How Ian Fincher Uses Data to Find What Manual Auditing Misses

Ian Fincher

Ian Fincher pulls up a spreadsheet. A New Orleans healthcare organization's transaction data. Twelve months of activity.

Thousands of line items.

A traditional auditor would sample maybe 50 transactions. Check those thoroughly. Assume the rest are fine.

Ian Fincher does something different. He analyzes all of it. Not manually.

Using data analytics tools to find patterns that human eyes would miss.

He's looking for the transactions that shouldn't be there. The ones that violate what he'd expect based on the organization's normal activity. That's where risk lives.

The numbers that don't make intuitive sense

Ian Fincher starts by building expectations. A restaurant with consistent daily traffic should have consistent daily revenue patterns. Some days are busier.

Saturday is busier than Tuesday. But you can model what's normal.

Then he looks at the actual numbers. A Tuesday that looks like a Saturday. A month that's way out of proportion.

This isn't proof of fraud or error. It's a flag. It says: investigate this.

Something changed. The question is what.

For a healthcare organization, he might look at patient billing. If you're billing the same amount every single day with no variation, that's suspicious. Patient activity varies.

Billing should too.

A manufacturing business should have proportional inventory changes. As sales increase, inventory should decrease (assuming they're selling existing stock). If inventory increases while sales are flat, something's off.

Ian Fincher uses data analysis to develop expectations based on the organization's history and industry norms. Then he tests whether actual transactions match those expectations.

Benford's Law and the red flags hiding in digits

This is where Ian Fincher's approach gets sophisticated.

Benford's Law is a mathematical principle that applies to naturally occurring datasets. The first digit of naturally occurring numbers follows a specific distribution. The digit 1 appears as the first digit about 30 percent of the time.

The digit 9 appears less than 5 percent of the time.

This isn't intuitive. But it's remarkably consistent across real-world data: stock prices, election results, river lengths, company expense accounts.

When you artificially create numbers, when you guess, or lie, or just make stuff up, you violate Benford's Law. Your digit distribution looks wrong.

Ian Fincher applies Benford's Law analysis to expense accounts. If someone's falsifying expense reports, their made-up numbers typically won't follow Benford's distribution. The analysis flags suspicious patterns.

This catches problems that a human reviewer might miss entirely. You're not looking at individual transactions. You're looking at statistical patterns.

Trending analysis that reveals what changed, and why

Ian Fincher builds trend analyses for all significant accounts.

For a nonprofit, he tracks spending by category month to month. Program expenses should be consistent. Administrative expenses should be consistent.

Fundraising expenses follow a pattern (higher before year-end appeals, for example).

When a month breaks the pattern, he asks why. Did staffing change? Did you start a new program?

Did someone make an error in categorizing expenses?

A trending analysis reveals what changed. The pattern-breaker gets investigation.

For a business applying for a bank loan, trending analysis shows whether financial position is improving or declining. For an audit, it shows where risk is highest.

Ian Fincher also uses trending to spot anomalies that might indicate fraud. Expense accounts growing out of proportion. Receivables aging but not converting to cash.

Inventory sitting untouched for months.

Why AI can't replace auditor judgment

Here's what Ian Fincher emphasizes: data analytics is a tool, not a replacement for auditing.

Analytics shows you the red flags. But you still need an auditor to investigate. A higher-than-expected expense might be fraud.

It might be a one-time purchase. It might be legitimate activity that the model didn't account for.

Ian Fincher looks at the data flag and then asks questions. He reviews documentation. He talks to management.

He uses professional judgment to determine whether the anomaly is a problem or just a quirk.

This is why data analysis doesn't replace auditing, it improves it. Instead of testing 50 randomly selected transactions and hoping you catch problems, you test every transaction and focus intense review on the ones that don't fit the pattern.

You find more problems. You miss fewer things.

For a regulated organization like a nonprofit, this matters. For a business managing risk, this matters.

Ian Fincher combines data analytics with old-fashioned auditor skepticism. That's the combination that actually works.