Digital Financial Reporting

The CFA Institute paper which calls for broader and deeper use of structured data in financial reporting points out that to achieve the vision that paper articulates, accountants and auditors need increased education. I agree with that assessment of the CFA Institute.

The paper calls specifically for increased education in "technology and analytical methods".  That is not very specific and I think it misses one very important skill. That skill is a discipline of philosophy, formal logic.

Also, there is a general movement going on to "get everyone to learn how to code".  That notion is misguided.  That is NOT what is needed. I am not the only one who believes this "learn to code" movement is misguided. What accountants need is to understand the rules of logic by which computers operate.

Besides, this Wired article, End of Code, has the sub title "Soon we won't program computers. We'll train them like dogs."  That is a very succinct and accurate statement.  But I do disagree with one thing Wired is saying.  Business professionals will train software using business rules, not code.  Financial reporting is not the sort of thing machine learning was designed for.  So don't believe the snake oil salesmen who tell you otherwise.

Further, if you are judging what you have to know based on the current software that is available to create structured data and you don't consider the Law of Conservation of Complexity and the Law of Irreducible Complexity, then you are apt to reach the wrong conclusion as to what professional accountants and auditors even need to understand. Most structured data tools today expose far, far too much technology "stuff" to users of the software. That technical stuff will be burried far more deeply within software in the future.

What professional accountants and auditors need is to understand how computers work and how to control the workings of computers to accurately understand what computers are capable of doing and what they are not capable of doing.  You want to get the correct training?  Go test drive a Tesla.  Literally.  If you go try out the driver assist feature and then think about your experience, you will learn way more that if you learned how to code.

Professional accountants and auditors have the vast, vast majority of what they need to understand dialed in.  Consider this question: how long do you think that it might take for a computer science professional to learn the domain of accounting and financial reporting? 

What professional accountants and auditors need is between a 3 and 5 day class on how to communicate effectively with information technology professionals. This is a bit of the discipline of philosophy, specifically formal logic.  It is also a bit of engineering, or specifically something like Six Sigma or Deming or systems theory.  Professional accountants need to understand how to communicate effectively with information technology professionals to help technology professionals build the software they really need and to be able to test that software to see that it is working correctly.

Additionally, accountants and auditors need to re-understand that they think differently than non-accountants.  The section A Very Brief History of Accounting will help them recall Friar Luca Pacioli's invention of double-entry bookkeeping and how it was engineered to detect errors and fraud. Don't make the mistake of taking this feature for granted or even worse, forgetting about it!

What professional accountants and auditors need to understand is the basics of how computers actually work and the mechanisms used to control a computer.  I understand this because I have a degree in accounting and am a trained auditor.  I also have a significant amount of formal training in information systems.  As it happens, I also received training in formal logic as an under graduate.  When working toward my graduate degree I studied Deming.  That combination was pure chance.

I got the XBRL ball rolling.  I invested in understanding XBRL by helping to create it.  Then, I spend the next 15 years trying to figure out how to get XBRL to work the way I know accountants and auditors need XBRL to work.  Along the way I took lots of notes.  I took what I learned; organized, summarized and synthesized that information as best that I can at this point.  This is what I came up with:

• Conceptual Overview of an XBRL-based Structured Digital Financial Report: Begin with the end in mind as the Seven Habits of Highly Effective People points out.  What is the objective?  What exactly is "structured data".  The CFA Institute paper does an excellent of laying out their vision, but they don't provide a lot of detail as to what structured data is or how to make it work.
• Comprehensive Introduction to Knowledge Engineering for Professional Accountants: This document helps you sort out the capabilities of computers and how to get computers to do what you desire them to do.  Computers work using the rules of formal logic, which is a discipline of philosophy which was invented in about 450 BC. Conceptual models provide frameworks for understanding. A framework is a set of principles, assumptions, ideas, concepts, values, rules, laws, agreements, and practices that constitutes a way of establishing how something operates.
• Comprehensive Introduction to Business Rules for Professional Accountants: Business rules guide, control, suggest or otherwise influence behavior.  Accountants and auditors understand human-readable rules that you get out of books.  They need to understand machine-readable rules and the sorts of rules machines need.
• Comprehensive Introduction to Intelligent Software Agents for Professional Accountants: Intelligent software agents is how work will be automated.  This helps you understand what an intelligent software agent is, how they work, what it takes to control such agents, and other important information about using these potentially helpful creatures.  It also helps you understand their limitations.  Don't be fooled by the snake oil salemen, be an informed buyer.
• Financial Report Semantics and Dynamics Theory: This theory provides a conceptual model of the mechanics of a digital financial report that I created and use because neither XBRL US, the FASB, or SEC provide such a conceptual model.  It explains the mechanics of a financial report.
• How XBRL Works: This five minute video helps you understand the difference between structured information and unstructured information.  It helps you see how computers can address specific information within a structured digital financial report.

If anyone has better information, please send it my way. I always work to improve.  I like to reach conclusions based on evidence and facts, not based on unsupported subjective opinion.

There are a lot more details if you want them, but that the information above is really the essence of what you need to get your head around the future of financial reporting which is digital financial reporting. With 3 to 5 days of work, professional accountants and auditors will understand how to get their heads around digital. Digital is not software or even a technology.  Digital is a mindset.

Accountants tend to try and understand XBRL from the perspective of what they are doing today and what they understand today.  That will not work.  Structured information is a transformational change, it is not an incremental change.

And don't make the mistake of thinking that accountants and auditors are in this boat alone.  Every business professional is in this same boat of needing to understand digital.  In fact, most information technology professionals are in this boat also.  Most information technology professionals are fixated on relational databases and don't understand how to leverage business rules appropriately.

Many accountants working with XBRL-based stuctured digital financial reports are actually ahead of the curve, not behind it.  Look at the first graphic on this page; anyone who creates 90% or more of their XBRL-based public company filings to the SEC is on the right track.

Become knowledgeable enough to differentiate a snake oil salesmen from someone that actually has something useful. Might save you a lot of money and frustration.

Also, protect your job.  The Economist reports information from a study that predicts that there is a 94% probability that significant job losses will occur for accountants and auditors over the next 20 years; that percentage is 98% for bookkeepers and accounting clerks. Understanding the capabilities of computers will help you keep your job!