A de facto good practices industry standard digital general purpose financial report metamodel is emerging.
By de facto I mean that while no one has actually published one specific document explaining that standard; practices that exist in reality, even though they are not officially recognized; standards do, in fact, exist.
By best practices I mean that a method or technique that has been generally accepted as superior to any other known alternatives because it produces results that are superior to those achieved by other means or because it has become a standard way of doing things. Sometimes people are offended by the term "best practices". A more politically acceptable term that means effectively the same thing is "good practices". The point is; they work better than existing known alternatives.
By industry standard I mean the normal or ordinary or typical or average manner of doing things within an industry. It is how an industry governs itself.
By digital I mean BOTH readable and understandable by machine based processes. PDF and HTML are technically readable by software applications...but they are not understood by that software. XBRL is machine readable and can be made machine understandable if the right machine readable information is provided. As a by product, digital is also readable and understandable by humans.
By general purpose financial report I mean a high-fidelity, high-resolution, high-quality information exchange mechanism used by an economic entity to report the financial position, financial performance, and liquidity of that entity. Typically general purpose financial reports are provided to parties that are external to that organization (i.e. external financial reports including policies, notes, and qualitative information).
By metamodel I mean a model that is followed by all report models. For example; Microsoft, Apple, Google, Amazon, Facebook, and Salesforce each have a report model that is used to represent their individual financial reports. But ALL of those reports follow the same common, standard SEC report metamodel. The ESMA also provides a standard report metamodel.
And so what is the basis for this de facto good practices industry standard digital financial report metamodel? Here is a list:
- US GAAP Financial Reporting Taxonomy Architecture: One of the very first general purpose financial report implementations was created by the U.S. Securities and Exchange Commission. XBRL US was charged with creating the XBRL taxonomy to support that endeavor. To achieve this, XBRL US created this architecture document. Mantenance of this taxonomy architecture was transitioned to the FASB. The original document published by XBRL US is better, but this document will do the trick.
- SEC's EDGAR Filer Manual: In support of XBRL-based filings of financial reports to the SEC, the SEC created this manual. It has some excellent information; but it also leaves a lot out. The result: quality problems. Bad practices that are allowed by the SEC are not allowed in the de facto good practices metamodel. But, we take all the good practices.
- European Single Electronic Format (ESEF) Manual: In support of XBRL-based filings of financial reports to the ESMA, the ESMA created this manual. Again, some excellent information; but it also leaves a lot out. The ESMA and SEC are more similar than they are different.
- XBRL International Open Information Model 1.0 (OIM): Useful information, but tending to be too technical in nature; the OIM helps one understand that a semantic model of a business report is necessary.
- OMG's Standard Business Report Model (SBRM): While not yet published, OMG is creating a logical conceptualization of a business report. There are drafts provided in UML.
- Logical Theory Describing Financial Report: This logical theory which provides axioms, theorems, and a world view was created by reverse engineering XBRL-based reports that have been submitted to the SEC and is the basis for the Standard Business Report Model. This is effectively the model I have implemented.
All of the above information was synthesized and implemented, generally with my assistance, by the following commercial software companies, working proof of concepts, and prototype software applications:
- XBRL Cloud: (Commercial software) Implemented about 95% of the metamodel. This can best be seen in their CleanScore Evidence Package. (Here is the evidence package, fundamental accounting concepts, and disclosure mechanics and reporting checklist validation for the 2017 Microsoft 10-K; Similar examples are provided here for Apple, Google, Amazon, Facebook, and Salesforce)
- 28msec/NTT Data: (Commercial software) Several years ago I helped 28msec implement XBRL in their database. They learned a lot, I also learned a lot and used that to help tune my model. Apparently 28msec was sold to NTT Data. 28msec implemented about 60% of the model.
- Pesseract: (working prototype) Pesseract is a working proof of concept that implements about 95% of the metamodel.
- Auditchain's Pacioli: (Commercial software) Pacioli is by far the best and most complete implementation of this metamodel. I have performed rigirous testing using multiple financial reporting schemes. The current interface is the GUI used for testing Pacioli; a new interface is scheduled to be put on and additional enhancements are planed as I understand it.
- Luca: (Pre-release commercial software) Cloud-based Luca is pre-release software for creating XBRL-based reports. The interface of Luca explains the fundamental report model very well. I don't know how to quantify how much this supports; it supports 100% of the report model but none of the supplemental validation stuff yet.
- Open source Microsoft Access Database: (prototype) This is a working prototype that I created.
- ACCIZOM: (Commercial software) I don't know a lot about ACCIZOM yet, but they say they support this metamodel and the Seattle Method. I have not yet confirmed that, but will.
- SCI: (Commercial software) SCI is building an expert system for creating financial reports and is leveraging the Seattle Method.
- Awther: (Commercial software) Awther is building an expert system for creating financial reports and is leveraging the Seattle Method.
OMG and XBRL International both call for only two implementations to support their standards. I have seven. Technically, you could say that the 30 software vendors that support XBRL-based reporting support this model because SEC filings support the model. Now, many of those SEC XBRL-based reports are inconsistent with the logic of the metamodel, but that is a problem in the creation of the reports, not the metamodel. Same can be said for ESMA reports, but I don't have any data on those yet.
It would be fantastic if a formal ISO standard for digital financial reports was created. Alternatively, be nice if XBRL International or OMG issued such a formal standard. For now, de facto is best I can do.
Try out my method, which I call the Seattle Method, for leveraging this industry standard for implementing automation within the enterprise.