Design, implement, and evaluate a digital secretary workflow prototype for SMEs

Thesis Abstract

This study investigates the design, implementation, and evaluation of a digital secretary workflow prototype aimed at enhancing administrative efficiency, information flow, and decision support within small and medium-sized enterprises (SMEs) facing resource constraints and increasing document handling demands. The problem addressed is the fragmentation of secretarial tasks due to legacy processes, inconsistent use of digital tools, and limited integration between communication channels, scheduling, and records management, which collectively reduce responsiveness and organizational agility. The aim is to develop a scalable prototype that automates routine secretary tasks, optimizes document routing, and integrates with existing enterprise systems to improve task completion times, accuracy, and user satisfaction. Specific objectives are to (1) analyze current secretarial workflows in a purposive sample of SMEs, (2) design a modular digital secretary prototype grounded in workflow and information-management theory, (3) implement the prototype using a cloud-based microservices architecture and standard interfaces (RESTful APIs, OAuth 2.0), (4) evaluate usability, efficiency, and accuracy through a mixed-methods approach, and (5) provide evidence-based recommendations for SME adoption and future improvements. A mixed-methods research design is employed, combining a quasi-experimental evaluation with qualitative insights. The population comprises SME administrative teams across manufacturing and professional services sectors in a metropolitan region, with a target sample of 20 organizations. Within each organization, a purposive sample of 2–3 secretaries and 2 managers participates, yielding approximately 50–60 user participants. Data collection instruments include structured time-motion logs and task completion metrics extracted from the prototype, system usability scale (SUS) questionnaires, think-aloud usability sessions, semi-structured interviews, and platform analytics. Validity and reliability are ensured through pilot testing of instruments (n=10), triangulation across quantitative and qualitative data, inter-rater reliability checks for coding qualitative data (Cohen’s kappa ? 0.70), and demonstration of construct validity via factor analysis for the SUS. Analytical procedures comprise descriptive statistics, inferential analyses, and qualitative thematic analysis. Descriptive statistics summarize workflow efficiencies, average task turnaround times, and error rates. Inferential analyses employ paired t-tests and repeated-measures ANOVA to compare pre- and post-implementation performance across key metrics (task throughput, response time, error rate), with regression analysis identifying predictors of user satisfaction and efficiency gains. Thematic analysis of interview and think-aloud data will elucidate user experiences, perceived barriers, and facilitators to adoption. The study will also apply the Technology Acceptance Model (TAM) and the Diffusion of Innovations (DOI) framework to interpret usage patterns and adoption readiness. A conceptual model integrating workflow automation principles and information governance is proposed to guide interpretation. Key expected findings include reductions in average task turnaround times by 25–40%, higher task accuracy due to automated routing and standardized templates, and improvements in perceived workload and user satisfaction (SUS scores increasing from below 65 to above 75). The research anticipates identifying critical design features such as adaptive routing rules, natural language processing-enabled email triage, calendar integration with conflict resolution, and secure role-based access controls. It is also expected to reveal contextual factors affecting adoption, including organizational culture, prior digital literacy, and alignment with existing ERP/CRM systems. The study’s contributions to knowledge encompass (i) a validated design and implementation blueprint for a digital secretary workflow prototype tailored to SMEs, (ii) empirical evidence on the impact of integrated workflow automation on administrative performance in resource-constrained settings, and (iii) practical guidelines for SME stakeholders on change management and technology integration. In conclusion, the research will demonstrate that a modular, cloud-based digital secretary prototype can meaningfully streamline routine administrative tasks, improve information governance, and support decision-making in SMEs without imposing prohibitive cost or complexity. Recommendations include adopting phased implementation with stakeholder engagement, investing in staff training focused on new workflows, ensuring interoperability with common SME software ecosystems, and prioritizing data privacy and security. Limitations relate to the regional focus and potential variability in SME caseloads, suggesting avenues for broader replication studies and longitudinal assessments to examine durability of gains and long-term organizational impact.