Email Sales

Email Sales

Need product support? Please visit our Customer Support page.





Back to Blog

XBRL US DQC Announces Second Set of Validation Rules

Lou Rohman, Vice President, XBRL Services, Merrill Corporation | June 30, 2016

The XBRL-US Data Quality Committee today revealed its second set of validation rules and guidance for SEC filers, inviting public comment on the proposed rules.

These new rules would have checked 800,000 data points submitted in the last year to determine if errors existed in those data points. Coming fresh off the encouraging report recently on the impact of the first set of rules, I’m excited to see these new rules further drive the upward trend in XBRL data quality.

Reducing inappropriate axis + member tag combinations

Among the five proposed rules, I believe the one with the greatest error-reducing potential takes aim at inappropriate combinations of axis and member tags. Here’s an example: A company tags the amount of sales from its Chemicals segment using the “Legal Entity” axis tag with the “Chemicals Segment” member tag. The problem? The Chemicals Segment is not a legal entity — the combination is invalid. The company should instead use the “Segment” axis tag with the “Chemicals Segment” member tag. In plain text format, an unknowledgeable XBRL preparer may not see the error. But for data analyzing software, the erroneous combination renders the information unsearchable and ultimately unusable.

At Merrill, we see these invalid combinations far too frequently in SEC filings. As a leader within the DQC, I’ve emphasized and worked at a very detailed level to properly include this new validation rule — and I’m hopeful it will significantly reduce errors.

Run the rules against all upcoming — and previous — filings

As I’ve said before on this blog, if you submit XBRL to the SEC, you should be running these rules before filing. It’s better to catch and fix errors yourself, rather than have an investor or the SEC discover that your XBRL is telling the wrong story.

Don’t assume the rules catch every error

Another issue we see often at Merrill: An organization runs its disclosures against DQC rules, assumes the rules have caught all errors, and is surprised when we subsequently identify other errors in its filing. Filers must be aware that there are plenty of errors, including axis + member combinations, for which rules haven’t been or can’t be written. There’s still no replacement for a proper detailed review by a seasoned XBRL expert.

I agree

This site uses cookies to offer you a better experience. For more information, view our privacy policy.