire and targeted multilingual training are positively associated with shorter call durations, higher first-contact resolution, and elevated customer satisfaction in language-concordant exchanges. Pragmatic strategies such as fund of language, tone calibration, and culturally aware repair mechanisms are anticipated to mitigate miscommunication risks in cross-language interactions. Differential effects across language groups may reveal that non-dominant languages incur longer training payoffs but yield significant gains in customer trust when effectively supported by agents. The study is expected to identify gaps in current training curricula, particularly in pragmatic abilities, cultural competence, and domain-specific terminology management. Theoretically, findings will contribute to the refinement of Speech Accommodation Theory and Networked Language perspectives in applied, multilingual service contexts, and will extend Sociolinguistic Alignment frameworks by integrating organizational constraints and technology-mediated communication. The study contributes to knowledge by providing an empirically grounded model linking language proficiency, pragmatic strategies, training interventions, and performance outcomes in global tech support. Practically, it offers evidence-based recommendations for staffing, targeted multilingual training, and the design of supportive technology (e.g., real-time translation aids and glossaries) to optimize multilingual service delivery. The conclusion emphasizes the necessity of continuous, language-aware performance monitoring and iterative improvement of cross-lingual communication protocols. Recommendations include implementing language-specific onboarding, ongoing proficiency certification, scenario-based CA-informed training, and organizational policies promoting linguistic inclusivity to enhance customer experience and operational efficiency. Suggestions for further research include longitudinal studies tracking the evolution of multilingual competencies and the impact of emerging AI-assisted translation tools on human–machine collaboration in tech support.