« Road Work Ahead: Last Mile of Finance | Main | Game Changer: XBRL Viewer Add on for Firefox »

Model SEC XBRL Filing

I have been struggling and struggling trying to figure out the best way to communicate some things I have been trying to communicate and have settled on a number of model SEC XBRL filings to articulate the message.  You can find the first model SEC XBRL filing here. (If the site is slow, check back; the hosting people are doing an upgrade or something and the site has been slow.)

The key thing that I am trying to achieve is to show CPAs how to correctly model an SEC XBRL filing from an information model integrity perspective. The message is that:

  • the balance sheet, income statement, cash flow statement, statement of changes in equity are tightly intertwined
  • the policies and disclosures provided details which sometimes are stand alone, but sometimes are details of the aggregated summaries on those primary financials
  • properly expressing these relationships from an information modeling perspective allows for features such as "drill down" and moving from one area of the statement to another area
  • paper-based and document centric notions such as "presented on the face of the financials" get in the way of getting the most from XBRL

We are rapidly moving from the two dimensional medium of paper and electronic paper to a more feature rich model which leverages things like the multidimensional model. I have heard a few people use the term "model driven reporting".

What I am trying to do is point to specific issues in order to highlight the issues so CPAs and others in the financial reporting supply chain can figure out the best way to apply XBRL to financial reporting.  How XBRL is used is a choice. Information about the possibilities helps get to the appropriate answers, all things considered.

So, I am creating a series of model SEC XBRL financial filings to point out some things. I am starting with a "semi well modelled XBRL SEC filing" to make some specific points.  That is the model above.

This PDF file describes the model SEC XBRL filing.  It provides a narrative which shows how things are tied together and certain choices which must be made by SEC XBRL filing creators.  I submitted the model to the SEC Previewer which shows the resulting renderings.

You can make your own way through the model, here are some suggestions on getting the most of the model and the key points I want to articulate:

  • The financials are intertwined.  This XBRL formulas validation report results shows how interrelated. You can go look at the XBRL formulas themselves, but they can be hard to read without a good application.
  • Look at the financial as the interrelated fact tablesthat they are. The meta data of one fact table needs to correctly relate to other fact tables in order for the validation report to show no errors.  Every SEC XBRL filing needs to be properly articulated for the information in the filing to be useful.
  • Properly articulated information results in properly rendered SEC XBRL filings. Walk through this Excel spreadsheet to see how a fact table is turned into a human readable rendering.
  • These relations will be much easier to see when good software exists for viewing and comparing SEC XBRL filings.  By looking at these details you can see the impact.  If you are not comfortable at the detail level of XBRL, try and use the Firefox XBRL viewer addon to read the XBRL instance. It does not render the model very well, but it does show you some of the drill down capabilities and how a financial report will be more dynamic in the future.
  • Go grab the ZIP file (line "I") from the model's home page and run it through the SEC previewer yourself.  Edit the XBRL taxonomy and XBRL instance and see how the SEC previewer rendering changes. Notice how the rendering is better if you have many small pieces rather then fewer larger pieces.

(Note that this model consciously points to some label linkbases where others should be used for an actual SEC filing.  I have to do this due to the taxonomy editing software I am using.  That causes one EFM validation error, I realize that. The final version will have that adjusted to use the proper label linkbase files.)

PrintView Printer Friendly Version

EmailEmail Article to Friend

Reader Comments (1)

I can't understand the lingo yet (distinguishing Concepts, Members, Axis, Domains, Roll Ups, Elements, etc.) I hope to become conversational in XML, and then XBRL soon. But as it is I am tagging data for financial companies, and struggle with the same "document-centricity", the questions like whether we would prefer to drill down by Member or by concept (in deciding how to structure taxonomies), and that XBRL isn't leveraged to its fullest in its current state. Just wanted to show you support from an interested reader. Thanks!
November 24, 2010 | Unregistered CommenterNate Anderson

PostPost a New Comment

Enter your information below to add a new comment.

My response is on my own website »
Author Email (optional):
Author URL (optional):
All HTML will be escaped. Hyperlinks will be created for URLs automatically.