CA ANZ Data Analytics and Insights study plan

Distinguish structured from unstructured data and classify ten sample data elements by type, source and storage format

Identify and document five common data quality defects (completeness, accuracy, consistency, timeliness, uniqueness) in a supplied dataset description

Build a simple data dictionary for a fictional accounts-receivable extract, defining field name, type, allowable values and owner

Write a repeatable validation checklist that another reviewer could apply to test data before analysis

Explain when to use descriptive, diagnostic, predictive and prescriptive analytics, matching each to a business question

Perform basic profiling on a small dataset using spreadsheet functions or a scripting tool: row counts, null counts, distinct values, min/max, outliers

Write and debug at least five query or formula constructs that join, filter, aggregate and group data

Interpret a correlation coefficient and explain why correlation does not establish causation in an audit context

Apply sampling concepts: select a random and a stratified sample from a transaction population and compare coverage

Design a full-population test approach for a duplicate-payment scenario, stating the test logic and exception criteria

Calculate and interpret basic statistical measures (mean, median, standard deviation, z-score) on a monetary population

Draft the working-paper structure that records data source, extraction date, test logic, results and reviewer sign-off

Map analytics tests to financial statement assertions (occurrence, completeness, accuracy, cut-off, classification)

Design three journal-entry testing routines: round-number entries, post-period-end entries and unusual user-postings

Build a three-way match test for purchase order, goods receipt and invoice, defining tolerances and exception handling

Analyse a revenue dataset for cut-off risk around period end and quantify the value of exceptions

Construct a variance analysis comparing actual to budget by cost centre and explain the two largest drivers

Build a simple driver-based forecast and state the assumptions and their sensitivity

Select the correct chart type for six given analytical questions and justify each choice

Redesign one poor chart by removing clutter, fixing axis scales and adding a clear takeaway title

Write an insight statement in the form finding, so-what, recommended action, for three separate analytical results

Structure a five-slide narrative that moves from question to evidence to recommendation

Apply appropriate precision and rounding when reporting monetary and percentage results to avoid overstating confidence

Deliver a short verbal briefing to a non-technical audience and record the questions raised

Apply confidentiality, privacy and data-minimisation principles to a scenario involving customer records, documenting lawful basis and retention

Identify threats to objectivity and professional scepticism when an analyst also owns the process being tested

Evaluate three emerging techniques (process mining, robotic process automation, machine-learning anomaly detection) against a stated business problem, noting limitations

Assess the risks of biased training data and unexplainable model outputs in a decision-support context

Consolidate weak areas by re-testing the two domains with the lowest practice scores

Complete a timed mixed-topic practice set and review every incorrect answer to its underlying concept

Summarise each domain into a single page of key definitions, formulas and judgement points

Rehearse explaining one end-to-end analytics case from data extraction to communicated insight