5.9 Management information systems

Data analytics examines data to identify patterns, explain performance or support predictions and decisions. A database is an organized electronic collection of related data that can be stored, updated and queried. Cybersecurity protects systems, networks and data; cybercrime is illegal activity using or targeting digital systems or information.

A management information system combines people, processes, databases and technology to collect, process and present timely information for planning, decisions and organizational control. Analytics is only as reliable as the database's accuracy, completeness and relevance, while cybersecurity protects confidentiality, integrity and availability from cybercrime and other threats.

Start with the management decision, identify the minimum relevant data and database fields, define access and quality controls, and decide which analysis can validly answer the question. Protect the system through proportionate technical, procedural and human controls.

A retailer queries its sales-and-stock database and analyses demand by store to plan replenishment. Access controls, backups and staff phishing awareness reduce the risk that cybercriminals steal or alter the data.

Raw data is not automatically useful information, analytics does not prove causation, and cybersecurity cannot eliminate all risk. A database is the organized data store, not the whole MIS.

Critical digital infrastructure includes data centres that house computing and storage, cloud computing that supplies scalable remote resources, and artificial neural networks that learn weighted patterns from data. Business technologies also include virtual reality (VR), the internet of things (IoT), artificial intelligence (AI) and big data.

VR creates simulated environments for training, design or customer experiences; IoT connects sensor-equipped objects that collect and exchange data; AI performs tasks involving prediction, recognition or decision support; big data describes datasets whose volume, variety or speed require advanced processing. These applications rely on infrastructure, connectivity, skills, quality data and governance.

Match the technology to a precise business problem, then analyse implementation cost, compatibility, skills, energy and infrastructure needs, vendor dependence, cybersecurity, privacy, bias, reliability and stakeholder impact.

A manufacturer sends IoT sensor data through cloud infrastructure to an AI model that predicts equipment failure. The system may reduce downtime, but managers need secure connections, reliable data, human verification and a fallback if the data centre or provider is unavailable.

An artificial neural network is one AI approach, not a human brain, and big data is not automatically good data. Cloud use transfers some infrastructure work but not the business's accountability for security, privacy or decisions.

Customer data can improve segmentation, personalisation, service and retention when collected lawfully and used transparently.

Tracking creates value only if the insight changes an offer or interaction; misuse damages trust and may breach privacy rules.

Identify the customer decision, consent basis, data minimisation and expected benefit.

A loyalty app recommends products from purchase history while allowing customers to opt out.

Personalisation does not justify collecting every available data point.

A customer loyalty programme rewards repeat behaviour and records transactions, which can improve retention, targeting and personalized offers; however, rewards cost money, may attract deal-seeking rather than loyal customers, and can raise privacy or fairness concerns. Data mining examines large datasets to discover patterns or relationships that inform segmentation, demand forecasts and decisions. Its conclusions depend on data quality and interpretation: correlation may not show causation, historical patterns may encode bias, and opaque collection can damage trust. Evaluate whether the insight changes a useful decision and whether consent, access, retention and customer benefit are proportionate.

Employee data can support scheduling, training, safety and performance decisions, but monitoring affects autonomy, fairness and trust.

Metrics may be incomplete or biased; workers need clear purpose, access and routes to challenge errors.

Check whether the measure reflects the job and whether the use is proportionate and explained.

A warehouse uses workload data to plan breaks, not to rank workers without context about equipment faults.

A precise metric can still be an unfair measure.

Digital Taylorism uses digital systems to divide, standardize, measure and closely monitor work, extending scientific-management ideas through real-time data, algorithms or automated targets. It may identify bottlenecks, support scheduling, improve consistency and raise productivity, but narrow metrics can ignore quality and context, intensify work, reduce autonomy, create stress and weaken trust. Employee data should have a clear purpose, proportionate collection, transparent criteria, human review and a way to correct errors or challenge decisions; productivity gains must be weighed against motivation, retention, ethics and long-term performance.

MIS can improve coordination, speed, forecasting and control, but creates cyber, privacy, dependency, bias and surveillance risks.

Good governance sets access, retention, audit, security and accountability; ethical use balances efficiency with rights and stakeholder harm.

Identify the benefit, threat, affected stakeholder and control, then judge residual risk.

A firm encrypts customer data, limits access and audits model decisions rather than assuming the software is neutral.

Compliance alone does not prove ethical or socially responsible use.