[Paper Review] Trends and Challenges Towards an Effective Data-Driven Decision Making in UK SMEs: Case Studies and Lessons Learnt from the Analysis of 85 SMEs

Abdel‐Rahman H. Tawil, Muhidin Mohamed+3 more|arXiv (Cornell University)|May 24, 2023

Big Data and Business IntelligenceBusiness, Management and Accounting3 citations

TL;DR

This study investigates data-driven decision-making trends and challenges in 85 UK SMEs, revealing that while data science offers significant benefits in productivity and innovation, SMEs face barriers like limited funding, skills gaps, and data quality issues. The research identifies key enablers—such as phased digital adoption, staff upskilling, and external support—to help SMEs build data cultures and leverage AI and analytics effectively.

ABSTRACT

The adoption of data science brings vast benefits to Small and Medium-sized Enterprises (SMEs) including business productivity, economic growth, innovation and jobs creation. Data Science can support SMEs to optimise production processes, anticipate customers' needs, predict machinery failures and deliver efficient smart services. Businesses can also harness the power of Artificial Intelligence (AI) and Big Data and the smart use of digital technologies to enhance productivity and performance, paving the way for innovation. However, integrating data science decisions into an SME requires both skills and IT investments. In most cases, such expenses are beyond the means of SMEs due to limited resources and restricted access to financing. This paper presents trends and challenges towards an effective data-driven decision making for organisations based on a case study of 85 SMEs, mostly from the West Midlands region of England. The work is supported as part of a 3 years ERDF (European Regional Development Funded project) in the areas of big data management, analytics and business intelligence. We present two case studies that demonstrates the potential of Digitisation, AI and Machine Learning and use these as examples to unveil challenges and showcase the wealth of current available opportunities for SMEs.

Motivation & Objective

• To examine the current trends and challenges in adopting data-driven decision-making practices among UK SMEs.
• To identify barriers—such as financial constraints, lack of technical skills, and poor data quality—that hinder data science adoption in SMEs.
• To explore how digitalization, AI, and machine learning can be effectively integrated into SME operations through practical case studies.
• To propose actionable strategies for SMEs to build data capabilities incrementally, including staff upskilling and phased technology adoption.
• To highlight the role of external stakeholders—such as government, academia, and business associations—in supporting SMEs through knowledge sharing and funding

Proposed method

• Conducted in-depth case studies across 85 SMEs, primarily from the West Midlands region of England, to analyze real-world data-driven practices.
• Collected and analyzed data on SMEs’ data collection, storage, management, and usage practices, including challenges in data quality and governance.
• Evaluated machine learning models (specifically decision trees) on oversampled datasets of varying sizes (500–2000 instances) to assess model performance and hyperparameter tuning.
• Used model accuracy and error metrics (e.g., 0.343 error at depth 21–30 with 1500 instances) to assess the impact of data size and tree depth on predictive performance.
• Integrated qualitative feedback from SMEs and independent commercial assessments to validate findings on data quality and model reliability.
• Proposed a framework for incremental data capability development, emphasizing culture building, training, and phased investment

Figure 1: Open Static Datasets Distribution Across Major Themes by Six UK Cities

Experimental results

Research questions

• RQ1 What are the primary trends and challenges in data-driven decision-making adoption among UK SMEs?
• RQ2 How do data quality, sample size, and model hyperparameters affect the performance of predictive models in SMEs with limited data?
• RQ3 What role do digitalization, AI, and machine learning play in enhancing SME productivity and innovation?
• RQ4 What strategies can SMEs adopt to overcome financial, technical, and cultural barriers to data science adoption?
• RQ5 How can external stakeholders such as government, academia, and business associations support SMEs in building sustainable data capabilities?

Key findings

• SMEs in the UK collect and store business data but often lack the skills to analyze it effectively for actionable insights.
• Model accuracy improved with increasing decision tree depth, stabilizing at a maximum depth of 21, with the lowest error (0.343) observed at depths 21–30 using 1500 oversampled instances.
• Small sample sizes (n < 100) and noisy data were primary contributors to high model errors, confirming the statistical challenge of limited data in predictive modeling.
• Despite limited resources, SMEs showed strong potential in adopting data science when supported by training, phased investment, and external expertise.
• A significant digital adoption gap remains between SMEs and larger firms, particularly among micro and small firms (10–49 employees), due to financing and skills constraints.
• External support—through funding, knowledge sharing, and open data—was identified as critical to helping SMEs build data capabilities and achieve long-term digital transformation

Figure 2: Lifecycle of BDC project support for SMEs