A generation ago, junior analysts spent their first years keying numbers from paper filings into spreadsheets. Today an API delivers ten years of standardized financials in seconds, and AI drafts the first pass of a research note. What does that leave for the human?
What's been automated
- Data collection: structured filings, market data feeds, and scraping have made gathering numbers nearly free.
- Bookkeeping and close: cloud accounting reconciles transactions continuously; the monthly close keeps shrinking.
- Screening and monitoring: alerts, anomaly detection, and factor screens do what teams of juniors once did.
- First drafts: language models summarize filings and earnings calls in seconds.
What's become more valuable
- Judgment: deciding which assumptions a model should make, and challenging the ones it did.
- Verification: automated outputs fail confidently. Someone must know what a sensible gross margin looks like — and catch it when the machine doesn't.
- Communication: turning analysis into a decision a board will actually make.
- Cross-functional fluency: managers who can read financial output without an interpreter move faster than those who can't.
The skill strategy that follows
Don't compete with the automation; supervise it. That means being strong enough in fundamentals to sanity-check any output: knowing how the three statements connect, what drives each ratio, and where models typically go wrong.
That fundamentals-first approach is exactly how our courses are built — from Finance for Non-Finance Managers for operators to Financial Market Research for analysts. Tools change; the accounting equation doesn't.