How To Analyze Bank Statements, Part 1: OCR Conversion

Investigator using a laptop

If you are a detective, investigator, accountant or attorney, you may face the arduous task of analyzing bank statements. If it is a complex case, there can be years of statements and multiple accounts. How do you isolate and categorize the transactions that are germane to your investigation?

Typically, you receive scans of bank statements from your client, some better quality than others. How do you go about automating an endeavor that may take weeks or months if done manually?

Here are 4 steps to put bank statements into a format that will facilitate analysis.

Step 1. Select an OCR tool.

If your statements are in pdf format, you need an OCR (Optical Character Reader) tool to convert them into a format that can be processed by a computer, e.g. Omnipage, Adobe, Nuance, Abby Fine Reader. There are many OCR tools that claim to convert documents into multiple formats, e.g. csv, txt, xlsx. Excel is a good choice because it simplifies analysis. Here are questions to ask:

(a)      How accurate is the conversion process?
(b)      Is it tuned for bank statements?
(c)       Can it process all bank statements? If not, what are the limitations?
(d)      What options are there for output format?
(e)      If the output format is Excel, are the transactions formatted in defined rows and columns to facilitate analysis?

Step 2. Process the statements.

Does the tool make it easy to correct OCR errors, such as:

(a)      Date is in proper format?
(b)      Numeric data contains only numbers, currency indicators and/or decimal points?
(c)       Debits and credits are in separate columns and are positive numbers?

Step 3. Reconcile the statements.

To insure there are no incorrect transaction amounts, check that:

(a)      Dates are in chronological order.
(b)      The account ending balance equals the sum of the opening balance minus debits plus credits.

Step 4. Prepare the data for analysis.

(a)      Decide which tools you will use for financial analysis.
(b)      If necessary, format the data. Check that date fields contain dates, numeric fields contain numbers and text fields contain text. This may sound obvious, but how can you analyze bank statements with data that is all text?
(c)       If you are using Excel, are all data fields in separate columns?

Now that you are ready to analyze your data, watch for How To Analyze Bank Statements, Part II, that will highlight different types of analysis tools you can use to speed up your financial investigation.

If you want to learn about a Bank Statement Analysis Tool that automates these tasks, give me a call at (239) 842-8224.