Technology and Analytics
Systems and data
An ERP integrates all functions on a single database, eliminating redundant data entry but creating implementation, cost, and change-management risk. Know the distinction between a data warehouse (central, structured, subject-oriented), a data mart (subset for one function), and a data lake (raw, unstructured data). ETL extracts, transforms, and loads data. OLTP supports daily transactions; OLAP supports multidimensional analysis and slicing of cubes for reporting.
Analytics types
- Descriptive: what happened (dashboards, KPIs, variance reports).
- Diagnostic: why it happened (drill-down, root-cause analysis).
- Predictive: what is likely to happen (regression, forecasting, classification, clustering, machine learning).
- Prescriptive: what should be done (optimization, simulation, decision modeling).
Big data is often described with five Vs: volume, velocity, variety, veracity, and value. Data mining discovers patterns; data visualization supports interpretation and communication.
Emerging technologies
RPA automates rule-based, repetitive, structured tasks - it does not learn. AI/machine learning identifies patterns and improves with data. Blockchain is a distributed, immutable ledger enabling smart contracts and enhanced traceability. Cloud models are IaaS, PaaS, and SaaS. Cybersecurity controls include multi-factor authentication, encryption, patch management, least-privilege access, phishing training, and tested backup and recovery.
Traps
Questions often test predictive versus prescriptive (forecasting the outcome versus recommending the action) and confuse RPA with AI. Another trap is SOC reports: SOC 1 covers controls relevant to financial reporting; SOC 2 covers security, availability, processing integrity, confidentiality, and privacy. Also distinguish disaster recovery (IT restoration) from business continuity (keeping the entire business operating).
Rule of thumb
Match the tool to the question's verb: report = descriptive, explain = diagnostic, forecast = predictive, decide/optimize = prescriptive.
Worked example
A retailer uses historical sales, promotions, and weather data to estimate next month's demand at each store - that is predictive analytics. Feeding those estimates into an inventory optimization model that sets reorder quantities is prescriptive.