Accounting Data Analytics with Python Training Course

Participants can identify the fields, keys and levels of detail required for accounting analysis.

• Double-entry structures, chart-of-accounts hierarchies and transaction grain in extracts
• General ledger, subledger and operational source-system relationships
• Document numbers, posting dates and unique transaction identifiers
• Master data fields for suppliers, customers, employees and cost centres
• Fiscal calendars, accounting periods and period-close timing issues
• Data dictionaries defining field meaning, format, ownership and permitted values

Participants can import common accounting extracts and inspect their initial structure.

• Reading CSV files with delimiters, encodings and missing-value settings
• Importing Excel worksheets, named ranges and multiple workbook tabs
• Inspecting DataFrame dimensions, columns, data types and sample records
• Parsing posting dates, invoice dates and payment-date fields
• Converting amount columns from text, currency symbols and parentheses
• Exporting controlled analysis outputs to CSV and Excel formats

Participants can assess accounting data quality and produce analysis-ready datasets.

• Missing-value profiling for mandatory transaction and master-data fields
• Standardising supplier names, account codes and cost-centre identifiers
• Removing whitespace, inconsistent capitalisation and hidden text characters
• Duplicate-record checks using transaction keys and business attributes
• Validating numeric fields against expected signs, scales and ranges
• Creating documented cleaning logs for repeatable data preparation

Participants can select, group and summarise ledger transactions for defined accounting questions.

• Boolean filters for entities, periods, accounts and transaction types
• Query expressions for readable multi-condition transaction selection
• Sorting journals by amount, posting date and preparer identifier
• Groupby aggregations for account balances, supplier spend and journal counts
• Pivot tables created from pandas aggregation outputs
• Period comparisons using grouped totals and variance calculations

Participants can join related accounting datasets and perform structured reconciliations.

• Merge operations using invoice numbers, supplier IDs and document keys
• Inner, left and outer joins for reconciliation purposes
• Unmatched-record reports for missing invoices, payments and ledger postings
• General ledger to subledger reconciliation using grouped balances
• Bank transaction matching against cashbook and payment records
• Reconciling control totals before and after transformation steps

Participants can build risk-based tests for manual and automated journal entries.

• Journal populations classified by source, user, account and posting period
• Manual journal filters based on user IDs and source codes
• Late postings identified after period-end and close-date thresholds
• Round-number entries and unusual descriptions screened systematically
• Entries posted to suspense, clearing and rarely used accounts
• Debit-credit balance checks at journal and batch level

Participants can identify potential duplicate, split and anomalous payment transactions.

• Exact duplicate tests using supplier, invoice number and amount
• Near-duplicate matching using dates, amounts and supplier reference fields
• Payment frequency analysis by supplier, bank account and approver
• Split-invoice tests around delegated approval thresholds
• Repeated invoice references across supplier and legal-entity records
• Duplicate-payment exception logs with review status and rationale

Participants can analyse purchase order, goods receipt and invoice matching exceptions.

• Three-way matching between purchase orders, receipts and invoices
• Quantity and price variances calculated against approved purchase orders
• Invoices processed without purchase orders or goods receipts
• Tolerance thresholds applied to value, quantity and timing differences
• Supplier invoices posted after receipt-date and invoice-date gaps
• Exception categories aligned to procurement control owners

Participants can build ageing schedules and investigate overdue balance patterns.

• Ageing buckets calculated from invoice date, due date and reporting date
• Receivable balances segmented by customer, region and credit status
• Payable balances analysed by supplier, currency and payment terms
• Overdue concentration measured through cumulative balance and count metrics
• Credit notes and unapplied cash identified within customer balances
• Ageing movements compared between reporting periods and collection stages

Participants can calculate and explain financial trends, variances and account movements.

• Monthly account movements calculated from periodised general ledger data
• Budget-to-actual variances measured in value and percentage terms
• Common-size analysis using revenue, assets or expenditure bases
• Financial ratios calculated from mapped statement line items
• Material account movements ranked using defined quantitative thresholds
• Variance narratives supported by transactions, drivers and source evidence

Participants can identify timing patterns and unexpected movements in accounting data.

• Monthly time series created from posted transaction dates
• Rolling averages used to smooth volatile expense patterns
• Seasonal comparisons matched to equivalent fiscal periods
• Month-end spikes analysed by journal source and account
• Trend breaks flagged against historical ranges and expected patterns
• Cumulative spend trajectories compared with approved annual budgets

Participants can apply Benford's Law appropriately as a transaction screening method.

• Benford's Law assumptions, uses and limitations for accounting populations
• First-digit frequencies calculated from suitable positive transaction amounts
• Expected Benford distributions compared with observed frequency patterns
• Second-digit and first-two-digit tests for deeper screening
• Mean absolute deviation used to assess distribution divergence
• Benford exceptions investigated with account, user and supplier context

Participants can identify unusual transactions and prioritise them for review.

• Z-scores calculated for transaction amounts within comparable populations
• Interquartile range method applied to skewed expense data
• Percentile thresholds used for large-value transaction screening
• Peer-group comparisons by account, supplier and business unit
• Risk scores combining timing, amount, user and control indicators
• Exception queues prioritised by value, risk rating and recurrence

Participants can test user access and transaction activity for segregation-of-duties conflicts.

• User-role matrices linked to supplier, payment and journal activities
• Conflicting access combinations identified through rule-based tests
• Journal preparer and approver relationships analysed for self-approval
• Supplier master changes matched to subsequent payment activity
• Dormant users and terminated employees identified in access listings
• Access exceptions documented with owners, dates and remediation status

Participants can present accounting analysis through clear charts, tables and exception summaries.

• Matplotlib charts selected for trends, distributions and category comparisons
• Bar charts ranking accounts, suppliers and high-risk exception groups
• Line charts showing monthly balances, spend and collection patterns
• Histograms and boxplots displaying transaction amount distributions
• Exception tables formatted with materiality, status and responsible owner
• Management findings framed around risk, evidence and recommended action

Participants can document accounting analytics work for review, rerun and audit evidence.

• Notebook narrative documenting objectives, populations, logic and conclusions
• Control totals recorded at import, transformation and output stages
• Assumptions, exclusions and materiality thresholds stated explicitly
• Reusable Python functions created for recurring accounting tests
• Peer review checks for logic, completeness and interpretation
• Version control principles for notebooks, data files and outputs

Participants can build, review and present a complete accounting analytics notebook addressing a defined finance or audit question.

• Business question defined with accounting assertions and decision users
• Source datasets profiled, cleaned and reconciled before analysis
• Analytic tests selected for identified financial and control risks
• Python notebook built with documented code, outputs and assumptions
• Exceptions evaluated through materiality, recurrence and supporting evidence
• Findings presented with prioritised recommendations and implementation actions