Now that public companies are required to use Extensible Business Reporting Language, or XBRL, in their financial filings with the Securities and Exchange Commission, the SEC is trying to make sure the data is appropriate and not hiding accounting gimmicks.

Lou Rohman, vice president of XBRL services at Merrill Corporation, a member of XBRL US’s Data Quality Committee, recommends that financial statement preparers take a closer look at the guidance being issued by his committee. Documents on the committee’s website provide guidance and validation rules that should be used for the filings.

“The SEC gets the information and puts it in a database and then users come in to use it,” he said. “The challenge is getting clean, good data in that database. That’s always going to be an issue. Any database has errors in the data.”

With XBRL, the SEC has put out a set of rules, but hasn’t yet enforced them to a great extent, according to Rohman. His company, Merrill, helps companies prepare their XBRL data before filing it with the SEC.

The Data Quality Committee is trying to address XBRL quality issues by providing an industry consortium where the main players in the industry can come together and recommend how filings should be done.

“Here are some automated rules that all companies can use, and here are some guidelines as well that companies can use,” said Rohman. “Those rules are free of charge, so why wouldn’t a company take them and run their filing each quarter through those rules before they submit the XBRL to the SEC, thereby correcting anything they see that isn’t appropriate. That’s one of the challenges. How do you get all companies to consistently and accurately file XBRL?”

He has been seeing some improvement in the XBRL filings in the years since the SEC first mandated use of the data-tagging technology in 2009.

“It’s getting better,” said Rohman. “The first two years had a limited liability, kind of a soft landing that the SEC gave them. But that went away a number of years ago. Since then, it’s tightened up, but there’s a gradual improvement. But the challenge is how do you get everybody to care when there’s no penalty?”

The SEC has been posting some observations online to try to prod companies to do the right thing with XBRL. In March, the SEC posted some staff observations about the misuse of custom axis tags by some filers.

“The SEC has come out with those types of observations to say they’ve noticed this problem,” said Rohman. “It tells you the SEC is looking at this stuff, and we’ve been told internally the SEC is using the data. So there’s a risk with filing information that isn’t proper and having the SEC use it for whatever purpose they use it for. That’s something companies are becoming more aware of and therefore improving the quality because nobody wants to be pointed out. Their observation talked about how you’re allowed to make unique tags for your own company if they don’t exist in the U.S. GAAP Taxonomy. That’s fair enough, but there are certain tags that talk about a category. Those types of tags, the SEC is saying, should be extremely infrequent.”

Rohman is also on the Financial Accounting Standards Board’s Taxonomy Advisory Group, or TAG, which provides advice to FASB on updates to the GAAP Taxonomy as new accounting standards updates come out for projects such as leasing, revenue recognition, financial instruments classification and measurement, and credit losses.

The SEC has also been using XBRL filings to look for outliers that could point to possible signs of financial shenanigans.

“They have what’s called their Corporate Issuer Risk Assessment financial model software, CIRA,” said Rothman. “It takes in tons of information and produces outliers that maybe an examiner should go look at more. Maybe they’ll find out everything’s OK, but it points out on a dashboard some areas. We’ve been told that by the SEC at some of their conference speeches. Part of the source of going into that is XBRL, probably not a large portion at this point, but it’s growing because they’re seeing the ability and the value of grabbing this extremely quickly, as opposed to some other sources. It’s directly from the filers so that’s the best way to get information, as opposed to going through an aggregator.”

The SEC has emphasized that the tool is just a risk assessment at this point and not the “RoboCop” it has been portrayed as in some quarters. Still, many of the details about the technology are not being made public at this point. “A lot of it is still behind the walls of the SEC, so it’s hard to know exactly how they’re using it,” said Rohman.