PwC Audit Analytics

Context

At PwC Czech Republic, audit engagements increasingly relied on analyzing large volumes of client financial data - not sampling a few rows and extrapolating, but working across full populations of transactions and accounts when the engagement team needed defensible conclusions.

Problem

Traditional audit sampling misses patterns that only show up at scale: odd clusters, boundary cases, inconsistencies between statements. Teams needed analysts who could run systematic analysis over P&Ls, cash flow statements, and related financial data without drowning in noise.

My role

Data analyst on audit analytics workstreams: building and running analysis that supported engagement conclusions.

Constraints

• Engagement deadlines: Analysis had to land while the audit clock was running.
• Messy client data: Labels, charts of accounts, and exports rarely matched textbook examples.
• Defensibility: Every cut had to survive partner and client scrutiny.

What I did

• Analyzed high-volume financial datasets across income statements, cash flows, and balance-sheet-related views.
• Helped engagement teams move from "we think something is off" to "here is what the full population shows."
• Documented logic clearly enough that others could reproduce and challenge the work.

Key decisions

• Population thinking: When data volume allows, test the whole picture - not only a comfort sample.
• Clean questions beat fancy tooling: The hard part was defining what "wrong" meant for that client, not the software label.

Result

Repeatable audit analytics support on engagements where data volume was part of the risk picture - practical "big data" work in a pre-hype enterprise setting.

What this shows

Rigor with real financial data under deadline pressure - a foundation for later KPI modelling, forecasting, and product metrics work where definitions matter.