In the face of major new accounting requirements, even the biggest technophobe can’t stand up to the freight train of change that necessitates technological solutions to achieve compliance.

The newest star on the stage of technological advancements—artificial intelligence—along with advancements like “optimal character recognition” and “natural language processing” have a key role to play in how companies adapt to new requirements for recognizing revenue and accounting for leases. In some large, complex business environments, the task of extracting the necessary data and detail from contracts would be daunting, even impossible, otherwise.

To comply with the new revenue standard, for example, which calendar-year public companies will do beginning Jan. 1, companies need to analyze and extract multiple data points from each of their contracts with customers before they can debit or credit the appropriate accounts at the appropriate times under the new rules. Tools using artificial intelligence can help with large volumes of contracts to look for consistent themes or patterns, says Sean Torr, managing director at Deloitte. It’s also useful to flag outliers.

With leases, some of the largest companies have leases numbering in the tens of thousands. “It takes somewhere from nine to 10 hours for a human to review a lease,” finding and collecting all the details needed to comply with the accounting requirements, says Mike Baccala, an assurance partner at PwC.

There aren’t enough humans trained in that analysis to complete the work in the timeframe necessary, says Baccala. To varying degrees, depending on the setting and the circumstances, tools like AI, OCR, and NLS are helping companies get structured data out of static documents. “It’s a real problem when you start to do the math on some of these areas,” he says. “We’re using a number of technologies together to look at the problem of GAAP change,” he says.

Rob Bruce, vice president at software developer Kimble Applications, says companies can use such advanced technology to “drive the right behaviors” in three important ways. First, technology like artificial intelligence, or augmented intelligence, prompts all the right behaviors to get contract information recorded in the ways necessary to comply with the standards. Correctly identifying transaction pricing is key to compliance with the new revenue standard, for example.

“This is not just a finance problem. Knowing whether an obligation is complete, or how complete it is, is important.”
Rob Bruce, Vice President, Kimble Applications

Second, it facilitates analysis, assuring data is both complete and correct. “Using diagnostics technology helps people look at what results they are getting, what forecasts they are getting, what are the trends,” says Bruce. Completeness and accuracy has been a big problem for companies trying to satisfy demands from auditors under existing standards, as auditors have been pressured through regulatory inspections under the Public Company Accounting Oversight Board to get better evidence to support their audit opinions.

Finally, it drives collaboration across departments in an organization. “This is not just a finance problem,” says Bruce, “Knowing whether an obligation is complete, or how complete it is, is important.” The new revenue standard allows companies to recognize revenue only as they fulfill separately identified performance obligations within each contract, and they must make some disclosures about obligations that are not yet satisfied on the reporting date.

It’s still early in the curve toward corporate adoption of advanced technologies to achieve compliance with the revenue standard, says Bruce. “People are daunted by the amount of manual work that may be associated with this,” he says. As a result, companies are showing interest, but are not committing in droves to advanced deployment.

The new standards are driving interest in automated contract management systems, says Nagi Prabhu, chief product officer at technology firm Icertis. Powered by artificial intelligence and machine learning technologies, contract management systems take in contract information in a way that automatically sorts data into the right buckets necessary for accounting compliance. It recognizes both the data and the language to apply the proper accounting treatment, he says.


Below are some recent statistics from the Harvard Business Review on companies’ data quality.
On average, 47% of newly-created data records have at least one critical (e.g., work-impacting) error. A full quarter of the scores in our sample are below 30% and half are below 57%. In today’s business world, work and data are inextricably tied to one another. No manager can claim that his area is functioning properly in the face of data quality issues. It is hard to see how businesses can survive, never mind thrive, under such conditions.
Only 3% of the DQ scores in our study can be rated “acceptable” using the loosest-possible standard. We often ask managers (both in these classes and in consulting engagements) how good their data needs to be. While a fine-grained answer depends on their uses of the data, how much an error costs them, and other company- and department-specific considerations, none has ever thought a score less than the “high nineties” acceptable. Less than 3% in our sample meet this standard. For the vast majority, the problem is severe.
The variation in DQ scores is enormous. Individual tallies range from 0% to 99%. Our deeper analyses (to see if, for instance, specific industries are better or worse) have yielded no meaningful insights. Thus, no sector, government agency, or department is immune to the ravages of extremely poor data quality.
Source: Harvard Business Review

Prabhu also sees growing interest in advanced technologies as a result of the onslaught of new accounting requirements. “Every second or third conversation with customers is asking us: what can you do to provide technology to solve this problem?” he says.

Those automated technologies may be helpful going forward, but they don’t address the mounds of paper contracts companies have today. That means the first big hurdle companies have to clear in the short term is to convert paper contracts into digital data, says Deloitte audit partner Will Bible.

Accounting leaders have reported companies are using a mix of automated systems and manual controls to achieve compliance with the revenue recognition standard when it begins to take effect on Jan. 1. The enormity of the work to comply with the new revenue rule is leading companies to think hard about how they’ll comply with the leasing standard when it takes effect a year later.

The experience of trying to comb through contracts for necessary accounting detail to comply with revenue recognition has given companies reason to move more expeditiously into lease compliance. “Companies are seeing how important it is to have an electronic format to manage their portfolio going forward,” says Bible.

Despite futuristic predictions that migration to AI and other advanced technology will replace humans, Henri Leveque, another PwC partner, believes that won’t happen. “None of these solutions eliminate humans,” he says. Instead, “the technology is making humans significantly more relevant and valuable.” It still takes human intelligence to read and analyze results and to make decisions based on what the data indicate. In fact, accounting and auditing standards over the past several years have moved away from routinized application of bright-line requirements into principles that necessitate more judgment and analysis.

Chris Stephenson, a principal at Grant Thornton, says both the technology and corporate adoption have a long way to go before companies will fully utilize the technology to achieve seamless compliance. Artificial intelligence suggests software is making decisions, and human intervention or expertise is not needed. Today, the environment is one where humans are complying with rules and relying on software to help. “Artificial intelligence is the other side of the goal post where humans aren’t there at all,” he says.

Companies still have a long way to go to digitize, assemble, and cleanse the data they have now, says Stephenson. A recent Harvard Business Review analysis suggests only 3 percent of corporate departments have acceptable levels of data quality. On average, 47 percent of newly created data records have at least one critical error, the analysis found.

That suggests continued corporate adoption of automated processes to originate contract data will go a long way toward moving companies into a realm where artificial intelligence is truly a factor in achieving compliance. “Once that happens, I see huge opportunity,” says Stephenson.

As machine learning gets deeper into contract data, maybe it could even learn to negotiate contracts, making decisions at lightning speed based on reams of historic data on pricing and terms. “In revenue, the contract negotiation goal is to be the most prepared person in the room,” says Stephenson. “A bot can grab internal data [and] historic data and can set terms even better than a human. With leasing, you could be in a position where the lessee and the lessor have bots doing the negotiation with a human approver.”

But back in the present, there’s a more timely, compelling reason for companies to get further along the curve of adopting artificial intelligence and related tools to perform their accounting processes under new standards. To respond to growing demands for better audit quality, auditors are adopting the tools to perform more robust checks on accounting and internal controls.

“In certain cases where there are high volumes, we deploy platforms that read contracts and can assist in audit procedures,” says Bible.