AI-DRIVEN BUSINESS ANALYTICS FOR FINANCIAL FORECASTING: A SYSTEMATIC REVIEW OF DECISION SUPPORT MODELS IN SMES

Abstract

The accelerating convergence of artificial intelligence (AI), business analytics, and financial management has redefined how small and medium-sized enterprises (SMEs) forecast cash flows, allocate resources, and navigate volatile market conditions. Yet, research on the breadth and depth of AI‐driven decision support models for SME financial forecasting remains fragmented. Addressing this gap, the present systematic review and meta-analysis synthesizes findings from 78 peer-reviewed studies published between 2015 and 2025, each investigating the deployment of machine-learning, deep-learning, or hybrid-intelligence systems in SME forecasting and budgeting contexts. Guided by PRISMA protocols, we searched five major databases—Scopus, Web of Science, IEEE Xplore, ScienceDirect, and Google Scholar—followed by rigorous title, abstract, and full-text screening. Eligibility criteria required empirical, quantitative evidence, explicit focus on SMEs, and sufficient statistical detail to calculate standardized effect sizes. Ultimately, 67 studies met all inclusion standards and were subjected to meta-analytic pooling using a random-effects model. The aggregated results reveal a robust, statistically significant improvement in financial-forecast accuracy, decision speed, and overall financial performance among AI-adopting SMEs. Hybrid frameworks—those combining human expertise or traditional statistical methods with machine learning—produced the largest gains, underscoring AI’s role as an augmentative, rather than purely autonomous, decision partner. Industry-level analysis highlights especially strong benefits in manufacturing and retail, where high-frequency transactional data supports granular demand analytics, while service-sector SMEs reported meaningful, albeit smaller, improvements in scheduling, pricing, and customer-engagement precision. Geographically, firms in digitally mature ecosystems attained greater returns than counterparts in emerging markets, a disparity linked to infrastructure readiness and data-governance practices. Beyond quantitative gains, qualitative evidence from case studies indicates that AI deployment fosters a cultural shift toward data-driven decision-making, elevating organizational agility in budgeting cycles and cash-management routines. Nevertheless, recurrent implementation challenges—limited analytic expertise, data fragmentation, and algorithmic transparency concerns—temper the pace of adoption. The findings collectively demonstrate that AI-powered decision support is not merely a technological upgrade but a strategic enabler capable of leveling competitive asymmetries between SMEs and larger enterprises. By presenting a thematic taxonomy of AI models, synthesizing effect magnitudes, and identifying contextual moderators, this review offers actionable insights for managers, policymakers, and researchers seeking to harness AI for resilient, evidence-based financial planning within the SME sector.