Digital Financial Reporting

The Committee of European Audit Oversight Bodies (CEAOB) has issued guidance, Guidelines on audits' involvement financial statements in ESEF, related to the audit of both the human readable layer and machine readable layer of XBRL-based digital financial reports.  Here is that guidance.

To give you a flavor for the requirement, see page 5 section titled "Accuracy" which includes this guidance:

• The marked-up information does not correspond with the human-readable layer of the financial statements;
• Numbers disclosed in the primary statements of the IFRS consolidated financial statements have been marked-up with an inaccurate context (e.g. year or year-end, currency; debit/credit; scaling (i.e. millions/thousands));
• Inappropriate elements from the core taxonomy have been selected; 
• A misrepresentation of the accounting meaning of the number or disclosure being marked-up arising from selecting an inappropriate element from the core taxonomy; 
• An extension taxonomy element created to mark-up a number in the primary statements is not anchored to the core taxonomy element having the closest wider accounting meaning and/or scope to that extension taxonomy element of the issuer; 
• Where an extension taxonomy element combines a number of core taxonomy elements, the issuer has not anchored that extension taxonomy element to each of those core taxonomy elements.

 Here is information that I provided related to the audit of XBRL-based financial reports.

I have use the best-practice based method I developed and framework I have created to construct a high-quality sophisticated example of an XBRL-based report.  This report happens to be a financial report; but it can help you see to possibilities of general business reports also.

You can interact with this sophisticated example XBRL-based report by:

• Looking through this human-readable representation created by a commercially available software application that is currently available.
• Loading the XBRL instance into the software application of your choice.
• Downloading the noncommercial working proof of concept software application Pesseract and using that application.
• Downloading and using the open source Arelle XBRL processor.
• Using the XBRLQuery web service.
• Downloading the ZIP file with the XBRL instance, XBRL taxonomy, XBRL linkbases, XBRL formulas, etc.
• Looking at this auto-generated Inline XBRL document or this human readble HTML version of the same Inline XBRL format (this has some issues related to improperly formatted XHTML, I will eventually fix those)

The techniques that I use are based on about 10 years of poking and prodding XBRL-based financial reports that have been submitted to the SEC (thank you public companies SEC!), my knowledge of XBRL, and my accounting knowledge.  A trained observer with the right tools can learn a plethora of information by observing the good and bad patterns of all those public company XBRL-based financial reports.

Here are some key things to understand about this report:

1. This report was validated using four different XBRL processors including XBRL Cloud's processor, UBmatrix XPE 4.0 processor, XBRL Cloud's evidence package, Pesseract, and the UBmatrix Taxonomy Designer.  All give consistent results per XBRL syntax and XBRL formula validation.
2. There are 53 XBRL Formulas that verify 123 different facts with consistent results (UBmatrix XPE; XBRL Cloud). Most of these XBRL formulas are roll forwards, member aggregations, or cross dimensional computations that XBRL calculation relations cannot explain or verify.
3. There are 39 sets of XBRL calculation relations all of which are verified to be consistent with expectation across all software applications. (UBmatrix XPE; XBRL Cloud)
4. There are 60 different disclosures and all 60 disclosures have rules that explain how the disclosure should be created and all disclosures are created as document by those machine-readable rules (Pesseract).
5. There are 7 high-level fundamental accounting concept relations that are verified to be consistent with expectation (Pesseract).
6. There are 55 sets of reporting checklist rules that verify that all required disclosures are provided, all disclosures that must be provided if a specific line item are provided, there are matching sets of level 1 notes text blocks, level 2 policy text blocks, level 3 text blocks, and level 4 detailed disclosures (Pesseract).
7. All of the XBRL presentation relations are tested against a set of rules to make sure they are valid relations.  There are 10 relations, all within one report fragment, that are "odd" to make a point.  Look at the representation which (a) is inconsistent with other fragments and (b) is not detected by XBRL validation.  Notice how you have to look at them a bit in order to figure out what is going on.
8. Note that every report fragment is logical and readable.  Now granted, there is room for improvement in the rendering and there are a few cases where a fragment, if moved to a separate network, would cause both separate facts sets to look better than the rendering where the fact sets are combined.
9. Note that 100% of the reported facts are properly mathematically interrelated, that there is no duplication of facts in the XBRL instance, and there are several rather unique use cases.  For example, there is a correction of an error (i.e. the same fact as originally reported containing an error and restated where the error has been removed).  Reporting information using two different reporting scenarios used. (actual and budgeted)  Note that all the mathematical relations tick-and-tie; cross caste-and-foot.  And 100% of these mathematical validations are visible in the report (the green) and can be seen to be correct.

Now, consider something.  Imagine thousands of these high-quality reports inside a repository of reports of some sort.  Here is another view of that repository.  None of this has to be "technical" or hard in any way.

If you are having trouble understanding that business reporting can use XBRL because this example relates to financial reporting, check out this Lorem Ipsum example. (i.e. any words or numbers can be represented using XBRL, not just financial words and numbers)

If you are considering purchasing XBRL-based digital financial reporting related software; check this example out throughly and then have the software vendor load this XBRL-based report and compare that software with what you see here.  Don't be an ignorant buyer.

There is a lot that can be learned by exploring this report.  There are subtleties and nuances that abound that you do not want to miss understanding.  I am thinking about doing a weekly digital financial report study group.  Let me know what you think of that idea.  Send me an email.

Workiva published a little ebook called Connected Reporting that is worth reading.  In the ebook Workiva does a good job of explaining that connected reporting should be utilized, why connected reporting should be utilized, and even the five qualities of connected data.  They even explain three broad steps that you need to take to get to the point where you can do connected reporting.

What is not explained are the implementation details of how to get to connected reporting.

What is needed is a proven method of implementing connected reporting. The method needs to be auditable.

But the first step is to up your digital maturity. This is not about pawning off some task to the IT department.  That is probably the worst thing you could do. Remember, "Digital maturity is primarily about people and the realization that effective digital transformation involves changes to organizational dynamics and how work gets done."

You need to avoid creating a fragile house of cards.  Recognize that 81% of business leaders do not really understand artificial intelligence.

"Connected reporting" is just a buzz word Workiva created that refers to the same thing others are talking about using other terms such as "financial transformation", the "modern finance platform", "continuous reporting", and other such terms.  Essentially, they are all talking about the same thing.

Start here.  Be wary of the snake oil salesmen that are out there ready to part you and your money. There is no rush.  There is no need to be on the bleeding edge.  But there is a need to start and to build in the right direction. Gradually increase your digital maturity so that you can be ready for accounting, reporting, auditing, and analysis in the fourth industrial revolution.

Also, see this survey by Workiva and the AICPA; 2019 State of Financial Reporting Survey.