Technology and information systems: formula sheet

Key formulas

System-to-level match

TPS → operational | MIS → tactical control | DSS → tactical/strategic decisions | EIS → strategic

A memory guide. Real organisations blur these lines, so state the main use and the user.

Data to information

Data → processing → information → decision/action

Information must be relevant, accurate, timely, understandable and cost-effective to be useful.

Control versus decision support

MIS = regular, structured reports | DSS = interactive, what-if modelling

Use this to answer the common MIS versus DSS comparison.

ERP core idea

Single database + integrated modules = one version of the truth

Always weigh benefits against cost, implementation risk and inflexibility.

The 4 Vs of big data

Volume + Velocity + Variety + Veracity

Define each V briefly, then apply it to the scenario's data. Value is sometimes added as a fifth V.

Four types of analytics

Descriptive (what happened) → Diagnostic (why) → Predictive (what will happen) → Prescriptive (what to do)

Sophistication and value usually rise along the sequence, but so do cost and skill needs.

Insight-to-action chain

Data → Information → Insight → Decision → Performance improvement

Use it to show that data has value only when it changes a decision or a measure.

AI and ML relationship

Machine learning ⊂ Artificial intelligence

ML is a subset of AI. Use this to answer 'difference between AI and ML' questions.

Evaluating an investment in technology

Net benefit = Quantified benefits − Quantified costs, then consider non-financial factors

Use only if the scenario gives figures. Always add risks, data quality and strategic fit.

RPA suitability test

Suitable if the task is rule-based, repetitive, high-volume and uses structured data

This is a rule of thumb, not a guarantee. Tasks needing judgement are poor candidates.

CIA triad

Security = Confidentiality + Integrity + Availability

Use it as a checklist when a scenario describes a data breach or system failure. Say which element is hit.

Data quality attributes

Accurate, Complete, Consistent, Timely, Relevant, Valid

Use these to assess whether information is fit for purpose. Name the attribute that fails in the scenario.

Net benefit of a new system

Net benefit = Total benefits − Total costs (compare on a discounted basis where cash flows span several years)

Include one-off costs such as migration and training plus running costs. Add qualitative benefits and risks separately.

Risk exposure

Risk exposure = Likelihood × Impact

A qualitative guide for ranking risks. Use numbers only if the question gives them.

Quick revision

• Information systems support different management levels: operational, tactical and strategic decisions need different detail and timing.
• Big data is often described by volume, velocity, variety and, in some views, veracity and value.
• Analytics can describe what happened, explain why, predict what may happen and suggest what to do.
• Machine learning finds patterns from data and improves with more data; its output depends on the quality of that data.
• AI can automate routine work and support forecasting, but it needs human oversight and judgement.
• Blockchain is a shared, tamper-resistant record; it suits cases needing trust between parties, such as supply chain tracking.
• Poor data quality leads to poor measures and poor decisions, however advanced the system is.
• Good data is accurate, complete, timely, consistent and relevant to the decision.
• Data governance sets who owns data, who can access it and how it is used and protected.
• Cyber risk needs layered controls: access limits, encryption, monitoring, staff training and response plans.
• Always weigh cost, benefit and risk, then give a clear recommendation tied to the scenario.
• Insights only matter if they change decisions or measures; link data to KPIs and actions.

Common mistakes

• Treating MIS and DSS as the same thing. Fix: Say MIS gives regular structured reports for control, while DSS is interactive and models what-if questions for semi-structured decisions.
• Describing ERP as just a bigger accounting package. Fix: Stress integration of all functions on one shared database, with real-time data across departments.
• Listing the 4 Vs with textbook definitions and no scenario link. Fix: After each V, add a clause using the company's own data, such as real-time sensor feeds for velocity.
• Confusing veracity with volume or treating it as 'truthfulness' only. Fix: Define veracity as the reliability and quality of data, including errors, bias and gaps, and say how it affects decisions.
• Treating AI, machine learning and RPA as the same thing. Fix: State that ML is a part of AI that learns from data, and that RPA follows fixed rules and does not learn.
• Giving a textbook description of blockchain with no link to the business. Fix: After defining it, name a process in the scenario, such as tracking goods or settling inter-company balances, and explain the effect.
• Writing a generic list of cyber threats with no link to the scenario. Fix: Choose only the threats the scenario supports, quote the facts, and explain the effect on that business.
• Treating data quality and data security as the same thing. Fix: Quality is about fitness for use. Security is about protection. Name which one you are discussing and use the right framework.
• Treating data and information as the same thing. Fix: Define both early. Show that information is processed, in context and useful for a decision.
• Listing features of dashboards without linking them to the scenario. Fix: Tie each point to the named company, its users and its strategy. Use scenario facts in every paragraph.

Exam tips

• Always name the user and the decision before naming the system. Examiners reward application over definitions.
• When asked to compare systems, use a clear structure: purpose, user, data used, output. This earns easy technical marks.
• For recommendation questions, include both benefits and risks, then reach a clear conclusion for professional skills marks.
• Link systems to performance management: KPIs, variance reporting, budgets and dashboards give the topic its APM flavour.
• Mention information quality and data governance briefly. Many answers miss these points.
• Always tie each V or analytics type to a fact from the scenario. Generic definitions score few technical marks.
• Use the professional skills marks: show scepticism about data quality and give a balanced, commercially sensible recommendation.
• When asked to evaluate, cover benefits and limits, then conclude. A one-sided answer loses marks.