E1 - Managing Finance in a Digital World: the finance function, data and technology
What E1 tests
E1 is the Operational level enterprise subject, sat alongside P1 (Management Accounting) and F1 (Financial Reporting). The objective test runs 90 minutes with 60 questions and a 70% pass mark. Questions are scenario-based: you are asked which technology, structure or data practice fits a described situation, not to recite definitions.
The finance function
- Core activities: transaction processing, stewardship and control, reporting, decision support and strategy. Automation shifts effort toward the last two.
- Structural options: shared service centres (high volume, standardised, low cost), centres of excellence (specialist expertise), embedded business partners, plus outsourcing and offshoring.
- The finance business partner converts financial data into operational decisions and challenges assumptions - the relationship the exam tests most often.
Data and information
- Data is raw; information is data given context and meaning; knowledge adds interpretation.
- Data quality dimensions to apply: accuracy, completeness, timeliness, relevance, consistency, validity, uniqueness.
- Big data uses the 5 Vs: volume, velocity, variety, veracity, value. Value is the commercial test - data nobody uses is a cost.
- Governance assigns ownership, stewardship, security, ethical use and legal compliance (data protection).
The digital world
- Cloud (IaaS/PaaS/SaaS), RPA, AI and machine learning, blockchain, IoT and 3D printing each change what finance can automate, analyse or assure.
- Digital transformation alters business models - platform, subscription, servitisation, disintermediation - not just the toolset.
Worked example
A shared service centre handles 50,000 supplier invoices monthly. 80% are purchase-order invoices arriving as structured data; 20% are scanned PDFs containing free text.
Which technology for which group? RPA for the 80%: the three-way match (purchase order, goods received note, invoice) is rule-based and deterministic, and RPA is built for structured, high-volume, repeatable steps. For the 20%, use OCR with machine learning to extract and classify free text, routing exceptions to a human reviewer. The reasoning is capability fit: RPA cannot interpret unstructured content or exercise judgement, so applying it to the PDFs fails.
Traps that cost marks
- Calling raw data "information" when no context is supplied.
- Treating RPA as AI: automating rules is not learning.
- Assuming cloud adoption equals outsourcing, or that it always reduces cost.
- Listing the 5 Vs but omitting veracity and value.
- Answering governance questions with "IT owns the data" - ownership sits with the business.